Rattys latest EFI project

Introduction

Firstly, a little background for those of you who don’t know what my Brother and I are currently up to – we are both building (or re-building) Bajas. Both of the cars are based around circa 1965 body shells but have undergone a huge number of modifications to make them even more capable than the basic Baja.

My brothers car has been rebuilt several times, each time to make the vehicle less likely to suffer from whatever had continually failed in the last ‘version’. This time it was a conversion to IRS as Dave had decided that he had better things to do than to spend every second weekend replacing Swing Axel gearboxes! Whilst he was at it he fitted 4-wheel disc brakes, converted the brakes to dual circuit, re-manufactured the lift kit, and repaired some rust patches and dents that had appeared since the last rebuild. The 2.1 litre Type IV engine was also refreshed as it was showing signs of uneven compression due in part to failing heads.

In the last ‘version’ of this car, the engine was fed by a modified Kombi EFI system – modified only in physical appearance (to fit into a Baja and fit over the Porsche fan shroud). The ECU and EFI components were straight out of a Kombi. This system worked fine for the most part (far better than any of the Carby combinations that he had tried), however the cold start function would cause the engine to hunt at idle, and it appeared to be running too rich at certain loads and too lean at others, most likely due to a changed (from stock) vacuum pattern.

Now, due to bigger valves and other ‘secret’ mods to the refreshed engine, Dave decided that the stock system was no longer going to do the job – it was time to go aftermarket. He had seen what I had done with my Delco/Kalmaker set-up and liked the look of it, so he suggested that he might be able to borrow the entire system that I had installed on my test engine so that he could get his Baja finished and over to Frazer Island for the Christmas/New Year period (as he does almost every year). At first I wasn’t too keen to just hand over this rather expensive system, but I then considered that it would give me an opportunity to finally get this EFI system out on the road and learn the ins and outs of tuning an EFI system from scratch…at no risk to my engine!

Saturday:

Dave had collected a number of JE Camira EFI systems (the basis of the Delco/Kalmaker system) from wrecked vehicles (thanks to Mud Bashing for his assistance here!), so, three weeks from his departure date, he handed me the wiring loom from the Camira and, whilst pointing to the empty engine bay, asked me to ‘make it fit?’. Not an easy job when the engine is not in place to make measurements from. Dave set-up a dummy engine with the manifold from the Kombi EFIed engine (which we were going to use again on this engine) and fitted it to the car. I stripped off the kilometres of tape that keeps this loom together in the Camiras and stripped out the un-necessary wires (wires for the dash warning lights, Auto trans control, starter, alternator, etc), then I laid the loom over the engine to see which way was the easiest to run the wires. Unfortunately, Dave already had a large-ish hole in the right side of the firewall from the Kombi EFI loom, and was keen to re-use it – I say unfortunately as I feel that if the loom was run from the left side of the car it would have fitted much easier and more neatly, however….

A decision was made to mount a number of components on the left hand side of the firewall. Dave made a neat metal plate that attached to the firewall, and could be removed from the engine bay, without the need to undo nuts on the inside of the car (2nd fuel tank in the way). This plate was to hold the coil (from the Camira), the MAP sensor, Camira under bonnet relay/fuse box (which is part of the loom), a couple of relays (one for the heated oxygen Sensor, and the other for the low pressure fuel pump) and a 6 way fuse box. I ran a couple of extra wires from this plate to roughly where I had guessed the plug from the oxygen sensor would end up, and also a couple of wires back up the loom, through the firewall grommet and to a small switch which would be taped to the outside of the loom and used to switch the low pressure fuel pump on to prime the surge tank (on initial start up and should the car ever run out of fuel).

I shortened a number of wires and removed a couple of incompatible plugs (Cylinder head temp, Oxygen sensor and the plugs for the fuseable links (unfortunately the fuseable links did not come with the loom – hence the 6-way fuse box)). At about this time I had to go to work, so I suggested that Dave mount the fuel pumps and components that were to be mounted on the metal plate so that I could accurately cut to length and terminate the wires in these areas – I also suggested that he mount the exhaust on the dummy engine so that I could do the same for the Oxygen sensor.

Sunday:

I had some sleep to catch up on and also a number of household duties to do, so Sunday started rather late! When I made it over to Daves place I found the car minus dummy engine, but plus most of the ‘plate mounted’ components, and the fuel pumps and surge tank were in the process of being fitted. Once this was complete I refitted the dummy engine and laid the loom out again. It was decided that the temp sender would be fitted to the RHS of the engine, so I was able to cut and terminate this wiring along with move two of the injector plugs over to the other side of the engine. It was determined that the wiring to the dissy would be ‘almost’ long enough (we are using the dissy from the Camira – but without the fan shroud in place we couldn’t tell which way the plugs on the dissy would face, therefore ‘almost’….)

Most of the rest of Sunday was taken up with ‘help me fit this …’ or ‘can you hold this whilst I…’ type jobs as Dave worked on almost every other part of the car. I should point out that the engine that was to be fitted was still in bits in the corner of the garage as the heads and crank had not yet arrived.

Saturday:

I wired up the oxygen sensor relay and the low pressure fuel pump relay. At my request, Dave had tidied up the vehicle wiring during the week (tail lights, gauges, voltage regulator, etc) so I cut and terminated the wires to the tacho and ignition switch. Dave was almost finished building the engine as all his parts had arrived during the week and he felt that the engine would be in the car by late Saturday or early Sunday. I thought that we should make a huge effort on Sunday, and try to get the engine to start before we called it a day. I began to terminate the wiring that was going to the fused side of the 6-way fuse box, but didn’t quite get it finished before I had to leave for work.

Sunday:

I arrived on Sunday to find the new engine mounted in the engine bay and Dave cursing and swearing…. he had been attempting to finish the fuel hose ‘plumbing’ and couldn’t find the fuel pressure regulator anywhere. He had found the old fuel hose, clamps and everything else associated with the regulator, but not the regulator itself. After another hour and a half of searching around the garage, we decided to go and extract a spare, untested, second hand reg from a box at the back of Daves garage (anyone who has seen Daves garage would know why 1) he lost the original reg and 2) why we didn’t just use the spare reg in the first place!). Having found the spare reg, and whilst looking for a screwdriver to remove a piece of old fuel hose, I happen to stumble across the original reg…bugger!

I went back to what I was doing when I left on Saturday and once this was finished we fitted the Dissy drive and distributor. It was now clear that one of the dissy plugs would just reach, but the other plug would be all of an inch short. After a quick search through my box of spare wire, I managed to find enough of the correct coloured wire to make the extension and I turned my attention to the wiring of the Oxygen Sensor.

Dave had purchased a 3 wire Oxygen Sensor (suit Ford Falcon ) and had visited a local wrecker to get the other half of the plug from a wreck. We still were not sure of the exact length of the wiring for this sensor as it had to run along the engine cage and around an exhaust heat shield, neither of which had yet been fitted. We took a guess and soldered the off cut plug to the loom and plugged in the sensor. As far as I was concerned, the loom was complete – however, the power supply wiring to the 6-way fuse box was not yet complete, and as this was all that was required to finish the job, I completed these last couple of connections.

Dave was still running around, attempting to fit an extra oil cooler to beside the gearbox, blead the brakes and a dozen other small jobs, so I headed home for dinner. When I returned the cooler had not been finished as the hose fittings wouldn’t fit with Daves new exhaust, but everything else was done. Dave had also fitted an LED in the dashboard to operate as a check engine light (CEL), but hadn’t wired it up as he wasn’t sure of which wire went where. Once this was sorted out I buzzed through the wiring loom using my multimeter to check for short circuits and unexpected open circuits (broken wires), not finding any we connected the battery (with the ECU disconnected) and tested for voltage in all the necessary places. It all appeared good, with the exception of the Oxygen sensor relay activating as soon as the battery was connected (it should only activate when the ignition is on). We decided that this wasn’t a major problem (it was just me trying to be smart and connecting the relays trigger wire to a wire that I thought would only be active when the ignition was on…but is is always active – lesson here volks: always double check with the wiring diagram and don’t rely on ‘should be’ logic), and as it was getting late we thought that we would press on for the start-up.

By this stage we had a crowd of interested onlookers who had decided to come and see what the fuss was about. Dave put about 10 litres of fuel in the front tank and I flicked the priming switch to fill the surge tank (there was still no back window fitted at this stage so I was standing beside the engine). We all did the duck and dive as we check the entire car for fuel leaks and then stood and looked at each other, waiting for the fuel pump to become quiet (indicating that it was pumping fuel). Five minutes later we were still listening to the ‘noisy’ fuel pump and staring at each other…..hmmmm, a problem perhaps? Dave decides to put fuel in the back tank as it is closer to the low pressure (noisy) pump and hopefully would allow the pump to start to work (he was assuming that the pump was unable to prime itself as it was so far away from the front tank).

As Dave returned to the front of the vehicle to switch tanks (activated by a switch on the dashboard), I switched on the low pressure pump using the priming switch to immediately hear the pump ‘go quiet’….the back tank had been selected all along. After we all finished laughing at Dave, who was pointing the finger at everyone else for selecting the wrong tank, we listened carefully to the tank, waiting for the air bubbles to stop as the surge tank became full. Once the bubbles stopped, Dave flicked the ignition on to prime the high pressure pump and fuel rail…but there was nothing but silence….there was no pump operation, not warning lights on the dash, no relay….nothing! This was one of those horrible moments when everything you have done flashes past you so fast that nothing makes any sense and you wonder how long the damage will take to repair and at what cost….

I grabbed my Multimeter and began testing: power at the fuses?; check. Power at the ECU?; check….er …what? Hmmm, Power at the ignition wire?; Nope Phew, not even in my wiring – its something that Dave has done, probably whilst fitting the new wiring to the dashboard. A quick check under the dash shows that there is power at the ignition switch, next stop: fuse box…er….why is there a fuse missing Dave? Yep, the ignition fuse was missing – a quick search of the car and garage turned up a replacement fuse and it was quickly installed as Dave began pointing the finger at everyone again for removing the fuse in the first place.

After the laughter died down Dave turned the key again – this time everything came to life. We switched the ignition on and off a couple more times to ensure that all the air was out of the fuel rail, then we switched everything off and removed the ECU power fuse so that Dave could crank the engine over to build up oil pressure before the start. This is when we discovered that the starter was not in the best condition as it took several twists of the key to get it to fire…however we were too close to having it running to be bothered looking for a better starter (or rebuilding the exiting one) and it was just after 11pm too, so no more time to waste.

Once Dave was happy with the oil pressure situation, it was decided to go for a start. I connected my Laptop and re-installed the ECU fuse. I made a couple of quick adjustments to the set-up (we started by using the Camira calibration), crossed my fingers, and Dave turned the key (a couple of times to get the starter to work!). On the first attempt the engine didn’t attempt to fire, but there was a lot of air returned to the fuel tank, the second attempt was the same as the first, but on the third attempt the engine fired and began to run. It was running lean and backfiring through the intake but before I could adjust anything it stalled.

I quickly double checked my settings and realized that I had accidentally set the cylinder size to 425cc and not 525cc as it should have been. The other issue is that we had no idea what the injector efficiency rating for the stock Kombi injectors was. The best information that we could come up with was to make sure that everything else was correct (engine size, etc) and then increase or decrease the efficiency rating until the engine idles reasonably well.

I selected to injector efficiency option and Dave turned the key several times again. When the engine came to life I decreased the injector efficiency (therefore the injector needs to be open for longer to inject the correct amount of fuel) and the engine began to idle smoothly. We quickly fine tuned the distributor to give the same advance reading as was shown on the Ignition Advance ‘Gauge’ on my Laptop. Fairly happy with the result we began to congratulate ourselves on how easy it had all been…then I noticed that the Check Engine Light was on.

As I began to check the error readout page in Kalmaker, the engine began to run rough – there was clouds of black smoke coming from the exhaust (not to be confused with the clouds of white smoke coming off the new paint on the exhaust!). I suggested that Dave shut the engine down whilst I check out this new development. It turned out that the Cylinder head temp was going high enough to trigger an ‘overheating’ warning, which, in turn, causes the engine to run richer, which then causes the ‘over-rich’ error to be reported. If nothing is done at this point, and the temp continues to rise, the ECU will cut ignition to random cylinders in an effort to allow them to cool (and I would suggest make the car almost undriveable too!). The overheat warning was set to trigger at around 85 deg (quite reasonable for a water cooled car) but the Cylinder head temp on Daves engine was up around 117 deg when we shut it down…no wonder it was upset!

At this point we decided to call it a day (it was 12:45am on Monday morning and I’m certain that the neighbours would not have been all that impressed with us revving a VW engine at that time!). Happy that the engine was running we all headed for bed.

Monday:

Dave and I decided that we would try to get a little more tuning in before we went to our Parents for our usual Monday night roast. I met Dave at his place just as he arrived home from work and we pushed the car out onto the driveway and connected the Laptop, Dave connected the battery and we turned the engine over. It started first go and idled smoothly – I then spend the next hour going through every coolant effected variable checking it to make sure that it would work with the higher temp readings that this engine would create. Once I was happy with that we gave the engine a rev and although it was reasonably smooth Dave was horrified to see the Ignition Advance gauge peek at 58 deg (!). We decided that we would have to have a closer look at the ignition advance map over dinner and packed everything away.

It would appear that the Camiras use up to 58 deg ignition advance – as many of you would know, VW engines should only have around 30 – 32 deg advance. We decided to modify the advance map to reduce any advance greater than 32 deg back to 32 deg. I did this on the Kitchen table and agreed to upload this new map on Tuesday when I was at Daves place re-wiring the Oxygen Sensor relay.

Saturday:

Dave messaged me at work on Friday night asking if I was keen to do a tuning run on Saturday, so early Saturday morning (well it was early for me – about 10am) I arrived at Daves. The car was now fitted with the engine cage, spare tyre, number plates, all the windows, seats, and even floor mats! Whilst I connected the Laptop to the ECU, Dave made a couple more quick checks (like fitting a nut to the steering wheel!). We then climbed aboard and turned the key. The engine came to life instantly and settled to a smooth idle. Dave selected reverse and we began to reverse out onto the road. We both noticed a loud clicking noise coming from the rear of the car as it moved, but as Dave has the ride height set as high as it will go, we suspected a worn CV. Selecting 1st gear and beginning to release the clutch the engine bogged in – we assumed the incorrect ignition advance settings initially, but as we rolled down the hill away from my Brothers place it became apparent that there was much more wrong than that…. For a start 2nd gear was not able to be selected…and that clicking noise was becoming louder…more of a bang now, and it was starting to shake the car every time it happened. I suggested that we not go too fast as I really didn’t need to rocket off into the bushes when that dodgy CV locked up. When we reached the bottom of the hill we stopped at the roundabout, Dave selected 1st and began to let the clutch out and the engine stalled. Several turns of the key later and the engine was running again, but now the gearbox was stuck in 1st and every time we let the clutch out the engine would bog in and try to stall. We eventually managed to get the car out of 1st and tried second from a standstill….it could be selected, but as soon as you began to let the clutch out there was a loud bang as it jumped out of gear and a grinding of teeth (both in the gearbox and Daves!).

The gearbox was stuffed (as you have probably guessed) – putting any more than a very light load on 1st gear would cause the box to lock up and stall the engine, 2nd gear didn’t exist at all, and 3rd was so noisy that you had to yell over it…we never bothered to try 4th. The diff was also missing teeth (which is what the bang was as the car was rolling) The short version of a long (and rather funny) story is that we towed the car home, pushed it into the garage and closed the door. There was no way that this car was going to be ready to go to Frazer Island in a week. I thought that that was the end of that, so after we had all visited the local bakery for lunch my Laptop and I went home.

Later that night I received a number of messages from Dave asking about the condition of a gearbox that I have in my possession that I used the Bellhousing from – ‘its stuffed’ I told him ‘large chunks of metal in the oil - you would be wasting your time with that one’…. ‘but is it worth a go’ came the reply. ‘Nope’ was my answer. ‘I’ll come over tomorrow and take a closer look at it’ was the return message…. OK, there was no way he was going to listen to what I had to say.

Sunday:

Sunday came and went without any sign of Dave… I assumed that he had seen the light and had given up.

Then, at midnight I received a message ‘found another gearbox that I had forgotten I had – have driven the car and it drives fine, do you recon it would be safe enough for me to drive it to work?’ I told him that the ignition advance is all over the place and I have no idea what the fuel settings are like, so basically, no…. ’but it drives fine, I don’t think it would hurt, what do you think’ I could see that I wasn’t going to be able to stop him so I suggested that he take it really easy and at the first sign of trouble turn it off and call me and I would sort something out…some people you just can’t tell.

Monday:

Received a message from Dave: ‘made it to work OK. Engine is really gutless and the ride is so rough that I feel that I need to see a chiropractor now, what are you doing this evening?’…I was impressed and a little surprised – remember that this engine was running on a Camira Calibration with the only real modifications being the top ‘chopped off’ the ignition advance so that it didn’t exceed 32 deg.

At 6pm Daves Baja arrived in my driveway. ‘ouch’ he said as he climbed out of the drivers seat holding his back ‘man, the ride in that thing is shit! I have to do something about that…’. He told me that the engine was running fine under throttle, but it was gutless and it stalled almost every time you pulled up at the lights. I plugged in the Laptop and we went for a short run through Helensvale.

It wasn’t long before I discovered what the problem was – the engine was running a total of 6 deg advance at full throttle and, of course, if you want the car to go faster, you use full throttle! If Dave had just backed off the throttle, the advance would have shot up to 32 deg and the car would have rocketed off. I spent about 5 mins sitting on the side of the road ‘re-designing’ the advance map so that it ran 32 deg everywhere except at very low throttle setting at anything above 1200 RPM. ‘Try that’ I suggested to Dave, who promptly selected 1st gear, buried the right foot and dumped the clutch (expecting there to be very little change in the engine performance). Two large black stripes and a cloud of tyre smoke along with a handful of opposite lock was the result. Dave snapped second for a similar result….’that’s a little better’ he commented! I noticed that the advance gauge was still only showing 28 deg max advance, but I wasn’t sure why. Dave dropped me off at home and promised to come back after his roast dinner (no roast for me this night 🙁 )

At 9:30pm the Baja was back in the driveway and I was uploading a couple more changes that I had made to the Calibration into the ECU. I had found another temp related variable that retarded the ignition if the engine temp was too high, and we were triggering this with Cylinder head temps of around 122 deg. A short run around the block had a smile on Daves face and a reading of 32deg on the advance gauge. We decided to do a long highway run to see how close to correct the fuel setting were so we found our way out onto the M1.

The reading were initially high (rich) and were a little erratic and as I began to think about changing settings something odd happened. The reading from the O2 graph suddenly dropped to the bottom of the graph and the engine lost power….’what did you do?’ was Daves first comment. ‘Er… nothing’ was my reply. Dave backed off and the graph indicated that everything returned to normal. Strange. Then only a few seconds later it happened again, and again, and again. We pulled over and the car idled smoothly. Hmmm. I spent about half an hour looking for some sort of problem in the fuel maps, but I was unable to find anything. I was also concerned that the overtemp situation had come back to haunt us again, but this was ruled out as it would make the engine run rich, not lean. Eventually I came to the conclusion that it was nothing to do with EFI settings as was probably a mechanical or electrical issue. I suggested that we should check the power to the high pressure fuel pump and the filters too as it just looked and felt like it was running out of fuel. I also suggested that we trick the temp sender by installing a resistor in the line between the temp sender and the ECU. Although not strictly necessary, it would eliminate the chance that the overtemp issue would cause any further issues.

Tuesday:

Dave took the Baja to work again today and spent some time lowering the ride hight, checking the wiring to the pumps, and replacing the filters – this is where something interesting came up. Dave had reused the filters that were on the car with the last EFI system, but they had dried out whilst the car was being re-built. When he checked the filters in the low pressure side of the system (2 filters – one for each tank) he noted that the element in the filter for the front tank was ‘sucked in’ toward the output side of the filter. It was really badly disfigured, but it hadn’t been like that when he had reinstalled them. The filter for the back tank was also showing sign of the same problem, but nowhere near as bad. When the hoses were disconnected from the output side of the filters only a few drop came out, despite the fact that they were still connected to the almost full fuel tanks. The filter on the high pressure line also only had a very limited fuel flow when it was disconnected (OK the pump was off, but most of the fuel remained in the filter). We think that the filters must have dried out and solidified – has anyone else heard of this happening? Dave fitted 3 new filters and after work took the car for a long run up the M1 before dropping into my work to pick up the resistor for the temp sender that I had set up for him during the day. He reported that there was now no problems. Whilst he was working on the car, he also noticed that the bolts holding the exhaust on had worked loose – we believe that this may have had something to do with the erratic reading we were getting from the Oxygen Sensor.

[ Edited on 29-12-2005 by ratty 63 ]

Saturday (Christmas Eve)

Dave had driven the car every day this week and it was running OK, but still stalling when you came to a halt at traffic lights.

Mid morning Dave arrived in my driveway again. This was our last chance to sort out any problems. I still wasn’t took keen on him spending a week and a half thrashing a partly tuned injection system on Frazer Island, but nothing was going to stop him now! We spent a couple of hours cruising up and down the M1 and making find adjustments to the injector efficiency setting. The graph only varied slightly when cruising (normal) and showed that the enrichment was working when you buried the right foot. A quick run around town showed that a few changes that I had made to the idle controls had improved the stalling issue, but not completely cured it (down from 85% to about 15% chance of a stall). Once again our ‘play time’ was cut short as I had to go to work.

Sunday (Christmas Day)

Limited time to be playing with and EFI system on Christmas Day, but I had had a couple of ideas whilst I was at work on Saturday night and I was keen to try them out. I did a quick upload of my altered Calibration late Sunday afternoon and Dave went for a quick run around the block. He reported that the car returned to an idle much better and it hadn’t stalled once. As he was leaving for Frazer Island early the next morning, this was the last opportunity that I had. We both agreed that this was as good as it was going to get under the circumstances.

Monday (Day of Departure)

Can’t tell you too much at this stage, but a message on my phone tells me that he made it to Frazer Island OK. Had two flat tyres on the way there – one of them repairable, the other had to be replaced – but most importantly the EFI system is running fine.

Thursday

Received another message. All is well. Using a little more fuel than he had expected, but he reckons that he shouldn’t have lowered it as it is now dragging its butt through the sand with all the gear on board.

[ Edited on 29-12-2005 by ratty 63 ]

Well that certainly is a story and a half.....
I hope he's taking plenty of pics.
Is he planning on being at Double Island beach trip on 7 & 8th Jan?
Graham

This has been a great read Ross.
What! Dave blame everyone else for doing something wrong no not Dave!

Will be waiting for info on how it is going .

Kath and I have a Xmas prizzy for Steven, We brought it down for the Light cruze but you were not there.
How is the family all good we hope?

Marc & Kath

Hey Ross
Have you heard from Dave lately?
and hows the BAJA going?

Marc

Great read .. look fwd to hearing how it runs

great effort mate, mind posting a few pics of the system you made inc controllers?

kkk

Great work.

Ross, apparently you fabbed your own runners. Any chance you could post some pics of them, especially the injector bosses.

11CAB wrote:Well that certainly is a story and a half.....
I hope he's taking plenty of pics.
Is he planning on being at Double Island beach trip on 7 & 8th Jan?
Graham

...and the story will continue after he returns from Frazer Island - I'm keen to clean up the tuning - mainly the ignition advance and idle to put the finishing touches on. The Auto-tune function should have taken care of the fuel map by that stage....

I don't know if he will be on that trip...I think that will be only a day or two after he is due home....I could be wrong though

Midlife crisis wrote:Will be waiting for info on how it is going .

Kath and I have a Xmas prizzy for Steven, We brought it down for the Light cruze but you were not there.
How is the family all good we hope?

...so far I haven't heard any more from Dave (and you can be sure that if there was some sort of problem with the EFI I would have heard about it by now!).... no news is good news! I will message him again tonight and see if I can get an update....

Sorry we wern't at the Xmas Lights cruise - my work wouldn't allow me to take the night off as they don't have enough Management staff to fill my shift - they did allow almost all my staff to take the night off though :bounce idiots! Michelle was planning on attending by herself, but as the time to leave approached it all became too hard.

I will try to catch up with you soon. We are all well and Steven is doing great.

koolkarmakombi wrote:great effort mate, mind posting a few pics of the system you made inc controllers?

...pics are no problem - any particular part of the system that you would like to see? I have pics of some of the work in progress, however as we began to run short of time the camera didn't get used as much as I would have liked. I can take pics of almost anything once Dave gets back from Frazer Island. I'll post what I have in the next couple of posts....

56astro wrote:Great work.

Ross, apparently you fabbed your own runners. Any chance you could post some pics of them, especially the injector bosses.

Thanks! The manifold that Dave used with this engine is simply an EFI Type IV manifold that has been cut and welded to rise up over the top of the Porsche style fanhousing. It still uses the standard Kombi Plenum chamber (with a couple of mods done to it to allow the Air Temp Sensor to be fitted). I will post what pics I have of this manifold, and will take some better ones once Dave gets back. Using the standard Kombi (type IV) manifold allowed Dave to use all the Kombi hardware (injectors, injector clamps, fuel rails, etc) to make the system look and operate like standard. If you are planning on using this system in a stock Beetle, check that the fuel rails and injectors will fit in the engine bay as they are a very close fit in the Baja - and it has a body lift kit fitted (so the engine 'sits lower' in the engine bay...

I can't wait to hear all the storys from this trip - I'm also particularly interested to hear of any problems they had with the vehicles and the EFI system. As yet I haven't driven the car, only been a passenger, so I am hoping that I will get the chance to drive it so that I can judge the drivability for myself...that might sound like a strange thing to say, but Dave sets his cars up very differently to the way I set mine up (he has only two pedal positions - up and down and the cars are set up to be driven like this too).

I'll let you all know how things are going as soon as I hear back from Dave....

R 🙂

This is the old, Kombi based, EFI system that Dave was running. I posted this pic mainly as the manifold is fairly visible. I haven't got any pics of the changes that were made to the plenum chamber at the moment - I will get a few and post them here once Dave returns.

R 🙂

Pic 1: The engine bay with the Dummy engine and manifold from the old EFI system fitted. This shot was taken very early on the first weekend. You can clearly see the 'exploded rainbow' wiring loom which I had layed over the engine to see where the best locations were to fit the various components.

Pic 2: Close up of Pic 1. The Camira dissy is fitted to this engine (actually, it had just been dropped in the hole as there was no dissy drive - it's not seated down as far as it would normally go.)

Pic 3: The Component panel in the engine bay. Unfinished in this pic. The Camira Coil is mounted on the left, then the Camira underbonnet relay/fusebox is mounted tot he right of the coil. To the right of that is the MAP sensor, with space for the Oxygen sensor heater relay and the low pressure fuel pump relay. On the right is the 6 way fusebox.

R 🙂

This pic is of the priming switch - it operates the low pressure fuel pump indepentantly of the ignition so that the surge tank can be filled should it ever become empty, without having to put the high pressure pump at risk.

R 🙂

...more pix to come....

R 🙂

I like the work. Keen to see if you make any changes.

Helbus wrote:I like the work. Keen to see if you make any changes.

As far as the mechanical side of things are concerned, the system is finshed, there will still be a couple of minor tuning changes (particularly to the idle)...

....just got a phone call from Dave. He has just blown his 4th tyre for the trip and is stranded in Gympie...looks like I may have to go for a drive with a couple of spare rims and tyres tonight...Bugger.

R 🙂

Nice work! Thanks for sharing the story with us 🙂

ratty 63 wrote:....just got a phone call from Dave. He has just blown his 4th tyre for the trip and is stranded in Gympie...looks like I may have to go for a drive with a couple of spare rims and tyres tonight...Bugger.

R 🙂

LOL
He's your little bro isn't he :P

great stuff.

You've probably got this listed on the other EFI thread. But that's a million mile long now. Three simple questions.

1. You've got kalmaker software on your laptop have you?

2. Where did you get it and how much did it cost?

3. What model of camira ECU are you using?

I can answer some off that the model is je its around 88 89

And next well this year now there will be a 2411cc type 4 getting this treatment i hope

56astro wrote:

ratty 63 wrote:....just got a phone call from Dave. He has just blown his 4th tyre for the trip and is stranded in Gympie...looks like I may have to go for a drive with a couple of spare rims and tyres tonight...Bugger.

R 🙂

LOL
He's your little bro isn't he :P

....is it that obvious? 😀

Turns out that he couldn't wait for me to finish work first (fair enough too!) so it became another job for Mud Bashing - how'd you go Tim? What time did you get back?

pete wood wrote:great stuff.

You've probably got this listed on the other EFI thread. But that's a million mile long now. Three simple questions.

1. You've got kalmaker software on your laptop have you?

2. Where did you get it and how much did it cost?

3. What model of camira ECU are you using?

The answers are on the other EFI thread, but so that you don't have to wade through a million posts...

1. Yes, the Kalmaker software is running on a cheap Pentium Laptop that I picked up a couple of weeks ago.

2. The software was suppied by Injection Connection in Perth. They are the distributors of the Kalmaker system. Check out their website and forum if you are interested: http://www.kalmaker.com.au/. The purchase price of the ECU and real time converted ECU is $1100 +GST. Not the cheapest around...but you get what you pay for....

3. The ECU is a Delco 808, which is found in a number of Holdens (and GM based vehicles). The ECU needs to be modified to run the real-time board, which then allows the ECU to be connected to a Laptop. The JE Camira ('88) is the best vehicle to use for parts, but the same ECU is also used in the JD Camira (single point injection), the N13 Pulsar (single point injection on the 1.6 and multi on the 1.8 ... but don't use the wiring loom!), the VP, VN and VQ Commodore also uses the same ECU, but the wiring loom is also very different and not as easy to convert as the JE Camira loom.

I am also told that the 1.7 EFI Lada Nivas use the same ECU, as do the 1.5 Daewoo Lanos (open the bonnet on a Lanos and compare what you see to a JE Camira....they are identical!). I don't know what the wiring looms are like in the Lanos (they don't look that different), but the Niva uses a pully mounted crank angle sensor and a coil pack to operate the ignition....if you come across one of these that no longer needs its wiring loom and sensors please let me know!)

There are also a stack of GM produced cars in the US that use the same ECU....

You can use a stock ECU from a Camira/Commodore if you burn the Calibration to your own EPROM, but this doen't allow you to make 'live' adjustments, or connect a Laptop so that you can see what is going on.

I will be purchasing another real-time converted ECU in the near future as I plan on using this ECU on all my cars, and as I now understand how it works (well most of it :duh), and I have the software, I might as well....

R 🙂

Well you know went i left i drove straight from then to gimpie and me haveing more ass then class i told dave that i would be there at 9.10 when i was just past the entertainment center exit and dave sent me a msg at 9.09 asking where i was and at that moment i pulled up next to him as the clock ticked 9.10 he said that i had more ass than class so after i chat about the fraser trip that i missied we got under way changing his treds and started on the trip home after droping the old treds and jack at daves house i got home at 12.45 and if you want to know how far it is from maries to the matillda servo at gimpie its 215km each way

i spent a bit of time with dave and barry on fraser
and well lets say he aint that happy with the new rebuild
verry happy with motor
but the tyres( 3 puntures in 4 days and nearly bald all round) ,starter, ride hight , steering are all getting seriously modified

good luck dave
see you at double island
with the new treads
terry

yeah i was up there today and he was starting on the mods his main complaint aside from the tires was the gearing sucked but he then went for a drive in the bush and loved it he said it just crawled its way up the stepest tracks and i don't think he will have a female co pilot this time she was not inprested with all the stopping to fix tires

mmmm so I heard he got bogged heaps and couldn't stay on the tracks.

Let him know I have a few subie engines around if he wants to upgrade to sumthin with real power 👍

ba ha ha ha ha ha ha now thats funny real power and i don't think having one will help his getting stuck problem in fact it would make it worse seeing they are heavyer than a type 4 hang down more than a type 4 so when he got stuck cause of one of those things it would sit there in the hot sand and sshshshshshshpooff oh shit there is all this green SHIT on the ground you should fill them up with just striaght water so when they boil you can have a cup of tea😁

dude the only time we needed a cup of tea was when we all were waiting for dave to get his car unbogged
he was in the same postition i was in last year

so yeah subaru did win this year HANDS DOWN
and will beat any vw powered car without effort in anything it does
even barry didnt get bogged as much
hehehehehehehehehe
terry

ever concidered using a michrotech? much the same price and a very simple and effective unit

76bug wrote:ever concidered using a michrotech? much the same price and a very simple and effective unit

Yep, but they didn't do half of what the Delco can. I have an old Microtec that I was given by a friend - it needs repairs but once this is done I will set it up on a dummy engine and see what it can do.

I spent months looking at the different ECUs, but when I saw a Delco/Kalmaker system in use in a friends car, there was no comparison. I also saw vehicles fitted with Microtecs, Wolfs, and a couple of others....although I feel that some of them wern't set up correctly, some of them didn't drive very well at all!

R 🙂

so wot can the Delco/Kalmaker system do?

Everything but boil water but well leave that to the subies trying to keep up:P

MUD BASHING ANYONE ?? wrote:Everything but boil water but well leave that to the subies trying to keep up:P

So how was the beach Tim ?? Must have missed your VW EFI powered Baja there .. hope it is going in time for my next back yard bash .. I need some cars to fill in the ditches all these subie powered offroaders are digging ....:kiss

76bug wrote:so wot can the Delco/Kalmaker system do?

NOT give your baja the ability to get up hills
and
DO give the ability to get your baja bogged everywhere
hahahahahahahahahahahahahahaha
terry

So how was the beach Tim ?? Must have missed your VW EFI powered Baja there .. hope it is going in time for my next back yard bash .. I need some cars to fill in the ditches all these subie powered offroaders are digging ....:kiss [/quote]

yeah that would be like me when i was there last year and me missing the mega manx thats taken how long to build its up to 8 years now and its now where do you know? and as for digging holes yeah that might be all there good for they never get anywhere unlike a aircooled manx that got up the hill which has stopped every water boiler that has TRYED it before it to bad that one didn't make it to the fourm on film
gees im a prick but a funny one

NOT give your baja the ability to get up hills
and
DO give the ability to get your baja bogged everywhere
hahahahahahahahahahahahahahaha
terry [/quote]

thats funny dave tells a different story that thermo fan must be rooted by now hay it was always on and a quote i heard from bazz oh just wait a minute i have to turn it off its to hot gees guys sign me up for one of those great engines i would love to have one ha ha ha ha

Now I know that you guys are just trying to stir each other up - I have heard both sides of the story and it sounds like everyone had problems getting bogged...not that that has anything to do with EFI.... so perhaps if you wish to 'discuss' this Subi Vs VW topic, you should start your own thread (where those who care can go to hear the facts 😀 ) rather than take over an unrelated thread?


I haven't had a chance to have a close look at Daves EFI system since he got back, we have both been rather busy. I do know that Dave has lashed out and fitted a full set of new tyres to the Baja (are you surprised?). I have worked out why it is still stalling occasionally (my fault for not tuning the entire deceleration enrichment map - only the part above 900RPM - I can explain further if anyone is interested). Dave also tells me that it is allowing unburnt fuel into the exhaust when he backs off from high revs, which then ignites with a large flame out the tailpipe... I'm not yet sure why its doing this, but Dave suggested we leave it like this for a while as its great for scaring other motorists that drive a little too close to the back of the car.... 😀

I will try to get a few more pics and hopefully spend some time fixing up the minor tuning issues over the next couple of days.... I'll post the results.

R 🙂

Hi Ross
Yep keep the details coming it is a great read.

Marc

Yep good read as always ross - you want to be careful or you might end up running a wiring loom modification and Delco tuning business 🙂

Any motor that is still air cooled and can leave black marks on the road whilst wearing 31x10.5R15's is making some serious torque. So I am impressed 🙂

is the delco a 32bit computer?

76bug wrote:is the delco a 32bit computer?

Nope. Unless you have everything in the car controlled by the ECU I can't imagine why you would need 32 bits...? Have a look at the Delco/Kalmaker web site for all the info on the ECU and software that I am using: www.kalmaker.com.au

Update:

I went for a drive with Dave the other day, just to get some idea of what was happening when the engine was returning to idle... It's a fairly simple issue that is (basically) being caused by the fact that we are not using a Vehicle Speed Sensor (VSS) at this time - we will be installing one in the future.... I'll try to keep this explanation as simple as possible....:

As an EFI vehicle idles, the ECU monitors the engine speed and makes changes to the IAC (Idle Air Control - controls the amount of air that bypasses the throttle plate) to keep the engine as close as possible to the requested idle speed. This part works great on the system that we currently have.

However, to enter a idle, a number of situations must be met: The throttle plate must be closed, the engine must be below 1500RPM and the vehicle must be stationary - now without a VSS, the ECU can't tell if the vehicle is stationary or moving, so it assumes that it is stationary...and this is where the problem begins.

If you decelerate back to below 1500RPM, with the car in gear and the clutch out the ECU assumes that the car is coming back to an idle. It trys to bring the idle speed down to the preset idle speed (900RPM in this case) as quickly as possible by reducing the amount of air bypassing the throttle plate through the IAC. The IAC is a stepper motor - the less steps that it is open, the less air that get through. Now if you are decelerating in 3rd gear, it will take, say, 15 seconds to reduce your speed enough to get from 1500RPM to 900RPM - by this time, the IAC has wound back to 0 steps (completely closed) as it trys to control the idle (this only takes about 2 seconds). If you then push the clutch in with the IAC at 0 steps, the engine has no air, and no 'force' from the gearbox to keep it moving, so it stalls. Unfortunatly the IAC wont react fast enough to catch the stall at that lower rev range.

Of course, if we fit a VSS, the ECU will know that the car is still moving, and will not try to return to an idle, therefore having no effect on the IAC steps...instead it will try to hold the revs at about 1500 RPM, assuming that you will either want to accelerate, or you are about to push the clutch in, but either way, it won't stall.

So, after all that, it would appear that we will have to fit a VSS - which we were going to do anyway.

It has been suggested that I open the throttle stop just slightly so that when the IAC is at 0 steps, the car will still idle (albeit just!) - this is a quick and bodgy fix, but it might just work until the VSS comes along.

As for the flames coming out the exhaust when backing off at high revs, I am fairly certain that that is being caused by their being too much deceleration enrichment (like an accelerator pump, but used when decelearting - stops popping/burbling in the exhaust when set right). I will be catching up with Dave tonight and will back the Decel enrich. off and see what happens - the flames do look cool at night though...you wouldn't want to get too close to the back of the car! :duh

I am also planning on reducing the amount of ignition advance in the lower throttle and RPM ranges to make acceleration at lower throttle setting less jumpy.

I'm having a ball with this system! I'll let you know how I go with this next batch of fine adjustments....

R 🙂

Ok Volks,

Just found a few more pics for those interested....

...its very hard to take pics of this engine as Dave has a spare wheel carrier mounted over the engine (attached to the engine cage) but here you go...

R 🙂

Pic 1: Modified manifold (passenger side of vehicle) - you can also see the electrical components board that we mounted the coil, MAP sensor, relays and fuses on.

Pic 2: another shot of the top of the manifold - the electrical connector in the top of the plenum is the Intake Air Temp sensor - this is mounted where the cold start injector was in the factory system.

a 32bit computer just gives u a better resolution i.e. wen u are setting up ur matrix map u will have a map that is 32x32 end result more rpm vs fuel point to tune.

really?

My VE map is 9 x 15 = 135 load points (with the rev limit set to 5500RPM - add another 9 points to the map for every 400RPM)...

...thats more than enough load points for most vehicles, after all, the ECU that I am using is the same ECU found in many of the GM produced vehicles - and they appear to drive OK....

Update:

Spent another hour roaring around the neighbourhood late last night and learn a few things:

1. Don't try sitting in the passenger seat of my Brothers Baja and staring at a laptop screen whilst tuning for WOT in the dark after eating a roast and a large lump of mudcake :puke

2. The autotune feature of the Kalmaker/Delco system only works when the 'coolant' temp (or head temp in this case) is between the upper and lower pre-set limits, therefore if you are going to change the way the coolant temp sender reads by inserting a resistor, make certain that it is still within range - the resistor that I used dropped the reading to below the pre-set limits and therefore it would only autotune when the CHT was in the 'hotter than normal' range....D'oh. Thankfully, like everything else on this system, I can change the pre-set autotune levels so thats not a major problem.

3. Not only was the deceleration enrichment too much, but the acceleration enrichment was not enough - particularly at higher revs - causing a rather unenthusiastic response when the throttle was opened further at high revs (like on the freeway) - this too was adjusted.

4 And finally, Daves engine likes to run rich at an idle (as to most engines) - leaning it out at an idle, even slightly made it run run like a pig!

We are still yet to try the throttle stop adjustment to see if we can fix the stall at an idle... I'll keep you informed.

R 🙂

i had a look at the website, seems like a great idea if u own a comodore as u dont have to change the computer, same concept as ECUTEK which i have played around with, as far as installing this system on a vw i think it is a little bulky.

Thsi is agood read Ratty and will be useful to me when I Foinally get around to installing my $80 Camira system on my Bug

Yes, VSS input and idle anticipation is CRITICAL to the drivability of the Delco/Holden PCM for the reasons you explained very well. I learnt that the hard way with a couple of Holden V6 and V8 conversions, and a transplant onto a Ford EFI V8, I was involved with.

You may want to drill a small hole (start with say, 3mm - I think the HSV had 8mm) in the thottle blade to assist the idle - HSV did this on the 200kW SV5000/VP GTS versions of the 5.0 in the late 80s/early 90s. Cranking open the throttle will upset the TPS... which will give you another set of dramas!!

VWCOOL wrote:Thsi is agood read Ratty and will be useful to me when I Foinally get around to installing my $80 Camira system on my Bug

Will it fire 8 plugs??

56astro wrote:

VWCOOL wrote:Thsi is agood read Ratty and will be useful to me when I Foinally get around to installing my $80 Camira system on my Bug

Will it fire 8 plugs??

Won't need to at first as I only intend to do a single TB conversion on my base 1916.

BUT I can see no problem with running one of Stans dissys fitted with the guts of the Camira distributor to splice the Delco to operate in team with 8 plugs. Or maybe use double-ended wasted-spark like the V6 configged with an crank angle sensor, and an ignition map yadda yadda from a 304 V8 to run another coil pack... Maybe?? That's a bit beyond my experiecne but I reckon it's possible...

Give Injection Connection an email on Injection Connection [tpiconn@space.net.au] (Alan Gibbs). He knows everything there is to know about Delco / Kalmaker.

56astro wrote:Will it fire 8 plugs??

I agree with VWCOOL - there is no reason why it couldn't - the Delco only tells the coil (or coils) when to fire. The ECU doesn't control which plug(s) are firing at any given time, so fitting one of those dissys that I have seen being used with the dual plug conversions shouldn't be a problem...or you could get really fancy and go for a multi coilpack set-up....

R 🙂

VWCOOL wrote:Thsi is agood read Ratty and will be useful to me when I Foinally get around to installing my $80 Camira system on my Bug

Yes, VSS input and idle anticipation is CRITICAL to the drivability of the Delco/Holden PCM for the reasons you explained very well. I learnt that the hard way with a couple of Holden V6 and V8 conversions, and a transplant onto a Ford EFI V8, I was involved with.

You may want to drill a small hole (start with say, 3mm - I think the HSV had 8mm) in the thottle blade to assist the idle - HSV did this on the 200kW SV5000/VP GTS versions of the 5.0 in the late 80s/early 90s. Cranking open the throttle will upset the TPS... which will give you another set of dramas!!

Yep - I anticipated some other sort of problem when we played around with the 'at rest' position of the throttle blade, but as it only requires a screwdriver and some spare time we will still give it a go... just so we know what sort of problems to expect! (If you don't try, you can't learn 😀) I hadn't thought of drilling a hole in the throttle blade, but it makes sence.

We are currently chasing prices on VSS's to suit the VW set up - for those who would like to know, a VSS to suit a VW is fitted between the back of the speedo and the speedo cable and provides the ECU with a set number of electrical pulses per wheel rotation. So far we have come across two different VSS that will fit the VW speedo. Their specs are different (different number of pulses per rotation), but as I can set the number of pulses/wheel rev in the ECU software this is not all that important. I am not certain about any other differences (does anyone here have any experience or knowledge about VSS's? Are they compatable with any ECU?) It would appear that there is not a lot of public knowledge about VSS's - most people just fit the one they are told to fit and it works. There is a $50 difference between the two models of VSS that we are looking at. The most info that we can get on either of them is the number of pulses/rotation and that they fit a VW speedo....

R 🙂

ratty 63 wrote:

56astro wrote:Will it fire 8 plugs??

I agree with VWCOOL - there is no reason why it couldn't - the Delco only tells the coil (or coils) when to fire. The ECU doesn't control which plug(s) are firing at any given time, so fitting one of those dissys that I have seen being used with the dual plug conversions shouldn't be a problem...or you could get really fancy and go for a multi coilpack set-up....

R 🙂

Ross,

Could you elaborate a bit on this. If you have a crank-fired system and 4 twin-post coil packs (wasted spark - 2x1/3 & 2x2/4) you would need a 4 channel ignition system, right?

If you use an 8-post dizzy, what controls the spark timing?

On the Injection Connection website it states that the Kalmaker software "features complete self tuning in Closed Loop and Open Loop."

This sounds like a huge benefit, especially when you compare it to the price of the Autronic set-up.

Do you know if is a similar system to the Autronic "auto-tune" (saving on hours of dyno time)? Have you used the self-tune feature Ross?

56astro wrote:Could you elaborate a bit on this. If you have a crank-fired system and 4 twin-post coil packs (wasted spark - 2x1/3 & 2x2/4) you would need a 4 channel ignition system, right?

I haven't yet set up my crank fire system, but as I understand it if you were running it with 8 plugs (two per cylinder) you would have to run a 4 channel ignition system.

If you use an 8-post dizzy, what controls the spark timing?

The Delco ECU only controls when the spark is fired. The decision as to which cylinder the spark is delivered is left up to the coilpack or dissy. The easiest way to set up the ignition on an 8 plug engine would be to use the 8-post dissy with an electronic trigger (I'm not all that familiar with the 8-post dissy - do they use points or some other way of triggering the ignition? I assume that both plugs in one cylinder fire at the same time - is that correct?).

As I understand it, the electronic trigger sends a reference signal to the ECU at a preset point before the spark is required. The ECU then calculates how much spark advance is required (dependant on MAP and engine revs (basically)) then works out how long to wait from the reference signal to the point that the spark is required. Once this time has elapsed, the ECU sends a signal back to the spark controller (either the coil or coil pack) to fire the ignition.

If you were using a dissy (normal 4-post or 8-post) the spark would be delivered to whichever post on the dissy that the rotor button is pointing to - exactly the same as a standard dissy.

The coil packs that are used with the Delco ECU have to be able to determine which cylinder to fire (and therefore which 'pack' to fire). This would be the tricky part of using a coil pack set-up with an 8-plug engine. There is no reason why it couldn't be done - it would just need some careful thought and planning.

Using multiple coil packs is possible - have a look at the Kalmaker web site - there is a pic and brief description of a V12 Jag that has been converted to run on a Delco. It uses 2x6 cylinder coil packs (from a Commodore). If they can make something like this run happily, then firing 2 plugs/cylinder on a 4 cylinder should be easy!

If you are interested in using coil packs rather than a dissy, perhaps a question on the Kalmaker forum would be a good place to start your research. There are heaps of people on there that know a lot more than me about these sorts of conversions! (Alan Gibbs, Frank 'Choco' Munday, etc)

56astro wrote:On the Injection Connection website it states that the Kalmaker software "features complete self tuning in Closed Loop and Open Loop."

This sounds like a huge benefit, especially when you compare it to the price of the Autronic set-up.

Do you know if is a similar system to the Autronic "auto-tune" (saving on hours of dyno time)? Have you used the self-tune feature Ross?

We are just starting to use the auto tune feature on my set-up. I would have liked to have used it right from the start (it was our intention) however an incompatability between the temp sender and the software settings meant that the auto tune function only 'switched on' when the CHT was higher than normal (ie: not very often).

The auto tune feature required a couple of things to be within range before it will activate. It requires the TPS reading to be stable for a period of time, the MAP signal to be stable for a period of time, and the engine must be up to operating temp...and that last point is where we had our issues. Normal operating temp for a water cooled vehicle would be around 80 - 100 deg - normal operating temp for an air cooled cylinder head was (an indicated) 115 - 130 deg, which would cause the ECU to go into 'overheat damage control' mode (runs super rich and causes the check engine light come on). Although it is possible to reset the temp ranges using the software, we decided to add a resistor in the wiring between the temp sensor and the ECU (a decision made due to time restraints). This resistor dropped the indicated temp down to around 60 - 75 deg - which was not hot enough to trigger the auto tune function.... ooops! :duh

I didn't realise that this was the case until just last week - so I have changed the auto tune temp range to suit the readings that we are getting from our engine and it is now auto tuning. I have not yet seen any results from the auto tune (haven't seen the car since I switched it on), but I will let you know how it works soon.

I haven't looked closely at the Autronic system, but I would assume that it would be similar... I do some more research and let you know....

R 🙂

:thumb

I don't know for sure, but I thought the 'cell-map learning' (auto tune) was only introduced in 1993 VR...? so wouldnt apply to teh earlier Camira/Plusar PCMs. Or is it there (or can it be 'enabled') on the earlier stuff as well?

Its there on the early stuff too - all the 808 series ECU's have it (or at least with Kalmaker they do). Ratty mentioned this on his first EFI thread.

Doug Sweetman wrote:Its there on the early stuff too - all the 808 series ECU's have it (or at least with Kalmaker they do). Ratty mentioned this on his first EFI thread.

Yep. It should be noted that the Kalmaker modified 808 that I am using is slightly different to the stock units in the Camiras/Commodores/etc. This is one of those changes....


I might clarify that by explaining that there are two maps that the ECU can 'automatically' update on the Kalmaker modified ECU. The stock unit can only update one of these maps.

The Modified ECU and the stock ECU have a VE correction map (or Block Learn) that is updated very slowly over a long period of time. This was included in the factory ECU to allow a correction 'value' to be applied to the VE map. This is done over the life of the engine to make allowances for slight changes in engine performance/quality/etc as the engine gets older. I believe that this Block Learn map is cleared when the power is removed from the vehicle...

The Kalmaker modified ECU also has a VE update function, which is basically a self tune function. It looks at the oxygen sensor voltage and then (providing you are meeting a number of other criteria) updates the main VE map so that the Air/Fuel ratio is correct.

When tuning a new car/engine with the Kalmaker system, the idea is to turn the Block Learn feature off, then turn the VE update function on. Then drive the car for a couple of weeks, trying to use all areas of the VE map for as long as possible (ie: low revs/low engine load, low revs/high engine load, high revs/low engine load, and every other possibility) This will allow the VE update function to see the output from, and correct as many VE maps points as possible. Once this is done and the maps are as correct as possible, the VE update function should be turned off, and the Block Learn should be turned back on. Then, assuming that you don't make any major changes to the engine and you got your VE maps fairly close (ie: you hit and autotuned almost every point in the VE map) you shouldn't have to touch the tuning again.

I still haven't had a chance to look/play with Daves EFI system since the last update as we have both been very busy with other projects...I hope to get a quick look at his car later this week if everything goes to plan....

R 🙂

With the VSS does the ecu want an eath signal or a power signal on rotation ? The VDO 412.009 unit works well on the Subie conversion we do, it provides an earth signal.

Doug Sweetman wrote:Its there on the early stuff too - all the 808 series ECU's have it (or at least with Kalmaker they do). Ratty mentioned this on his first EFI thread.

sweet! Not bad for 1987, huh...

Yep, very capable ECU with the kalmaker software behind it. I've heard of it such a diverse range of vehicles - from a stroker Harley Davidson to the V12 JAg previously mentioned !

Doug Sweetman wrote:Yep, very capable ECU with the kalmaker software behind it. I've heard of it such a diverse range of vehicles - from a stroker Harley Davidson to the V12 JAg previously mentioned !

...and it will even work on VW's! :P

I am hoping to spend a couple of hours on it on the public holiday Thursday (tomorrow). My intention is to get the little issues sorted and then let it self tune for a couple of weeks - I now need the time to work on my own Baja as my Shell finally came back from the bead blasters yesterday (its been away for 11 weeks - not bad considering I was told that I would have it back by then end of the first week! :bounce )

I'll let you know how it goes!

Brad, the wiring diagrams are delightfully vague about the voltage output from the VSS, but upon reading the 'testing the VSS' section of the Camira repair manual, it would appear that the stock Camira VSS outputs a positive signal (between 2 and 10 when wheel is rotating), so it looks like the one you have will not work....although with a little signal inverting circuit....?

R 🙂

[ Edited on 25-1-2006 by ratty 63 ]

Update

Hi all!

I spent Aust. Day cruising around Jacobs Well (northern Gold Coast farming/fishing community) sitting in the passenger seat of Daves Baja with my Laptop on my lap (really? :duh).

I was spending most of my time watching the Auto VE update function do its thing - it's so simple and so quick to update the maps! All Dave had to do was try to hold the engine as close as possible to each load point in the VE table (fuel map) for as long as possible and the ECU would do the rest.

The VE Update function has been turned on and working 100% since Tuesday and already most of the VE maps in the 'normal driving' range have been changed. The maps in the upper rev range and WOT areas have not been changed as the VE update only works in the areas where you have the AFR set to 14.7:1 - WOT is set to run richer than that so no auto tune....

I did smooth out the spark advance at lower revs (it was set to achieve 32 deg at any engine load at anything above 1200 RPM - whilst Daves engine was quite happy to run under these conditions, it was almost impossible to smoothly accellerate from an idle (not that Dave ever drives his cars smoothly from an idle...😀 ). I reduced the advance only a couple of degrees at lower revs and the difference was very noticable!

Since the VE update has been turned on, the 'afterburner effect' that caused so much amusement to those of us in the car (and scared the crap out of anyone who was too close to the back of the car!) has stopped happening 🙁 .... oh well, at least we are not wasting as much fuel now!

During our tuning time on Thursday I tried everything I could think of in relation to fuel in an effort to improve the high rev acceleration at WOT, but nothing appeared to work. I then had a close look at the spark map and noticed that the area of the spark map that would be used at WOT was a couple of degrees down on the 'normal run' area of the map (a leftover from the Camira spark map that we used as a base). I increased all the points on the spark map at 3000+RPM to 32 deg and there was a slight improvement. I would like to try to push the spark advance up another couple of degrees at WOT to see what effect it will have on both the power that the engine makes, and the CHT, but for now I have left it at 32 deg so that the auto VE update can do its thing. Dave also told me that when he built the engine he installed the cam advanced...he was wondering if it wasn't too advanced and once the engine reaches 3500 RPM it was 'going off the top' of the cam.... has anyone had any experience with this sort of thing? His last engine, which was similar in size but didn't have the go-fast crank and big cam would create more power than his current set-up at 3500+ RPM (at 32 deg advance).

As we were returning home, we took a bush track over a rather steep hill just behind where we live. The engine makes an incredible amount of torque at idle. We were climbing up hills at an idle and having trouble with traction at the same time (ever seen a Baja spin both back wheels at the same time whilst climbing a hill ...at an idle?!? :thumb )

I will try to get a download of the VE maps next weekend (after our trip to the Toowoomba Swap) as see what changes have been made...I might even see if I can figure out how to create a 3D graph of the changes and post it here.....

R 🙂

Ross, I really love this whole topic, must be something in my personality. Thanks again for sharing this .
Do you think that you will produce some starter kits or
something like that to help those "not so technical" people.?

Hi Glenn,

Thanks for the encouragement (although those I live with will tell you its not needed! 😀)

I had considered pricing some of the parts that I have used for sale to anyone who was planning to do the same conversion - things like the manifolds and wiring looms - but I wanted to experiment with a number of vehicles/engines so that I understood the finer details of EFI. This way I can design and offer a product that will better suit particular applications... does that make sense?

I am still working on my Baja (which will be powered by a Type 1 based 1835), as you know, I have been working on my Brothers 2.1 litre Type IV in his Baja, and Mud Bashing is also working on another 'big' Type IV that we plan to inject. Once we have done this I feel that we will have gained enough knowledge to confidently design and offer a product (or products) that work. It's still a little way off, but in the meantime I would be happy to share what I already know. Just ask the question!

Just as a matter of interest, what would you expect to recieve in a starter kit?

Update
Dave has been driving his Baja for the last couple of weeks with the ECU in auto tune mode (auto VE update is the technical name used by Kalmaker). It was obvious to us within a very short period of time that the ECU was reducing all the settings in the VE table (ie: leaning out all the load points), which in this case is a good thing as the engine was getting about the same fuel economy as it was when it was fitted with twin Webbers (read: disgusting). After just a few hours the VE table looked very different to how it started out. I haven't downloaded the latest VE table for some time now as Dave and I have been very busy over the last few weeks (trips to the Toowoomba swap, preparing for the Dubs by the Pub event, etc), but I expect that when I do it will be almost unrecognisable when compared to the original (Camira) VE table.

As I think I mentioned before, the Delco/Kalmaker system will only self tune the areas of the VE table that have the AFR set to 14.7:1, so the other areas (WOT and closed throttle) are not being changed. I believe that this could be causing a problem with a stumble that Dave has reported. It happens just as you crack the throttle open - i believe that the engine will be recieving dramatically different amounts of fuel as it changes from closed throttle to open throttle. Smoothing out the VE map around this area should fix this issue, but I won't be sure of that until I have some more time with the car in a week or so.

It is also possible that I have stuffed up one or more of the idle settings, which could also be related. The reason that I say this is that the car is now occasionally hunting at idle - something that it has not done in the past. The last time I had the laptop connected to the ECU, I made some big changes to the idle variables in an effort to stop it stalling when slowing the vehicle to a stop. I thought that it didn't have a lot of effect (some variables don't) so I left them where they were... it looks like I may have been wrong 🙁 Oh well, this is how you learn.

On the bright side, Dave was telling me that his Baja was using around half a tank of fuel between his place and the south side of Brisbane and back again - a trip that he does regularly, a distance of about 120-140 km all up. When we went to Toowoomba for the swap a couple of weeks ago (with a couple of weeks of auto tune time on the car) it used about a third of a tank on the trip from home to Toowoomba - a distance of about 200km (220km would probably be a closer estimate) and this included the climb up the Toowoomba range - so as you can see, the ECU, with no help from me, it making some major improvements to the fuel economy. This is all with the lean cruise option turned off, so we should be able to make some even bigger improvements in the future!

The other issue that we have been dealing with is the lack of top end grunt. The engine will spin freely to about 3500RPM then just seams to hit a wall - it will eventually hit the 5500RPM rev limit but it takes forever to get there. A bit of back to basics thinking presented what we believe to be a rather obvious answer: Dave is using big valves in the heads on this engine and the ID of the intake manifold is considerably smaller than the intake valve size (sorry, I can't remember the sizes that Dave mentioned) - we feel that we may have a problem with the engine being unable to draw enough air through the manifold to happily operate at above 3500 RPM. Dave mentioned that the EFI Kombis also tend to become a little breathless at about this point too - perhaps a self limiting thing, done on purpose by VW to help save the engines? Anyone have any thoughts on this? Our plan in to custom make a manifold with much larger diameter pipes (and a re-designed plenum chamber so that the TB is in a more accessable location) and see what difference this makes.

Thats all for now, I'll keep you all informend as/when I have another chance to look at the car.

R 🙂

yes , experience on a number of different set ups,
would be very valuable - especially to come to grips with all the variables / options on the system.
-
The Complete Delco EFI - VW package.
--------------------------------------------------------------
Info from Ross / Glenn / Others......
other contributions much appreciated
1.Use the EFI wiring loom from the JE Camira - it is seperate from the vehicle wiring loom and is easily removed and modified to fit a VW. Sensors would be best sourced from the same vehicle, but all vehicles that use the Delco 808 ECU use the same sensors, so you could use sensors from a VP/VN/VQ Commodore or even a Daewoo Lanos...
Use the ECU from the Camira (that you removed the wiring loom from) as an exchange unit and buy the complete Real Time ECU from Injection Connection.-Ratty63.
2.Kalmaker software.
3.Laptop computer.
4.The Commodores (VP/VN/VQ and later) use an in tank high pressure fuel pump - unless you plan on fitting this pump into your tank (possible, but not the easiest way to do it) you would be better off purchasing an external pump from a VL Commodore. Use a new filter - they are not expensive. - Ratty63.
5.for simplicity would use the standard distributor with the points removed.
6.Vehicle Movement Sensor.(cheap version to be advised)

7. Surge tank.If you don't run a surge tank you are likely to have issues with air entering the fuel lines whilst cornering or accelerating/braking hard, when the main fuel tank is close to empty. I have heard a number of people say that they are not necessary on road cars, but one of those people always tells me that they cannot let their fuel tank go below quater full otherwise the engine surges when he is cornering...er....hmmmm! (according to him, apparently it is a problem with something else, and not air in the fuel lines - same person has also had a number of EFI pumps fail 'for no apparent reason' - EFI pumps will fail if they get too hot from being run dry).
If you run a surge tank, you will need a low pressure electric fuel pump (and filter) to fill the surge tank and expel any air that finds its way into the tank.-Ratty63.
8. Run minimum 1x 8mm fuel line through car - preferably
through tunnel & use standard fuel line as return line.
Or run two new lines 8mm or bigger.
9. fit o2 sensor in exhaust.
10. fit injectors into manifold.
Yes you can use the Camira ones - but have a think about what sort of fuel rail connection you want to use - the Camira injectors are clipped into the fuel rail. Another alternative is to use injectors from a VL commodore (or Kombi) which use fuel hose to connect them to the fuel rail.
I removed the plastic from the back of the Camira injectors (where the clip usually goes) and attached fuel line - this also works.- Ratty63.

11. Don't forget that you still require a return line to be plumbed into the bottom of the fuel tank (regardless of whether you use a surge tank or not) - this will require the tank to be modified by someone with the necessary equipment to evacuate the gasses from the tank - an EFI specialist can usually help with this.- Ratty63.


[ Edited on 21-2-2006 by OvalGlen ]

...to answer a couple of your questions:

From point 1:
Use the EFI wiring loom from the JE Camira - it is seperate from the vehicle wiring loom and is easily removed and modified to fit a VW. Sensors would be best sourced from the same vehicle, but all vehicles that use the Delco 808 ECU use the same sensors, so you could use sensors from a VP/VN/VQ Commodore or even a Daewoo Lanos...
Use the ECU from the Camira (that you removed the wiring loom from) as an exchange unit and buy the complete Real Time ECU from Injection Connection.

From point 4:
The Commodores (VP/VN/VQ and later) use an in tank high pressure fuel pump - unless you plan on fitting this pump into your tank (possible, but not the easiest way to do it) you would be better off purchasing an external pump from a VL Commodore. Use a new filter - they are not expensive

From point 5: As far as I know, there would be very little/no loss of efficiency using a dissy rather than a coil pack system. The dissy just looks messy and has moving parts which will eventually were out. Using a dissy is the easiest way to do the conversion.

From point 6: If you don't run a surge tank you are likely to have issues with air entering the fuel lines whilst cornering or accelerating/braking hard, when the main fuel tank is close to empty. I have heard a number of people say that they are not necessary on road cars, but one of those people always tells me that they cannot let their fuel tank go below quater full otherwise the engine surges when he is cornering...er....hmmmm! (according to him, apparently it is a problem with something else, and not air in the fuel lines - same person has also had a number of EFI pumps fail 'for no apparent reason' - EFI pumps will fail if they get too hot from being run dry).
If you run a surge tank, you will need a low pressure electric fuel pump (and filter) to fill the surge tank and expel any air that finds its way into the tank.

From point 10: Yes you can use the Camira ones - but have a think about what sort of fuel rail connection you want to use - the Camira injectors are clipped into the fuel rail. Another alternative is to use injectors from a VL commodore (or Kombi) which use fuel hose to connect them to the fuel rail.
I removed the plastic from the back of the Camira injectors (where the clip usually goes) and attached fuel line - this also works.

Don't forget that you still require a return line to be plumbed into the bottom of the fuel tank (regardless of whether you use a surge tank or not) - this will require the tank to be modified by someone with the necessary equipment to evacuate the gasses from the tank - an EFI specialist can usually help with this.

You also need to find some way of attaching the TB to the manifold.

I think that you almost have everything listed there...

R 🙂

Hope you dont mind me using all your answers in the post above.
I know there are a million ways to do things, but can the return fuel line be plumbed into the top of the tank if desired.?
I know this debate has been done before but if the in tank pump is used would it be mounted solidly and thus suffer from the fuel starvation as without surge??

OvalGlen wrote:Hope you dont mind me using all your answers in the post above.

thats fine - the more info we can get in one place, the easier it is for those who are interested.

I know there are a million ways to do things, but can the return fuel line be plumbed into the top of the tank if desired.?

Physically - Yes. However, fuel evaproation as the hot returning fuel drops into the bottom of the tank could cause some problems. Everyone I spoke to about fitting return lines to fuel tanks recommended that the fitting be located in the bottom of the tank. There could also be a problem with the noise of the fuel dropping into a partly empty tank if the return is at the top....

I know this debate has been done before but if the in tank pump is used would it be mounted solidly and thus suffer from the fuel starvation as without surge??

From my understanding, in-tank pumps usually sit in a 'swirl pot', which is like an in tank surge tank (uses a series of baffles to keep the pump submurged), therefore eliminating the need for a surge tank. If you were to just fit an in-tank pump, without a swirl pot (or some sort of anti surging device) then it is likely that it would suffer from exactly the same problems as running an external EFI pump without a surge tank.

Update ... well sort of 😀
I spoke to Dave again today (still no 'hands-on' with the car yet), and he tells me that he has discovered a down side to EFI in an off road car....

....When the car was fitted with Carbies, then engine would loose power/stall when the vehicle was asked to go up very steep slopes, therefore limiting the vehicles ability to climb stupidly steep embankments - EFI does not suffer from this problem. During the week he pointed his Baja at a very steep embankment and opened the throttle - as he made his way up the embankment the front wheels lifted away from the face of the embankment and the rear wheels attempted to overtake the front of the vehicle! With his new high ground clearance rear cage now on the vehicle there was nothing to stop the overtaking manouver that was made by the rear wheels except Daves quick reaction (off the throttle and on the clutch). This was all that stopped the vehcile from rolling over backwards...during all this the engine never missed a beat. Dave considers this 'run at any angle' ability to be the only down side to EFI that he has discovered so far!

There was no damage done to the car in this incident, however I'm not so sure that the same can be said for Daves undergarments 😀

R 🙂

Update

Hi all,

Well, its been a few weeks now since I switched the 'Auto tune' function on and I finally organised to catch-up with Dave to see what has been happening. He told me that there were a few issues that needed attention, but generally the car has been running great. The fuel consumption has been improving continually (it was set up to run rich initially, avoiding issues with overheating/pre-ignition). He now thinks (by his calculations) that it would be good enough to get to Sydney (from the Gold Coast) on two tanks of fuel - not too bad considering he used 2 tanks of fuel to get to Nambucca last time! (and that was when it was still carbied).

The issues that we addressed on Sunday were:
- Stalling when returning to idle (still/again)
- Surging at low revs and low throttle settings
- Knocking (sounds like piston slap) when cold - goes away when warm.
- Far more grunt when cold than when warm (same time as point above).
- Still firing flame balls out the exhaust when throttle is snapped closed at revs (he, he, he 😀)
- Noticable extra vibration when engine is under high load at low revs

This is how we have attempted to sort out the problems:

- Stalling when returning to idle:

As we are not yet running a Vehicle Speed sensor, which would almost certainly fix this problem, it was suggested that we should adjust the throttle stop so that the engine idles with the Idle Air Controller almost closed (at about 10 steps - it was initially set to around 42-45 steps at idle). The stalling is being caused by the ECU trying to make the engine return to 850RPM (requested idle speed) whilst decelerating. The ECU slowly winds the IAC in until it reached 0 steps. If you push the clutch in whilst the IAC is at 0 steps the engine doesn't have enough air to keep it running, and therefore stalls before the ECU can react and open the IAC again.
We re-set the throttle position until the IAC was at 10-12 steps (it fluctuates continually) at idle. This appears to have worked as now, when the IAC is at 0 steps the engine will still run (slowly - down to around 650 -725 RPM, but at least it won't stall!)

- Surging at low revs and low throttle settings:

Upon looking at the new 'auto tuned' VE (Fuel) map it was obvious what was causing this. The 'Auto Tune' function only works where the AFR map is set to 14.7:1 (ie: everywhere except at very low throttle and at WOT). The 'Auto tune' had made a big reduction to the amount of fuel at 30Kpa (manifold vac), but as the 20Kpa range was set to 14.5:1 it was not 'auto tuned' - in other words, as the throttle was opened the amount of fuel delivered between these two points in the VE map varied enormously, causing the engine to accelerate - this sudden acceleration surge would then cause the manifold vac to change back again, causing the engine to decelerate, and around we go again.
Rather than mod the VE map by hand, we opted to set the AFR map to 14:7 right the way across all the ranges (both Kpa and RPM) - so that the ECU can 'Auto Tune' the entire VE map and flatten out the 'lumps' at both ends
This also appears to have stopped the 'flame balls' too.

- Engine vibration at low revs/high throttle settings & knocking when cold.

The Delco system allows you to remove ignition advance when the coolant temp is too high to help to reduce pinging. It can also be used to increase the ignition advance when the coolant temp is low (is the glass half full or half empty?). I noticed that when we started the engine cold, the advance gauge was showing between 15.7 and 19.2 deg advance at idle and the knock was quite audable. This was despite the fact that I have the idle advance set to 15.0 deg! A quick look at the advance correction (in relation to coolant temp) showed that at about 35 deg, there was an extra increase of up to 4 deg. As the engine temp increases the extra advance is removed until it reaches operating temp where it remains at 0 deg (until the coolant gets to over 105 deg).
I flattened out this map, so that it doesn't have any effect on the ignition advance until the coolant temp is too hot and instantly the knock went away as did a lot of the vibration.
I also took the oportunity to remove some of the advance from the low rev ranges too as we felt that this would make accelerating gently a little easier. This change in ignition advance when cold will also reduce the 'goes harder when cold' effect too.

So, after all this, I strapped myself into the passenger seat and we went out for a quick run. The car now no longer stalled when returning to idle - regardless of what we did, it still wouldn't stall. Cool! :thumb
The surging a low throttle was still evident initially, however it had dissapeared within only a couple of minuites (the 'auto tune' function took care of that!).
The extra vibration at low revs/high throttle was also gone and the car accelerated smoothly and was noticably quieter right throughout the rev range - it even reved out further on the highway (I bet the bloke in the fully-sick Honda didn't expect to have a Baja overtake him on the freeway entry at Helensvale - especially when he was making so much effort/noise!:bounce)

The ECU will be left in 'auto tune' mode for another couple of weeks and I will again check for issues before I flatten out any bumps in the VE map and switch the 'auto tune' function off (and the auto correct function on). At this point I will consider that the engine is tuned and will only be reconnected to the laptop if an issue shows up that requires resolving or Dave builds his bigger manifold.... we're almost done! 🙂

R 🙂

well done mate!

well written too.

Out of interest what is the software worth?

koolkarmakombi wrote:well done mate!

well written too.

Thanks! I'm glad that you can understand what I have written - I have tried to post as much info here as possible, written in such a way that most people can read it and have some idea what is going on!

koolkarmakombi wrote:Out of interest what is the software worth?

For all the latest prices and info take a look at the Kalmaker website: www.kalmaker.com.au

The forum/message board is also a great help for those interested in EFI.

R 🙂

wow,
a good friend of mine had a JE camira they were a pig of car but i didnt realise how much you could tinker with them when u know what ur doing

Joel

did you end up taking any pics of the finished install.

pete wood wrote:did you end up taking any pics of the finished install.

Yep - I will post them here in the next couple of days, along with an update of the latest info....

R 🙂

Hi all,

Firstly, I haven't forgotten about the pics - I am hoping to get a couple of pics in the next couple of days of Daves engine bay without the spare tyre mounted over the top so that you can see everything easily. I feel that this would help to make everything clearer than a half dozen close-up pics of different components. I'll post them as soon as I have them.

Update
OK, Its amazing what hindsight allows you to see isn't it?

I have always felt that there was something amiss with Daves EFI system - the engine never ran as smooth as my test engine did, but we just put that down to it being a different engine... how wrong were we?!

Dave had called me on Thursday and mentioned that his engine now had a fairly large flatspot just as you accelerated from idle. I had put this down to the fact that he had run the engine without the exhaust connected and that this had probably caused the ECU to make changes to the fuel map to correct the strange reading it was seeing. I was fairly confident that it would fix itself with a bit more run time (with the exhaust on this time!). As both Dave and I were planning on attending Klubfest in Brisbane, I offered to take my laptop and have a closer look whilst we were there (as well as take the opportunity to show a couple of interested people what it could do).

At Klubfest on Sunday I connected the laptop to the ECU and questioned Dave as to exactly what it was doing. He commented that it was running "like crap" - a lot worse than it was on Thursday. It now felt like it had a stumble every couple of seconds, regardless of engine speed and load, and it almost would not idle. This didn't sound good!

First thing I checked was the VE (fuel) map to see what the ECU had done to it... it was almost unrecognisable with sharp peaks and holes everywhere. Some points on the map were almost maxed out (around 98%) whilst others, right beside them, were set to single figures (around 4%) - No wonder it was running like crap...but why?

I decided that this fuel map was way to 'mucked-up' to bother trying to fix, so I recalled and reloaded an old map with a nice smooth VE map. The car had run OK in the past on this map so as far as I was concerned, this was the best and fastest way to return the engine back to a driveable state.

We then started the engine and let it idle - it was still rough and the O2 sensor graph indicated that it was running at around 14.7:1 - which is too lean for an engine to idle (but is what you are aiming for when not idleing). Now the fact that it was too lean on a known-to-be-good VE map was still worrying me, but I couldn't see anything else wrong... this was very strange.

As a temporary fix, I increased the VE at the points in the map that the engine idles at and everything smoothed out (as it should). The idle was now smooth and the engine would now accellerate through the rev range (with no load) just fine. But there was still a slight stumble if the revs were held constant, regardless of the revs....hmmmm.

We decided that this was the best we could do without pulling things apart and double checking everything (and this was not something that we wanted to do in the middle of the show!), so at this point I shut down the laptop and suggested to Dave that he sees how it goes and to let me know.

A couple of hours later I recieved a phone call from Dave - he had just arrived home (he had stayed in Brisbane for a few hours to catch up with friends) and the car was now running so badly it was almost undrivable ... WTF? :bounce I suggested that we connect the laptop again and have a closer look and (as expected) he agreed. A few minuites later Dave was parked on my front lawn and the laptop was connected.

Once again, we found that the VE map was all over the place, however the car was running better now than it had on the way back from Brisbane... I decided that it had to be a loose connection of one of the sensors failing, so we started to check everything. Whilst I checked the electrical items Dave checked the fuel supply ... strangely, everything appeared OK. The only thing we hadn't checked was the injectors - perhaps one (or more) was not firing properly. I grabbed my Camira service manual (remember this is based on a Camira EFI system) and my multimeter and began to test the resistance of each injector. The manual says they should be 16 Ohms, but every one of Daves measured 1.2 Ohms ... Bingo! It would appear that the standard Kombi injectors are low resistance injectors whereas standard GM injectors (like most production cars) are high resistance injectors. So this was definatly part of the problem, but why had it suddenly become worse? Dave decided to go and have a chat to his local EFI bloke and have both his Kombi injectors checked and also have a spare set of Camira injectors cleaned and checked. The plan was to install the Camira injectors in place of the Kombi units and sell the old Kombi ones.

The next day Dave dropped into the EFI bloke and told him the strange tale. The EFI bloke mentioned that he didn't think using low resistance injectors with an ECU made for high resistance injectors would cause a problem (although I have books stating that it will), but when he tested the Kombi injectors they were, in his words "completely knackered". Then when presented with the Camira injectors and told what they were out of, he handed them back immediatly and told Dave "they are too old to be used, just replace them"... er... really? He then quoted Dave a good price on a set of second hand late model Ford Falcon injectors that have the same flow rate as the Camira ones. Dave gave him the go-ahead and a couple of hours later he had a set ready to be installed (cleaned and the tip changed to the same shape as the Kombi injectors).

Dave installed the injectors that arvo, and when he started the car the difference was obvious. It now ran MUCH smoother (than it ever had in the past) and upon driving it, it now has much smoother acceleration and a heap more grunt from low revs. High revs are also smoother, but the amount of grunt is still down slightly (probably due to me mucking around with the ignition advance at high revs - oops!). The engine also sounds different - far smoother than ever before.

Is Dave happy with it now - you bet!

So there you go, I never felt that this engine ran as smooth as my test engine, or responded to changes as well, and it would appear that I was correct. It will be interesting to see what sort of mileage Dave can get out of it once the VE map had sorted itself out over the next couple of weeks. I'll let you know how it progresses (and whether we can get any more grunt out of it at high revs)

R 🙂

purely out of interest what were the new injectors worth?

I have a camira around the corner that I keep looking to see life to offer the usual case of crownies for. I dont have a beetle yet but am collecting the bits...........

love your work ratty

kkk

koolkarmakombi wrote:purely out of interest what were the new injectors worth?

Dave did tell me, but I can't remember :duh I'll ask him again and let you know....

koolkarmakombi wrote:I have a camira around the corner that I keep looking to see life to offer the usual case of crownies for. I dont have a beetle yet but am collecting the bits...........

Well worth the expence - a number of people have told me that it is a waste of money, but when you see what a difference EFI has made to Daves car, both on and off road, in terms of both drivability and fuel economy, there is just no comparison!

koolkarmakombi wrote:love your work ratty

awww, shucks 😊

R 🙂

koolkarmakombi wrote:purely out of interest what were the new injectors worth?

OK, spoke to Dave last night - the injectors were 2nd hand, out of a late model Ford Falcon - cost him $96 for the set (of 4). I believe that this included cleaning and the tip being changed from the Falcon style tip to the Kombi tip.

Dave tells me that the car has used much less fuel this week (having had a week to self-tune with the new injectors). The flat spot, just off idle has almost disappeared and the engine will now rev out further through the rev range before running out of breath than it used to with the old injectors. (Still won't easily reach the rev-limiter though 🙁 )

The only thing that hasn't improved is the hunting at idle, which is entirely my fault as I increased the points on the VE table where the engine idles when the old injectors were still in the car, to try to smooth things out. I haven't changed them back again since the new injectors were installed. I hope to have a quick look at this issue on Saturday.... I'll keep you posted.

R 🙂

Hi all,

Time for another installment in the EFI saga! 😀

After Dave installed the new injectors everything ran well for a week or so, then midway through last week I recieved a phone call from him at 11pm. It went something like:

Dave: "...Dude, its running like crap and is slowly getting worst - is there some way I can re-set the ECU so that I can drive it home. It's almost undrivable in this condition..."

Me: "Whats it doing .... and where the hell are you at this time of night?!?"

Dave: "I'm in Brisbane - its surging and its also backfiring through both the air filter and the exhaust at any rev range except if I have my foot flat to the floor or right off..."

Me: "Sounds like a vac leak in the manifold.... have you checked for this yet?"

Dave: "Nope - why would there be a vac leak....?" sounds of walking towards the back of the running car and things being poked, pulled and pushed, "Oh fugger, the number 1 injector is loose - it's almost fallen right out...how the #%&* did that happen..."

...it turns out that the seal on the injector had compressed down after it had been installed, causing the nut that holds the injector to loosen off and the injector to rattle loose. However, as Dave had continued to drive the car with the injector loose (and a fairly large vac leak) the ECU had attempted to re-tune the engine to suit. This had caused some strange results (to say the least) in the fuel map with some points of the map almost set to 100% whilst the point beside it was set to single figures - hence the surging that Dave had complained about. The only problem now that the injector was beck in and tightened was that the fuel map was so badly stuffed up that the ECU was going to struggle to set it back to normal again. Dave decided to drive the car home from Brisbane using only full and zero throttle (😀) and leave it for me to manually adjust as soon as I had a chance.

On Saturday I wandered over to Daves place and downloaded the VE (fuel) map from the ECU - after we stopped laughing I uploaded another map that we created as a "runs but not perfect" or starting point and I left it to start self tuning (again). We went for a short run to make certain that it was going to be OK - this short run should be a post on its own .... ever driven a really high Baja on wet roads with Mud Terrain tyres on it ... tear-off undies should be standard equipment! And I won't mention what happens when you forget to lock the bonnet down properly....🙂

Dave did two long runs to Coolangatta for the Wintersun Hotrod Festival on Saturday and Sunday and reported on Sunday that it was MUCH better - even better than when he had just installed the new injectors - but it was beginning to hunt at idle.

As Monday was a public holiday, I decided to have a quick look at the fuel map again to see if I could sort out the hunting at idle. I downloaded the map in the ECU and noted that it had made a few minor changes, but it basically looked the same as when I had uploaded it on Saturday. The most notable thing was that the 800 RPM part of the map had been leaned out. I started the engine and was surprised to see a couple of things happen.

1. The ignition advance was 18.7 deg - despite the fact that the main ignition map was set to 12.0 deg. I was certain that I had removed the spark correction when cold, but I checked it again anyway and found it to be set back up to Camira specs again (+6.7 deg at 40 deg C coolant temp). Then I realised... I had turned off the spark correction and not saved this change to my laptop, so when I uploaded our 'starting point' map .... you get the idea :doh

2. The O2 sensor was showing a lean mixture, but as the engine came up to temp, the mixture improved (as did the smoothness of the idle).

3. The ECU appeared to be having trouble staying in idle mode (where the IAC controls the engine speed) - as it switched into idle mode the engine would drop in revs, the mixture would lean out, and the IAC would try to fix the problem by increasing the revs. The revs would rise so fast that it would go out of the rev range that the IAC controls and the idle mode would switch off. Then engine would then begin loosing speed again and ..... go back to the start!

I decided that the engine needed more fuel at idle so I pushed up those points in the 800 RPM section of the fuel map - all of a sudden the engine was idling smoothly - no hunting and with the IAC controlling the idle speed - Simple!

I decided that, in order to stop the ECU from autotuning the fuel at idle that I should change the AFR's in the 800RPM range from 14.7 to 14.6 (the ECU will only tune the parts of the fuel map that are set to 14.7). I also made a couple of adjustments to the warm-up fuel AFR's (to make it richer when cold) then I saved the map (this time I remember doing it!) and suggested to Dave he lets me know how it goes.

When I saw Dave on Thursday, he told me that it was still idling perfectly and the engine is getting smoother and smoother - it has also only used less than a quater of a tank of fuel all week - which he tells me is absolutely amazing!

I have decided that I will give it another week to self tune, then we will take the time to smooth out any obvious problem areas in the fuel map before turning the self tune function off and the auto correct function on. This will mean that if it should ever 'stuff-up' the fuel map whilst Dave is out and about, he can just remove the fuse from the ECU, wait 10 seconds and replace it and the correction map will be erased - reverting back to the 'normal' fuel map.

Dave also wants to have a fiddle with the ignition mapping again now that we have the new injectors in it as he feels that it is still a little lacking below 2000 revs due to a change I made to the ignition map to try to correct the problems caused by the old injectors..... We will see what happens!

R 🙂

PS: I'm still trying to get those pics - it just seems that the harder I try the less time I have to take the pics! I will get them .....

Hi Volks,

Had a chance to have a quick look at Daves Baja on Sunday. Dave agreed to let me 'plug in and have a look' as long as I "didn't bugger anything up" He followed this up with: " It's been running the best it ever has so lets not make any major changes".... I couldn't agree more! 😀

I downloaded the latest info from the ECU and had a close look at the VE map - its still self tuning and appears to have modified/self tuned about half the map so far - I had expected it to have done more by now, but the car has only done short "home to work" runs (where it would be lucky to get to operating temp), and long, constant high speed freeway runs, so I s'pose thats why.... (I've posted a screenshot of the current VE and spark maps at the end of this post).

Dave mentioned that the engine would hesitate when you open the throttle from an idle - only a short hesitation, more like a cough, but it's much worse when the engine is cold. I realised earlier in the week that the AFR when the ECU was in idle mode and the AFR when the ECU was in run mode were very different at the 800RPM band (Idle: 13.0:1 Run: 14.6:1 with the engine at normal operating temp) - this would (to my way of thinking) cause the engine to lean out suddenly when the throttle was opened (which causes the ECU to switch from Idle mode to Run mode) - so I made the AFR in run mode (at 800RPM) 13.0:1 - suddenly the cough dissapeared - but the engine was warm, so the big test was going to happen the next morning (today).

We then went for our usual test thrash to see what else we could find - Dave pointed out that the amount of power that the engine has is improving slowly, but it still feels like it needs a little more spark advance down low... and at full throttle. I was hesitant to change the 'down low' advance until we had seen if the cough I mentioned above had gone, but he convinced me to add a couple of degrees to the entire top end of the spark map (from 1800 RPM upward and from about 50 KPA upward). I did this whilst we were moving, and Dave gave it a thorough testing, but we both feel that there was little change ... Dave wanted to add another couple of degrees, but I rememinded him of the "don't bugger it up" clause that I was working under - so he agreed to leave it at that for now... until next time 🙂

I recieved a message from Dave this morning, telling me that the cough is not gone yet (with the engine cold), but is much better than it had been. I spent a couple of minuites scanning through the hundreds of variables that I can play with and I have found a variable that adjusts AFR dependant on engine temp ... and its currently set to a much leaner AFR all the way through the temp range than what I have had to set the idle AFR too, so this will need changing next time I see the car.

Despite this large post, things are looking brighter than ever before 🙂 The car is now getting better fuel economy than it ever has in the past and is really smooth - smooth in relation to both throttle response and vibration. Of course there are still a few more little issues we have to tackle, but we have learnt our lesson about making massive changes all at once, so sorting out these problems will be done slowly and the results recorded for future reference.

I'll keep you up to date!

...and no, I haven't forgotten about the engine pics - I'm still trying to convince Dave to remove the spare tyre rack off the back so that I can get a couple of good, clear shots... perhaps soon! 😀

R 🙂

The following pics are:

1. The current 3D VE map - still a few bumps and lumps that need to be smoothed out, but looking good!

2. The current 3D spark map - this is the map after changing the max spark advance from 32 to 34 deg. You will notice that there is a bump back down to 32 deg at about 6000 RPM as I haven't bothered with this part of the map - the engine is rev limited to 5500RPM and rarely even gets there.

....I thought that, just for interests sake, I would show you what happens when you have your ECU set to auto-tune and you have an injector come loose... and you have no alternative but to drive the car home before you can fix it.:puke

These pics are screenshots - taken with a digital camera from my LCD monitor ... I know, not the best way to get a good clear pic, but as Kalmaker is DOS based I don't believe that I have any other alternative) - I have tried to keep them as large as possible so that they are still readable

R 🙂

The following pics are:

1. The 3D VE map after the trip home with a loose injector - compare it with the map in the post above :puke What you can't see is the MASSIVE hole at 90KPA between 2800 and 3200 RPM

2. The 2800 RPM range - Note the 90KPA setting (self tuned). KPA reading are across the bottom and the VE percentage is up the side of the graph

3. The 3200 RPM range - Again note the 90 KPA VE setting. No wonder it ran like a pig! 😀

Hi all,

I thought that I should update this thread for those of you who are wondering how this project is going, now that it’s almost 1 year since it began. I also had the opportunity to drive Daves Baja a couple of weekends ago (the weekend of our club show) – this is the first time that I have been able to drive it any distance by myself (I’ve moved it around the carpark at Daves work before, but that was all), so I thought I would share my thoughts about the vehicle in general…

I’ll discuss my thoughts about the set-up of the chassis at the end of this post, as I was more interested in how the engine and EFI system behaved – after all, that was the topic of this post in the first place!

I have to say that I am impressed with the engine and EFI system operation – and not just because I set it up either! It felt very smooth and predictable once I got the hang of the way Dave has the rest of the car set-up – you’ll understand what I mean when you read the end of this post. The Type IV has a truckload of torque, making it possible to drive around town in 3rd gear, taking corners and roundabouts at idle, then just squishing the go pedal and smoothly accelerating away to well over the legal speed limit on the highway! Asking the engine for power from an idle, even in 3rd and 4th gear gave a very smooth response, with no sign of the ‘shudder’ that indicates that you really should have shifted back a cog.

I did a short blast up the freeway and found that there was also plenty of power for overtaking manoeuvres, although it did feel like the snappy throttle response that is present in the lower rev ranges was missing – this is something that my brother has commented on in the past - he feels that the stock Type IV manifold may be restricting the flow of air to the engine at higher revs. Personally, I agree in part – but I also think that we still need to have a closer look at the ignition advance at higher revs. We have peaked the advance at 34 deg, but having recently done some research on stock dissys I feel that we could add around another 10 deg in some areas of the advance map without doing any harm to the engine, but potentially giving it a good, hard kick in the ass where it is currently lacking.

Cold and warm/hot starting is instant and simple – just turn the key and it goes. It would appear that we have finally managed to get the cold enrichment set-up correctly, as the engine behaves the same regardless of the temperature. The requested idle speed is set to 850 RPM (when warm) and that’s exactly where it sits, indicating that we have the VE and AFR correct in this area too – that took a little fiddling (mainly due to a few incorrect assumptions on my part – I should read the instructions more carefully!)

So, generally, I’m very happy with the way that the engine is running now – I feel that there are still a few areas which we could further refine, however as the car is in constant use and Dave is also in the process of building a bigger diameter manifold to replace the stock Type IV unit, I’m not going to spend any more time on tuning until Dave can make the time for us to sit down and go through everything step-by-step. I should point out that the last time I had the laptop connected to the ECU was months ago, so not a lot has been done recently! I would also like to set-up the lean cruise function so that it will become more economical on long steady runs, but this requires the rest of the tuning to be completed first so….

Now, to the chassis – as many of you will know, this is a Baja. It is based on a Semi-auto floor (IRS rear and ball joint front) with a heavily modified king/link front end bolted on to it. There is also a large number of mods to strengthen almost everything and to give better ground clearance. It runs a 1600 Kombi gearbox, 4 stud Beetle front disk brakes, and VP/VN Commodore rear disk brakes. There is also a body lift kit between the floor and the pan to strengthen the body, and give greater guard to tyre clearance.

I can sum-up the way that the chassis behaves in one word – disgusting :puke. It bump steers horribly, the steering wheel twitches constantly when driving along a smooth section of road, and it YELLS at you if the road/track should be bumpy. It also pulls from side to side when you apply the brakes, which side and how hard it pulls depends on how hard you apply the brakes and how far the front suspension compresses.

Then there is the issue of the accelerator pedal – because of the location of the Throttle Body (almost against the firewall, low down on the RHS of the engine bay) the throttle cable has to do a sharp turn once it has entered the engine bay, this is causing the cable to bind. To ensure that the throttle closes as soon as you step of the pedal, Dave has fitted a monster of a return spring to the TB. All this means that the amount of pressure that is required on the accelerator pedal to open the throttle is between excessive and ridiculous! Making fine adjustments to your speed requires a fair bit of concentration, and leaving the lights without looking like you are completely unco is almost impossible unless you are willing to spin the back wheels! (ie: just stomp it and dump the clutch – a great way to brake CV’s when you do this with off-road tyres!). I also found the clutch take-up to be unpredictable too, however this may have been because Dave had been out playing in the mud during the previous day and everything was inches thick, including the clutch cable. This didn’t help me look any more co-ordinated when leaving the lights either!

So, there you go – If this engine was in a car which didn’t have a mind of it’s own I feel that it would be fantastic to drive – smooth, powerful, reliable and predictable – everything that you could want!

I’ll attempt to keep you up-to-date with the manufacturing of the new manifold and my ideas on ignition advance – it will be interesting to see which has the biggest effect.

Hey, how about that – this is my 1000th post! I finally made it!

R 🙂

Thanks for the post mate it is a great read.

I am dead keen to do the same FI in a year or so. I have a 2yo and one due any day now so I am being realistic. I am hoarding bits though..... Plan is to rebuild the T1 1600 standard to get it all working then prob big bore once its working and I can retune using selftune. By then I will have ironed the bugs out of me and the new FI and will be able to retune easily- I may be dreaming...

As a guestimate what is the kalmaker chip/handset etc worth. I know they have a site but a guess is fine.

Im glad you described the chassis. Ppl often forget how much effect the rest of a car has on driveability.

I wonder aloud about the selftune. How early in the tweaking can you set this and let it set itself up? As I understand it this sort of feature separates the higher end cpu's from the kit types microsquirt etc.

Could you post a pic of the engine? I am keen to put a pic with all those fine words.

Great read :thumb

kkk

After months of throwing ideas about, we’ve finally done it! “Done what?” I hear you ask… We have changed the manifold from a re-shaped version of the factory EFI manifold to a much larger, completely custom-made unit.

Ever since installing EFI on Daves Baja we have felt that something was lacking at high revs. The engine just didn’t have the grunt that it used to have when it was fed by Carbies – it still ran OK, but it just felt a bit flat. This was made obvious by the fact that it was not possible to hit the 5500 RPM rev-limiter with the vehicle in gear (something that this engine should be able to do easily – it would get to about 3700 RPM and just stop climbing.

We noticed very early on that the vac reading from the MAP sensor would go out to 100% almost as soon as the throttle was opened with the engine revving at anything over about 2500 RPM, which we felt indicated that the engine was being starved of air (ie: we were reaching the maximum flow rate of the manifold).

After a couple of months of planning and a few weeks of collecting parts and organising to have a spare weekend to spend doing the manifold changeover Dave finally got around to doing the swap over the Xmas/ New Year break.

He had already created a new plenum chamber from a couple of pieces of steel pipe and a couple of flat ‘end caps’. He had also had a local exhaust shop bend up a couple of new, larger diameter manifold runners (from the plenum to the injector mounting/casting on the head), so all he had to do was fit it all together and then fit it to the engine.

The only part of the factory system that was re-used was the casting with the injector mounts in it, but the inside was opened up as much as possible using a die grinder. Dave also spent some time matching the casting to the intake port of the head to ensure that there was the best possible air flow being delivered to the combustion chamber.

The rest of the parts were simply cut to size (the runners were created to be too long on purpose so that they could be ground down to make them fit perfectly), and welded together.

Dave also took this opportunity to change the location and operation of both the TB and the throttle cable, in order to address the problems with the accelerator pedal that I mentioned in my previous post. This means that instead of the TB being against the firewall, low down on the RH side of the engine bay, it is now located at the other end of the plenum chamber (above the cooling fan). This allows the accelerator cable to run in a straight line. He also fitted a secondary ‘swivel’ which, he tells me, will help to improve the feel of the throttle. The MAT sensor was relocated to the bottom of the manifold (previously mounted on the top of the stock plenum chamber, where the factory cold start injector once was).

Once welded, the whole thing was painted in a crackle finish paint (not my choice, but anyway…) and allowed to dry. During the paint drying period, Dave removed the engine to replace the flywheel seal and clutch, which had been giving him grief for a while. He also took the opportunity to remove a set of cheapo brake pads from the rear discs and replace them with better quality pads in the hope that it would stop once again! He then fitted the manifold to the engine and replaced the engine in the car. (Note: it was not actually necessary to remove the engine to change the manifold, but as it was already out he took advantage of the easier access).

Thankfully, the minor changes in location of the TB and MAT sensor did not require modification of the wiring loom, and all the other electrical items were still located in the same position as before.

Dave had planned to use an intake duct and filter housing (and filter) from a VL Commodore in order to ensure that the only restriction in the intake system was the throttle plate. However having fitted the engine back into the car, it became obvious that for the ducting and filter to fit the spare tyre rack (mounted over the engine) would have to be modified to allow more clearance, so Dave decided that (for now) he would fit a Pod style filter directly to the TB and would consider redesigning the intake and filter system at a later stage.

Upon starting the car, the first thing that was noticeable was that the induction noise was reduced (even with only a Pod filter mounted on the TB)! The engine also sounded smoother at idle, but it also idled faster too. After double checking everything for tightness and for fuel and oil leaks Dave headed out onto the roads for a test drive.

At first he felt that the car was running very lean as it simply didn’t have any power at all, however after about half an hour it began to feel better as the Auto VE update began to remap the fuel to suit the new ‘easier breathing’ manifold. Feeling that it was looking good, he parked the car for the night and let me know (via SMS) of his success.

During the next week the car was mainly driven to and from work, and towards the end of the week Dave tells me that it was feeling much better than before – he had even given it a couple of runs at full throttle and he tells me that it easily spins to 4500RPM and is still climbing when he shifts gears, however he doesn’t want to go too much further just yet as it still feels lean at the top end.

He tells me that the small flat spot just off idle is back again, but I suspect that this is due to the fact that the fuel map for idle will be slightly incorrect now that the vacuum has changed in the manifold (and the auto VE update has been told not to make changes to the fuel maps in the idle range). I hope to be able to plug in my laptop and fix this small problem in the next couple of days.

It will be interesting to see just how much power we can get from this engine with the new manifold once everything is tuned correctly. It certainly sounds like it will return some (if not all) of the lost power. I’ll keep you up to date as it progresses….
I have included a couple of photos of the old system (just before the manifold was removed) and of the new manifold in the building stage. I haven’t yet been able to take a pic of the new manifold fitted on the engine in the car, but I will endeavour to do so ASAP.

I have included a couple of pics (over the next couple of posts of the old manifold, then the new manifold....

Pic 1: this is how the engine looked just before it was removed to perform the above operation - this pic shows the position of the air filter, which you can follow down to the TB in the lower RH side....

Pic 2: ... then if we remove the air filter you can see the electircal board that houses all the relays, fuses, MAP sensor and ignition coil...

Pic 3: ... as you can see, getting to the fuses with the air filter in place was not an easy job - thankfully, removing the air filter was very easy.

Pic 1: the Camira distributor, as fitted to the Type IV engine - it's a tight fit, but it works fine!

Pic 2: Here is the Plenum with a spare TB attached and the MAT sensor screwed into the back. All custom made from pieces of pipe by Dave.

Pic 3: With the runners removed from the injector mounting/end casting Dave was able to remove some of the material from the inside of the casting to allow more air flow. Here the RHS has been ported and the left is next in line. In all these shots the new manifold is fitted to Daves dummy engine (just a case, barrels and heads, no other internals)

Pic 1: The Runners and Plenum, before the runners were welded to the injector castings - the plenum chamber is attached to the runners by rubber tubes (same as the factory system only bigger)

Pic 2: The Runners and Plenum after welding and painting (once again, mounted on the Dummy engine). This shot taken looking from the pulley end...

Pic 3: ... and finally, the top of the plenum chamber, now with the throttle cable control (pivot) from the Camira welded in place.

R 🙂

Nice work Ratty.

Well done to you and your brother. Thanks for the pics and the speal.

What size tubing are you using for the runners?

Let us know how you guys get on.

Cheers
Jeremy

Hi Ross

Dave was taliking about you two doing this at our weekend at Levuka.

Looks great.

Marc

Jeza wrote:Nice work Ratty.

Well done to you and your brother. Thanks for the pics and the speal.

What size tubing are you using for the runners?

Let us know how you guys get on.

Cheers
Jeremy

Hi Jeremy,

Thanks for your praise! :blush

The runners are made from 50mm tube and the plenum is 100mm tube.

We took the Baja for a test run on Friday (Aust Day) to try to smooth out a couple of minor issues that have appeared since fitting the new manifold - keep in mind that we simply just swapped manifolds without changing anything to do with the tuning, so we expected a few minor problems...

There are three main problems that we wanted to deal with:
1. A slight hesitation (like a cough) when the throttle is opened from the closed position - almost like a faulty accelerator pump on a carbie.

2. The engine will vary in speed between 1500 and 2500 RPM even if the accelerator is kept still.

3. Reducing power, the higher the revs, the less power the engine appeared to have - This was similar to the problem that we were having before the manifold changeover.

Problem number one was reasonably easy to fix: When the old manifold was fitted, the fluctuations in manifold vacuum would cause the acceleration enrichment to activate most of the time, so I made the trigger vac change greater (so that the vac would have to vary more to trigger the enrichment). Now the new manifold is fitted the vac change is less so I had to 're-sensitise' the trigger by reducing the amount of vac change required to trigger the enrichment. This appeared to fix the problem 99% of the time - but when we set this up I fiddled with almost every variable that related to acceleration enrichment and I feel (and the reading from the O2 sensor backs this up) that it could use more fuel for enrichment - this is something that I will improve next time I 'plug in' to the ECU.

The second problem was simply caused by the steps in the (auto tuned) VE map being too big from one point in the map to the next - this is common with auto tuned VE maps that are not tuned in some sort of systematic way (as this Kalmaker instructions explain). It's an easy fix though, I just adjusted the VE map to make the graph rise smoothly from 0 throttle to WOT and also from closed throttle vacuum to WOT vacuum (it's a 3D map) and the problem disappeared except at very low throttle settings. The ignition advance is set up quite aggressively on this engine and I feel that if we were to 'flatten' the curve out a little that this would help eliminate this surging, but Dave doesn't want to loose his low down torque (It is a Baja after all 🙂 ), so we have left it alone for now.

Then we tackled problem 3 - so far we did all the changes whilst sitting on my front lawn, but for this one we headed out onto the M1 so that we could pull full throttle at high speeds for a while.
The engine pulls really well to around 3200RPM, then the power begins to taper off - don't get this wrong, it's far better than it was with the old manifold as it will easily rev to over 5000 RPM without any problem, but it appears to be badly effected by load (ie: up hills, towing a trailer, a car full of fatsos, etc). This was quite noticeable when we approached a slight rise on the M1 - the car, whilst at WOT, would only accelerate slowly from 90km/h to 110km/h, but once over the hill (with the foot still buried), it just took off.
At first I increased the fuel supplied at WOT at around the 2800 - 3500 RPM ranges by reducing the the AFR map, but it didn't seem to have any effect... so out of curiosity I increased the AFR at these points, and the engine felt slightly better... strange! I also increased and decreased the ignition advance at these same points to see what would happen - interestingly, the engine did feel slightly better with the ignition reduced from 34 deg to 32 deg at WOT - but it was hardly noticeable.
So - it would appear that the engine will give more power at WOT with less fuel and less spark advance.... does this seem strange to anyone else?

We stopped and had lunch at about this point, and we discussed what we had discovered. I considered doing a couple of other adjustments to the settings, but Dave was not keen to change it too much as he is heading to Frazer Island later this week and was reasonably happy with the way that the engine performed - we will look at those changes once he comes back.

Dave also mentioned that the cam that he used in this engine was set-up retarded as he had told the cam grinder that it was going to be used in an engine that required heaps of grunt at low revs - we are beginning to wonder if, perhaps, this cam is so retarded that the engine has simply 'run out' of torque by the time it gets to 3500RPM. Has anyone had any experience with something like this? I would be interested to hear your thoughts as it's not out of the question that Dave might consider re-building this engine again in the near future in order to change the cam....

OK, thats all for now - I'll let you know how his trip to Frazer Island goes.

R 🙂

[ Edited on 30-1-2007 by ratty 63 ]

Hi Rosco,

Do you have any details on the Camira TB in the workshop manual you copied?

I have to find out what the hose connections are for. The switch side of things is sorted.

Cheers

56Astro, the Engine management section of the Gregory's manual doesn't cover the vacuum hose locations (perhaps its covered somewhere else in the manual?), however the ports for the hose fittings are labeled on my TB .... this is what I have figured out so far:

- Starting with the big one (labeled ENG) - crankcase vent;

- the two, just above the throttle flap are PCV and VAC - PCV goes to a PCV valve (which we don't have), and I suspect that the VAC port is supposed to connect to the charcoal cannister;

- the ports around the base of the TB are labeled AC, ENG and MAP - I can't find anything in the manual to 100% confirm this, but I believe that the AC port connects to a Vacuum tank that supplies vacuum to the heater/AC/fresh/recirculate/etc mech behind the dash. The ENG port probably has something to do with the charcoal cannister and the port marked MAP is connected to the MAP sensor.

The only two that I have connected are the MAP port (going to the MAP sensor) and the crankcase vent, all the others are blocked off as I don't have any of the other stuff fitted to my car.

I have just had a quick look through a Gregory's manual for a VN/VP Commodore (as they use a very similar TB to the Camira) and I can't locate a 'hose map' in this manual either - they seem to cover the vacuum hoses in the sections of the manuals that relate to where the hose is connected and have descriptions similar to "vac hose to engine" - not really a lot of help!

I hope that this info is helpful....

R 🙂

Sweet

Thanks

The TB is with the engine at the fabricators so I can look at the labels.

Thanks again.

Hi all,

Just a quick update (as quick as my updates can be anyway!)

Dave spent a week on Frazer Island, giving the engine, with it's new manifold, a good flogging. Whilst he was doing this the ECU was self tuning (against my recommendation - if something were to go wrong with the fuel map whilst it is set to self tune, the only way to fix it is to plug in my laptop, which was at home with me).

However, apart from a crack in the exhaust, which Dave fixed ASAP it all went well. The extra load that driving in sand put on the engine enabled the ECU to reset the fuel map at a number of points that it had not previously had an opportunity to do. Also, the higher revs for longer periods of time enabled it to sort our the fuel requirements at the top of the fuel map - as a consequence, the engine will now easily spin out to the rev limit of 5500RPM in the lower gears, however it still feels like it is struggling (although not as much as it was before the trip) at highway speed - something that Dave still believes is being caused by his cam set-up.

Dave tells me that the rev limit (which is referred to as a soft-limit) ...er... isn't so soft. When he triggered it for the first time (whilst doing a burn-out in first gear 😀) he thought he had thrown a rod through the side of the case - I suspect that the 'violence' in the soft limit could be reduced if I make the gap between the cut-out and cut-back-in RPM smaller (at the moment it is about 500RPM). On the positive side though, Dave tells me that the rev-limit is so violent that you kind of make a point of not hitting it ... is this such a bad thing? I might have a play with this in the future.....

So, unless the engine begins to find more grunt at highway speeds, it looks like another engine re-build is on the cards, with the intention of changing the cam to something that is a little less advanced... which will, of course, mean that the fuel map will have to be re-set again to suit .... this seams to be never ending! ... Oh well, it's all good practice!

R 🙂

Is the rev limiter fuel or spark?

56astro wrote:Is the rev limiter fuel or spark?

My brain is a bit fuzzy today, so I had to charge up the laptop battery and check....

...it's fuel, and whilst the cut-out (high) revs setting is 5500RPM, the cut-back-in (low) setting is 5400RPM, not 5000RPM like I thought... perhaps it needs a bigger gap between the high and low settings to make it less violent.... hmmmmmm 🙅 😵‍💫

R 🙂

Can it be changed so that it takes away advance rather than fuel?

If I remember correctly, the WBX that my parents owned had a rev limiter that shut down the injectors as well. When the rev limiter was hit, it started to "chug" but it wasn't violent like you describe.

It is more normal to cut spark - I believe. The potential problem of cutting fuel is that if you are right on the rev limit for sustained periods you may get partial fuelling and go lean at WOT Ideally cutting both is better. Many rev limiters on early EFI cars were part of the rotor button. At least when you cut spark the engine will never run lean.

Hmmm, yes, I agree - however, I suppose that as long as you have the cut-out (high) RPM and cut-back-in (low) RPM values far enough apart the scenario that you describe shouldn't happen - of course, if the rev-limit is as violent as my brother describes then you probably wouldn't sit on the limit for too long 😀

Astro, the Delco ECU will only cut fuel for the rev limit - I figure that they used this exact ECU in dozens of different production models (and therefore hundreds of thousands of vehicles) without any apparent problem so it should be OK for our humble VW's.....

R 🙂

Complete and instant OE style fuel cut doesn't result in a 'lean' condition so there's no real worries there. But spark cut can lead to exhaust damage - the fuel explodes

Hi all,

Well, since the new manifold was fitted things have been going great - the engine appears to be far more willing to rev out to the rev limit and the fuel consumption is improving over time - the only real issue that has bugged us has been a slight hesitation just as you open the throttle (from idle).

We noticed that this hesitation was becoming more and more noticeable as time went on - I had already checked that the fuel and spark settings were the same in the idle and low rev (800RPM and below) ranges to ensure that the hesitation was not being caused by a big change in fuel or spark when you switch maps, but they were identical.... Then we noticed that the hesitation was beginning to appear when you open the throttle at any revs... it was slowly getting worse 🙁

We had also made a change to the throttle stop setting (on the TB) as we had noticed that the IAC was showing too few steps (between 0 and 5 - it should be showing around 10 - 12) - remember that we have to do this because we are not using a VSS. I thought that this was odd as this adjustment hadn't moved since the last time we set it up, but I can only believe what I read on the laptop screen.... Since we made this adjustment the car had begun to stall when allowed to decelerate quickly and the flame thrower in the exhaust was back again when over-running at high revs - this indicated to me that the adjustment we had made was way too much.... but the IAC steps on the laptop screen were correct.... hmmmm :o

After much discussion, we decided that the IAC must be sticking, so we decided to swap the entire TB for another one that we had sitting on the shelf so see if this fixed the issue - but as Dave was unplugging the vac hoses and wiring from the TB on the car he noticed that the vac hose running from the TB to the MAP sensor had collapsed (sucked flat) - figuring that this would have to be the cause of the problems that we were experiencing, Dave removed the offending plastic hose, and replaced it with a piece of fuel hose.

Bingo, problem solved - the hesitation has gone and the stalling and backfiring is also gone.

So the lesson we have learnt today is: don't use cheap plastic hose as vacuum hose in an engine bay!

R 🙂

lol... had the same kind of problem with a Holden V8 I converted - rats had eaten the MAP sensor vacuum line!

VSS is very useful - critical - for the Delcos, escpecially with auto trans. Can you fit one easily? It might make a lot of differecne - as I found out

VWCOOL wrote:VSS is very useful - critical - for the Delcos, escpecially with auto trans. Can you fit one easily? It might make a lot of differecne - as I found out

I could imagine that it would have a big effect on an auto car - much bigger than a manual. I know that we have had to put our thinking caps on to get around a couple of issues caused by the lack of VSS in Daves car - so far we have managed to achieve a fairly smooth running engine without it - but there are a few other features that I would like to set-up that require a VSS so it will probably be the next big update....

R 🙂

Highly recommended - enables lean cruise, better idle anticipation and quality, quicker ST andf LT fuel trim learns (I believe) etc etc etc etc!

Hi all,

Not really an update this time - as nothing much has changed.

Dave has flattly refused to allow me to take the ECU out of "Auto VE update" mode (auto tune) for reasons that I can't understand - all up now it has been auto-tuning for almost 18 months which is a ridiculously long period of time (it is supposed to be used for a couple of days/weeks until the VE maps are sorted then switched over to "auto correct" mode).

The main differences between "auto tune" and "auto correct" is that "auto tune" effects the VE maps directly and is quite quick to make changes, and unless you have your laptop with you, the changes made by the "auto tune" function cannot be easily changed back.
The "auto correct" mode creates a correction table that is then applied to the VE maps - the changes to the correction table are made over a long period of time (the correction table is supposed to allow for changes in engine condition over the life of the car) - the correction table can also be erased by removing the power from the ECU, so should the "auto correct" function make a meal of the correction table (not that it is likely too), then simply removing the fuse (or the battery) will set everything back to zero again.

Obviously, I am keen to switch over to "auto correct" so that if/when something does go wrong with the tuning process, Dave can fix it himself on the side of the road (or track as the case often is!). I have been waiting for a phone call from Dave when he is on one of his trips (like the one to Frazer Island that is featured in the latest VWMA) telling me that he damaged the exhaust and that this has upset the tuning and now the car isn't driveable... or something similar - which almost happened on his Frazer trip....

It would now appear that another engine rebuild is on the cards as Dave feels that the cam fitted to this engine is incorrectly timed and causing the engine to run out of puff at about 4500 RPM... but there are a number of other projects that have to be finished first....

R 🙂

I am getting close to cranking up my Delco inspired injection and have not even considered a VSS-in fact I have just finished taping up the modified Camira loom and would hate to have to open it up to run some more wires.

nbturbo wrote:I am getting close to cranking up my Delco inspired injection and have not even considered a VSS-in fact I have just finished taping up the modified Camira loom and would hate to have to open it up to run some more wires.

The VSS is not strictly required, but as previously mentioned, it does allow you to take full advantage of the Delco ECU. Just see how you go - we have had Dave's Baja on the road for almost 18 months now, and having no VSS has caused a couple of minor problems, but nothing we couldn't get around (once we understood what it was doing!).

Let me know if you have issues with your system and I will try to help out.

R 🙂

PS: got any pics of your system - I would be interested to see what you have done so far....

Yes,did some research today and don't need the VSS.Bloke setting up my computer can rig one up for me if needed-was thinking of fitting a cruise control to it.Just tidying up the loom at the computer end-pulling out all the surplus wires in the multi connector ECU plugs and I can then fit the loom to the car.I am using a TPS that came with the Throttle Bodies and it will be OK with the Delco.Using a Camira ignition module and have mounted it as close to the Dizzy as possible to keep the wires short.Mounted the fuel pump on the RH frame horn next to the transmission and the MPS underneath as well,just below the starter.Mounted the temp sender in the blanking plate over the original Type4 fuel pump mount.Mounting the ECU under the passenger side back seat as well as 3 relays to run everything.Will drill a 2inch hole in the floorpan and use the original Camira rubber grommet.Loom is just long enough to reach.Should be ready to fire it up this weekend.

[ Edited on 24-5-2007 by nbturbo ]

Fuel pressure reg is from a Toyota and the dizzy is from a 244 Volvo-machined down to fit the VW motor.Bringing the fuel from the pump into the engine bay on the drivers side and out of the reg. on the passenger side-I have drilled the fuel rails in the centres and run a connecting line between them across the back and low on the motor-almost out of sight.I ran a 10mm supply and 8mm return line in the tunnel when I was modifying the pan for the auto trans.

[ Edited on 24-5-2007 by nbturbo ]

Wiring loom is all done now so will fit it and the ECU and wire up the relays tonight.A pic of the air temp sensor I fitted to one T/Body.

A thing of beauty!

Keep us up to date with this project - I can't wait to see it all together!

R 🙂

nbturbo wrote:Yes,did some research today and don't need the VSS.
[ Edited on 24-5-2007 by nbturbo ]

*sigh* Been there, done that... Lost of 'experts' only have experience with aftermarket ECUs that don't depend on the VSS in the same way the Delco does. Save yourself heaps of hassle with significant idle and driveability issues: Fit the VSS


[ Edited on 3-6-2007 by VWCOOL ]

Ratty-got it running OK but having problem with real low vacuum signal at idle for the MAP to work properly-making it idle too rich.Because of the large T/Bodies/short manifolds and mild cam,can only get about 5in/Hg at idle and it's pulsing really bad.Have ordered some small one way valves and made a small collector to try and boost the vacuum.Have actually connected no 1 and 4 cylinders only together,and pinching the hoses to restrict the size, can get about 10in/hg at idle but still pulsing too much for the MAP.Other than the idle problem,I can't believe how the thing goes and the throttle response.Just love the way it fires up as soon as I push the start button without having to piss around pumping the throttle and feathering it to keep it idling like I had to with the Webers.If it works will make something a little neater.Will let you know how I get on with the idle issue.

[ Edited on 26-7-2007 by nbturbo ]

[ Edited on 26-7-2007 by nbturbo ]

Thats great to hear!

Interesting that you have a low vac signal at idle ... I will be interested to see how you get on in regards to this problem as I am looking at helping a friend do (another) EFI conversion on a Type IV, and we were discussing the idea of running with twin TB's.

Are you using the Kalmaker Street Pro 3 software or are you having someone else tune it using the workshop software?

Isn't that instant start fantastic - regardless of engine or air temp, it just starts - no messing about.

It's even nicer to know that once you have it set up you won't have to re-tune anything for a very long time... 🙂

Great stuff!

R 🙂

[ Edited on 27-7-2007 by ratty 63 ]

I'm starting to regret not persevering with a single centre throttle body.I'm not using Kalmaker(YET).The bloke doing this for me is using something he has built himself,and has done heaps of cars,but always with a normal intake system.He want's to be able to beat this one,as it will allow him to get into some hotter set ups.He has just Delco'd a V12 Jag motor thats gone into a Holden 1 tonner ute.At the moment he has put a variable resistor into the MAP circuit,so I can fine tune it at idle to stop fouling the plugs while I try to play with the idle vacuum.

ratty 63 wrote:It would now appear that another engine rebuild is on the cards as Dave feels that the cam fitted to this engine is incorrectly timed and causing the engine to run out of puff at about 4500 RPM... but there are a number of other projects that have to be finished first....

R 🙂

Found this thread that talks about timing and a Type IV motor running above 4500rpm. I thought you maybe interested.

http://www.shoptalkforums.com/viewtopic.php?t=76389&highlight=ignition+timing

Sometime it can be hard to get people out of their old habits.

My Brother made a few changes to the manifold on his Baja a couple of weeks ago - nothing serious, but it was enough to change the vacuum at idle - which in turn caused the engine to hunt slightly (when at idle). Instead of coming to me as asking me to sort it out (by resetting the fuel map in the idle range), he adjusted the throttle stop on the TB (like he used to do when the engine was fitted with a carby and was idling too low). This is something that every manual I have every seen (that covers the Delco EFI system) says you should not do - and I have explained this to Dave in the past.

Although it appeared to fix the problem at the time, the next day the hunting at idle was back ... and it was worse than before.

Several adjustments of the throttle stop later (making the situation worse every time) and I get the phone call that I should have received at the very beginning.

When I downloaded the contents of the ECU onto my laptop I noticed that the Autotune feature (which is still switched on) had created a slightly 'sawtoothed' pattern throughout the VE (fuel) map - so I decide to spend some time smoothing out the map. Once this was done, and with time rapidly running out (I had to go to work) I uploaded the new 'smooth' map back to the ECU... or so I thought.

The following day Dave rang me again to tell me that the engine was still hunting - exactly like it was before I had adjusted it - confused, I cranked over my (T)rusty laptop and had a close look at the map that I thought I had uploaded - it looked OK to me....

Dave brings the car back to me a couple of days later and I, once again download the contents of the ECU to the laptop and compared it with the download from last time (before 'smoothing') - it was identical :duh

I gather what I did was, rather than uploading the 'smooth' map from the laptop to the ECU, I downloaded the contents of the ECU to the laptop again by accident 😊 Oh well, no one is perfect 🙅

Whilst I was connected to the ECU this time I carefully, and quietly (without telling Dave), switched the Autotune feature off and the Auto Correct feature on (as it would be when installed in a 'factory' vehicle) - this means that the ECU will no longer change the VE (fuel) map, instead it will create a separate map with any changes it deems necessary, which are then applied to the VE map each time the fuel calculation takes place. The biggest advantage to this is that the separate (auto correct) map is volatile, so should something stuff up in the future (air leak, faulty injector, faulty sensor, etc) and cause the fuel calculations to be way off, it can all be reset to the 'smooth' map by removing power from the ECU for about 30 seconds.

So for most of us this would be the end of this topic - job done. Of course, this isn't the case as Dave is getting closer to doing a rebuild on this engine, with plans for a completely different cam, changes to the heads, and possibly even a little forced induction (although that last one is more of a 'wouldn't it be cool if we....'), so no doubt, this won't be my last post in this subject! 😀

So, how is everyone else going with their EFI systems?

R 🙂

Slowly, unfortunately!
:jesus

Ratty-mines running like a dream.Didn't have any joy with the Delco-tried to get a base file for an AJUN chip-but no go.So bought a Haltec and played with that for a while with little success-so took the car to a Haltec agent.He's got it running really sweet.Will get it out to run the motor in and get used to driving it.It's pulls really hard thru each gear with only light throttle opening and the gear changes are really quick and tight.

Hi folks,

Just a quick one today....

Last night my brother mentioned to me that his car was lacking even more (than usual) grunt at high revs, and he had noticed that on a long run on the weekend (around 800km) that the car felt like it would run out of fuel if WOT was used for any more than approx. 20 seconds. He wanted to hook up the laptop and see what the oxygen sensor was reporting, but I suggested that he double check everything to do with the fuel supply system before I hooked up the laptop and wasted another couple of hours sitting in the passenger seat of his Baja (only to diagnose a mechanical problem rather than a tuning problem...again).

He has just called me to tell me that during his lunch break he removed the fuel filters (2 X low pressure filters that feed the surge tank and 1 X EFI filter, mounted after the HP pump). He said that the low pressure filters were clearly 'flow restricted' (ie: blocked), and the HP filter weighed about 3kg once the fuel was drained from it.

At this stage he hasn't given the car a run (ran-out of lunch break 😀), so we are not certain that the new filters have fixed the problem... but it has to help!😂

I'd love to know what sort of mud laden fuel he uses, as most of my cars will go for years without a fuel filter change.

R 🙂

Hi all,

Dave tells me today that he is having trouble with the engine hunting at idle ..... but only when the headlights are on! 🤪

I feel that it is obviously going to be a voltage related issue (which, coming to think of it has shown it head very early on in this conversion... does anyone else remember?). I have just checked the requested idle (vs battery voltage) variable and I have noticed that there is about a 100 RPM step between the requested idle less than 9.9 volts and the requested idle at greater than 12.7 volts... so I don't believe that this is what is causing the problem.

I suppose that I should put my multimeter on his battery and find out what voltage is coming out of his Porsche alternator - he tells me that the volt gauge does drop quite dramatically when the car is at idle (as opposed to running at revs), but not as far down as 10 volts, so I wonder if his Alt is looking for attention (again - this will be the 4th or 5th time we have had to look at/repair this alt if its not right this time 😬 )

Anyone else have any suggestions? Has anyone else had any experience with Porsche Alternators? Are they reliable (or has Dave ended up with a lemon)?

R 🙂

They dont like dust. My Boss used to have to have the Porsche Alt repaired every second race or so when he was running the Porsche motor.

11CAB wrote:They dont like dust. My Boss used to have to have the Porsche Alt repaired every second race or so when he was running the Porsche motor.

Graham, what was the most common fault? Was it just bearings or was it something else (brushes, windings, diodes)?

R 🙂

I think it might have been the diodes.....just wouldnt charge

11CAB wrote:I think it might have been the diodes.....just wouldnt charge

Yep - been there and replaced (some of) those. This is what I suspect is causing the problems that Dave is having now. I'll just have to find the time to have another look at it.

R 🙂