hey Wes I was just checking out your article and I think you have over engineered the system. Do you think its possiable that your pum was sucking harder than your gravity feed line could flow? I really don't see the poin of a lift pump in this application. The way I would make one would be to have three ports in the can. The first, just above 3/4 way would act as a return to the tank to allow for vapour and fuel return. The second at 3/4 would be a -10 or so fitting from the tank itself. The last line would be at the base of the surge tank to the rail. The return would be pumped to the actual fuel tank. I can see no problesm with my thinking and ask as this is what I have planned and am obviously keen to avoid problems so feedback is appreciated 🙂 hey are you doing to take me for a spin on action day? what if I ask extra nice :bounce
Herbie ·
Did this fix all the problems Wes?
I have done a few efi conversions for turbo rotarys into old Mazdas and something to keep in mind is air likes to stay at the top so you need to be sure there is a good vent from the surge tank to the main tank, if the vent has loops or kinks in it when the tank goes in the you may be fighting an air pocket that keeps topping up each time you uncover the drain in the main tank, IE hard corners, braking or accellarating.
this is a really simple idea my friend ran with great success and I adopted it as my setup.
A Decent size hose with no loops and a vertical sort of angle so avoid trapping air from the Main tank to the surge tank of 1-2 litres or get a fuel trap welded to the bottom of the Main tank so it catches a litre of fuel and wont allow it to slop out.
then down thru the pump to the injectors slowing at the regulator and then returning to the top of the Main tank so it cools as it sprays into the Main tank.
It doesent need to be any more complicated than that, new cars run a similar setup with no extra tanks etc.
The more simple the setup the less hassle it will be.
Of course this wont help you much now you have finnished your setup, but its a good setup to duplicate,I know it feeds 400 hp with no dramas so it should suit anything we are doing.
Cheers Tony
Brad ·
I have to ask the obvious question here, well obvious to me anyway.
WTF are you doing with crap hanging off it everywhere ? It seems you have taken a simple concept and turned it upside down.
Is there any reason you can't simply plumb the return line back into the main fuel line ?
The HP pump has a one way valve and is use to maintaining line pressure, the returned fuel will be slightly hotter than thetandard stuff but the decrease in volume should mean no real increase in pressure.
Seems like a simple idea and was what I planned to do on mine.
Or do you need to allow your fuel system a chnace to breath ? I can't see why you would as I can't really see it getting that hot as to casue problems.
So have I lost the ploy ??? 🙁
Baja Wes ·
ok.
The first design did not have any air bubble problems. The line from the tank to the surge tank was basically straight down to the top of the surge tank. It was a short 10mm fuel line. It was not a restriction, as the car ran absolutely perfect for up to 1 hr. After that the heat from the return line heated up the surge tank. So the high pressure pump was not sucking harder than the gravity feed would flow.
Herbie, yes it has fixed all the problems. The only problem I had was the fuel in the surge tank getting too hot. Your design seems ok, but you would need multiple large lines going from the main tank to the surge tank. I did a lot of reading about this. Basically Volvo ran a similar set-up to you, thinking it would work. They ran into 2 problems.
One being that the hot return fuel going straight to the main tank caused the fuel to vapourise in the main tank, and a lot of fuel vapour was lost out of the tank breather. Modern cars avoid this by having a surge pot in the main tank. The return fuel goes back into this surge pot where it is basically trapped and cannot vaporise away as easily.
The second problem was the High Pressure pumps seemed to die relatively quickly (talking 1-2 years or so of daily use). It seems if they have to suck at all then it strains them. The high pressure pumps have ridiculously high flows, and gravity feed will not keep up unless you have really really really big lines going from the surge tank to the main tank.
They then changed their set-up to something similar to what I am running, with a low pressure feed pump. Their problem went away.
My set-up isn't really that complicated. The low pressure pump is really a recirculation pump, and doesn't need to be that big.
Brad ·
Why can't you just plumb your fuel return into your out line and let the fuel keep going round and round till it is used ?
Baja Wes ·
That's essentially what I had, but that causes the fuel in the surge tank to slowly heat up and cause the problems I had. Weren't you paying attention?
If you have your return line going into your main tank, your low pressure feed pump or your gravity feed line (depending on setup) needs to flow more than what the HP pump flows. This will usually be around 4 liters per minute, depending on the pump. That's a lot of fuel. So that's a pretty big feed line if you want gravity fed.
The way I have and had it set-up, my low pressure pump / gravity feed only had to keep up with what the engine was actually using, which is a much smaller flow of fuel.
Andy42 ·
I was wondering. In the set up you have would it be better to plumb the outlet of the Lp pump into the bottom of the surg tank so it was more likely to supply the cooller fuel from the main tank to the inlet of the hp pump???
Baja Wes ·
Possibly, but I already has a spare fitting in the top of the surge tank 🙂
I don't think it would make much of a difference in the end.
Herbie ·
I hadn't thought about the vapour loss, that makes lot of sense, and I guess it would encourage a smelly system.
does the fuel after the regulator flow at 4 litres/min, I suppose that sounds reasonable,
Buggy Brad,
Hot fuel is similar to hot air it has poorer efficiency.
and you can eventually boil the fuel as fuel boils at a lower temp than water, the you get vapours making your system fail as you replace liquid with gas and it doesnt do the job.
I actually have a little radiator that was an oil cooler from a motorbike and it has the right size piping for decent fuel line,
I am aching to be able to plumb it into a system someday, but have gone prehistoric in my car collection.
Cheers Tony,
PS Wes, As long as you got your problem sorted then its all good.
Andy ·
Mmmmm….. I’m far from an EFI expert, but what you have is basically what Herbie suggested, and will heat up your tank as you described anyway. The only difference is you replace a large gravity feed hose with a small hose pumped under pressure, which only adds to heat.
As the LP pump most likely puts out more fuel than the motor uses, it will circulate the hot fuel into the main tank to keep the temp down as it seems to be doing.
As I see it, Herbie’s suggestion is simpler, but will require modifying you main tank and header for a larger connection (or rely on a number of smaller hoses). Too overcome the fuel vapor problem, run a breather line back to your inlet some how?? Maybe through a small filter or condenser to trap some fuel before it escapes
In the stock Kombi FI system, the return from the regulator went into the bottom of the tank, not the top. This might help reduce vapor from the tank also.
Better still, fix the problem at the source. I assume the heat generator is the pressure regulator. Run a finned copper line at it’s outlet before it enters the tunnel at the back. If it gets a bit of air flow, it should help the problem.
Oh, and on your alternative suggestion, wouldn’t pressurizing the header tank affect the output pressure of the HP pump?? Will this affect the regulator? It will also again add to heat.
My 2 cents worth.
Do I get a spin at action day also!! Maybe you should sell rides to people.
Andy
:thumb
Andy ·
Tony, too cold fuel can also be a problem. I tried to fix a problem of vapor lock on a small motor bike by cooling the carby (admittedly it will be different for FI). Fixed the problem, but the fuel economy became appalling and the plugs began to fowl. I decided to live with the occasional vapor lock rather than try to re-jet etc.
Brad ·
That's essentially what I had, but that causes the fuel in the surge tank to slowly heat up and cause the problems I had. Weren't you paying attention?
I am suggesting you plumb the fuel return in down stream of your pump, ie not into the surge tank but into the line. This will mix the returned / warm fuel with the fresh / cold fuel and send it back into the rail.
I don't see how it can cause vapour lock if it is kept in the line under constant pressure.
I have actually done it thisway and it worked fine. Daniel also did this on the EJ20 in his kombi with no issues.
[Edited on 2-9-2003 by Buggy Brad]
Baja Wes ·
Andy,
The source of heat is the fuel rail itself. The heat from the pumps and the regulator is nothing really. The fuel rail gets quite hot to touch after the car has been running for a while. Fuel vapourises at a relatively low temperature. When you back off the throttle, the injectors don't open at all, so all the fuel goes through the hot rail and back to the surge tank. This is when the fuel really gets hot. The more economic the car, the worse the prob (cos more fuel gets returned).
If the hotter fuel goes straight back to the main tank, it vapourises a lot more easily. This flows through the vent line and overloads the carbon cannister. Excess fuel coming out of the carbon cannister overflow has been known to cause engine fires.
That is why car manufacturers return the hotter fuel back into the surge tank area (although it is internal in a regular car), and not return into the main tank area. The hotter fuel will mix with other fuel and cool down before it reaches the tank vent line. This is what car manufacturers do, so I assume they know what they are doing.
The low pressure pump would raise the pressure in the fuel rail slightly, but not much. A low pressure pump will churn out 3-7psi. A HP pump puts out about 90psi, and the regulated reduced that to 45psi. And extra 3-7psi will make the engine run a little richer. If you installed a regulator on the LP pump, or replaced the EFI regulator with an adjustable unit then you wouldn't have the problem anymore. But I wouldn't and did use this option anyway. Some race cars use this option because the pressurised surge tank raises the vapour pressure of the fuel and prevents any probs with the HP pump cavitating under racing conditions.
Baja Wes ·
Brad,
Are you talking about the fuel return going into the HP side of the HP pump? This will not work, and then the supply and return from the fuel rail will be at the same high pressure?
If you are talking about returning it into the discharge of the low pressure pump, then that is really no different to what I am running now.
I don't understand what your saying?
Herbie ·
Where do the air bubbles come from when you boil water??
they are a physical change resulting from increased energy input.,
and the heat comes from the close proximity of the fuel rail to the Hot engine,
it needs to be close to minimse the time needed for the fuel to quickly jump through the injector/s the less mass of fuel trapped near the injector then the less fuel heating up and it can move quickly to where its needed, otherwise the pulse of the injector has expired quicker than the fuel could move.
If you loop the line back into itself then its not much better than just blocking off the line and your fuel pressure would go bananas, as there is no escape for the fuel once it passes through the pump, the regulator cant do anything as it has equal pressure on both sides.
A regulator should run around 40 psi and then match the rising rate of your turbo boost, if you are turbo, and depending on the source of the regulator and the fuel pump will run maybe 120 psi if it has no escape, the pump will not appreciate the locked system and you must be lucky not to have troubles.
As far as I understand, a regulator is governed by a spring and a diaphram so any extra presure from a low pressure pump would be negated by the regulator reaching max presure earlier and thus just letting go quicker,
And the pressure from any low pressure fuel pump would escape through the surge tank breather anyway, as once the surge tank fills it will overflow back to the main tank
The best Idea is to copy a production cars specs as a car company would have paid a lot of money to be sure their cars did not have trouble.
Cheers Tony
[Edited on 2-9-2003 by Herbie]
[Edited on 2-9-2003 by Herbie]
Brad ·
Won't the fuel go into the engine thus reducing the pressure in the rail ?
Herbie ·
only at the rate that the motor can use it, but its being supplied at about 60 or more psi than needed for safety to avoid a leanout.
Andy ·
Wes,
Run the return line back under the car and let it cool as you go through the puddles 😀😀
Seriously, cooling it on the return run not an option?
:thumb
Baja Wes ·
Andy,
I run my return line in the tunnel where it is safe, but won't get cool. You could run it through a small radiator, but why bother when a LP pump is cheap.
Brad,
EFI pumps like to be flowing near their full flow the whole time. They are completely different to a low pressure pump. If you have them pumping to no-where, they will crap themselves. An EFI regulator is also different to a low pressure regulator, and need to bleed back to something with no pressure, otherwise they won't work. Dan must've had his kombi set-up differently to what your thinking.
An EFI pump will run near it's max flow nearly all the time. That's why if it is entirely fed by the LP pump, the LP pump will have a hard time keeping up. Having the return into the surge tank reduced the duty of the LP pump to only the flow the engine needs.
Even naturally aspirated EFI engines have regulators which vary according to manifold pressure. A little line runs from my manifold to the pressure regulator. Although it is much more important on a forced induction motor.
dan ·
I'm almost frightened to enter this discussion hehe...
Simply because my set-up is about as complicated as a thickshake!
I don't use a surge tank firstly. I use one hp pump and pipe the excess back to the tank....I haven't had any problems what so ever on the freeway and as for cutting out going round corners...well from experience, the tank has to be extremely low, I'm talking, running on vapours... and the lack of power is momentary before it's back to 100%. I have no intention of adding a surge tank thats for sure...tho I understand why your using one...4wd-ing and the likes!
So what are your highway temps nowadays? we'll have to tee up a few races...on private property or course hehe
could be fun?
Andy ·
Dan,
Kombi’s had FI from ~72/73. Even if you don’t have a specific FI tank (they had a return port next to the outlet), they were designed for it. The fuel actually collects in a recess with like a wall around it inside the tank (they do vary through the years though). Similar to what Wes says modern EFI cars have.
:thumb
Baja Wes ·
Hi Dan, good to see you still visit these parts. So your HP pump is not returning to itself like Brad thought.
My friend runs no surge tank in an old falcon, and it cuts around corners and up hills sometimes. But you've got a kombi so it probably doesn't see the same g forces 🙂
Mine is running fine temp wise. Gets up to the half way mark then stays there. I think I need a bigger expansion tank to suit my bigger water capacity though. When it warms up it sometimes overflows out the overflow bottle, then when it's cold it emptys the overflow bottle. Then when it warms up from there it fills the bottle again. Not enough capacity for the volume of expansion I am getting.
As for racing, you realise you don't stand a chance don't you 😀
kombi_kid ·
im about to put a V6 into my kombi.
can use explain to me in english (not this double dutch stuff use are talking) the basics of HP and LP pumps and the size that the surge tank has to be etc?
cheers
rhys
p.s. do all u guys just run standard tanks with more outputs/inputs put into it?
dan ·
alright wes...first too byron with 4 mals & 3 shorties & 7 mates wins 🙂
thanks andy... it suddenly all makes sense now hehe...
Then read it, read it, and read it again. It is probably the best car book I have ever bought. All the VW ones are full of opinions and old VW myths. This book is full of facts and extremely useful information. I recommend it to everyone, even if you don't have an EFI car yet.
Basically EFI needs a very high pressure pump to operate. It also needs a surge tank / swirl pot to ensure the pump is continually fed with fuel. Air bubbles in the high pressure fuel line is a bad thing. Also, a carb can have it's fuel flow interupted and not car because you keep running off the fuel in the float bowl. EFI does not have a float bowl, as soon as the HP pump sucks air, the injectors lose fuel pressure and the engine cuts out.
Andy ·
Dan,
Make that 8 mates (leagally) and I'm in on the race :-)
Herbie ·
Yeah, Wes I agree,
thats a great book, my brother had a copy at one stage.
I wouldn't mind a copy of my own.
Cheers Tony
kombi_kid ·
do u guys know of any fuel tanks with injectyion that had a similar patter to a Kombi???? or do u guys reacon getting a commodore fuel tank chopping it open and get out the surge tank???
cheers
rhys
amazer ·
get yourself a tank from an efi kombi.
this is my old surge setup. I just welded some ends onto a short length of stainless steel pipe for the pot. Both pumps were EFI kombi. My theory being that the one pumping no pressure would outflow the one pumping hp. pump 1 sucked from a 10mm fitting brazed into the bottom of the tank. All the EFi pumps I have seen have 10mm inlets. when the surge pot filled it just overflowed back to the main tank.
It worked. But the injectors were maxed out and kombi pumps didnt flow enough in the end, so I bought a VL turbo pump and bigger injectors. Could only afford one pump so I removed the pot and decided I would just run with 1/2 a tank of fuel minimum. And that worked too until valla where the bloody thing surged with a near full tank.
If I ever finish uni I'm going to get a 100 litre alloy boat tank and mount that under the back seat of my kombi. Then I can use all that space currently wasted by a fuel tank for turbos and intercoolers. Perhaps the alloy tank would be a better thing for any kombi EFI setup.
tassupervee ·
Sorry in advance guys for the long post but there are a few issues here to cover.
Wes
You are running a good system by feeding the surge tank via an aux. LP pump. Running the return from the HP pump back to the surge tank is almost unnecessary tho.
Nothing wrong with this system however. Blends warmed fuel from the engine with cooler fuel from the main tank and ensures the EFI pump is always provided with a head of fuel.
Surge tanks are necessary when there is no swirl pot in the main tank or mainly in the case of circuit racing cars that, due to high cornering "G" forces, will see the fuel surging up the sides of the tank and well and truly uncovering the fuel outlet as the tank level drops around sweeping bends. It would be adviseable in an offroad machine as well as these things wil often experience high angles for extended time and may not necessarily full of fuel at the time.
This type of return from the engine and the LP feed point of the surge tank should be well above the HP pump inlet or outlet depending on the setup, to prevent any heated fuel directly returning back into the HP pump and any air that may be introduced into the system from surging in the main tank from making its way into the HP pump. This is bad
It is acceptable to feed the surge tank/EFI pump by gravity and a 12mm line is adequate provided the run is short and directly down. Bigger is best here.
EFI pumps, by their nature, take a very dim view to "sucking" and this can easily result in cavitation, noisy running of the pump, which leads to overheating and very premature failure. They should always be sited below the main tank to ensure a "head" of fuel at all times.
Ignoring the surging around corners issue of some aftermarket EFI setups, with no surge tank or swirl pot in the main tank and the resulting cavitation, will lead to premature wear of the pump and possible failure of the pump.
This is the reason for having a "swirl pot" in the main tank, in the case of pretty much every EFI car ever made or a remote surge tank, either gravity fed, if it is possible to get the surge tank directly under the main tank, with a vent line from the very top of the surge tank breathing back into the top of the main tank or, (often in most cases), if not, then a aux. LP pump feeding the surge tank (which can then be located anywhere)
The return of the surge tank can be teed into the return line from the engine and this is usual practise. This return line "should" discharge right down in the bottom of the main tank. The return from the engine should never be plumbed directly back into the inlet side of the EFI pump as the fuel will slowly increase in temp leading to the problems with fuel pressure that Wes has described.
There are plenty of dodgey setups with the return fuel splashing into the main tank from the top of the tank that still work just fine.
The issue if vapour and heat expansion/breathing of the main tank is usually dealt with via a regulated/ restricted vacuum source to the tank.
This is why you are often greeted with a loud hissing sigh when you open the tanks at the fuel stop. It is air breathing IN.
Anothe simple strategy to deal with tank breathing is to plumb the main tank vent line into the inlet somwhere between the air filter and the throttle body.
The issue of say, a Facet type of solid state LP pump (which I think Wes is running there) not having the flow capacity of the HP EFI pump is not correct.
Those small solid state pumps are flow rated wayyy above any EFI pump flow demands.
I use a small 40 dollar solid state pump designed for atmo 6 cylinder carby cars as an aux lift pump in my Nissan FJ20det powered Econovan. This engine is turbocharged and dyno's at around 150 KW on stock boost setting.
I dont even use a swirl pot or surge tank and the little lift pump simply blows directly up the bum of the EFI pump as there is no bottom feed in the tank for the pump. This would mean the EFI pump must "pull" the fuel out of the tank which is no good and results in EFI pump damage and eventual failure.
You will find that the EFI pumps are rated with no restriction and as the EFI fuel pressure pressure required rises, say typically around 45 psi or so and well above such as in a rising rate EFI system, the output flow rate reduces proportionately.
The issue of feeding 7 psi up the bum of the EFI pump via an aux. pump in series, and resulting in upping the eventual pressure fuel output pressure of the EFI pump is only correct in the case of an unregulated non-returning or static system. pretty much like a standard carby setup
In all EFI setups, all the fuel pressure regulation is done after the fuel injector rail, on the return line to the tank. So the pressure differences will just normalise between and after the two pumps as they share the load,
The regulator will simply bleed off excess pressure by providing a relatively unrestricted return path of fuel that exceed the flow rates the pumps can supply so the flow rate increases and the pressure simply drops untill the regulator reaches its preset output pressure. This type of regulation results in a constant flow of fuel thru the pump that lubricates and importantly, cools the pump. Also, EFI pumps are unregulated, positive displacement pumps and if the output of the pump is blocked off, as in a non-returning or static system, the pump will pretty much stall and burn out in short time.
It is often practice to use 2 EFI HP pumps in series, one near the tank and another somewhere along the line or in the engine bay being a "helper pump" that will act as a flow/pressure backup especially in say, a rising rate turbo EFI system where the flow demands with a rising rate regulator will exceed the output of a single pump resulting in fuel pressure drop.
VL turbo Crappadores are an example of a dual inline EFI pump setup.
Wes's system is pretty much the last word in a bulletproof fuel system but worrying about the heat buildup in the main tank and the resultant extra plumbing is pretty much unnecessary.
The heat buildup in the main tank is not that great bearing in mind the radiating surface area of a typical fuel tank so the idea of adding a fuel cooler to the setup is totally unnessary.
The third setup detailed by Wes is more than adequate to handle anything you can throw at it and my preferrence would be, in the interests of simplicity, would be to Tee the surge tank return into the EFI return.
Wes has fabricated a square tank but a far simpler to make solution is to use a length of 3" or so exhaust pipe and simply weld on a square plate on the bottom and if, as in Wes's case, using an in-tank submerged type of pump, you will need to make the top plate of the tank removable so the pump can be attached to the plate and wires and outlet pipe and such can run thru it.
There is alway the issue with this arrangement of successfullky sealing the wiring coming thru the lid but this can be helped out by lowering the breather return line from the surge tank well below the very top of the surge tank. You then only have to deal with splashing of the fuel rather than a head of fuel trying to weep out past the wires/seals.
This is probably an unnecessarily complicated way to do things and sealing up the surge tank and using a simple external, inline, hosetail type of EFI pump plumbed directly off the bottom of the surge tank is by far the easiest way to go.
I suspect that Wes happend to have the in-tank submerged type EFI pump floating around on hand and has the patience and facilities at hand to sit down and bother to fabricate and successfully seal a removeable lid to his tank.
By simply brazing 2 square flat 2mm plates to the top and bottom of the 3" tube is shit easy and then it is a doddle to braze on the various inlet and outlet hosetails needed. These square plates can be extended and bent at right angles to make simple mounting brackets if need be.
Sealed surge tanks are available commercially in various capacities starting around 120 bucks in mild steel and a fair bit more for ally iffen you wish to polish it up for wank value.
They take about an hour to make and cost a few dollars for some steel hosetails, the zorst tube and end plates
Too easy.
Have fun dudes
L8tr
E
Baja Wes ·
All good comments. There's obviously more than one way to skin a cat.
The only point I disagree with is this one...
tassupervee wrote:The issue of say, a Facet type of solid state LP pump (which I think Wes is running there) not having the flow capacity of the HP EFI pump is not correct.
Those small solid state pumps are flow rated wayyy above any EFI pump flow demands.
My facet fuel pump is rated at 2.2 L/min flow. Check their website it agrees with that figure (132L/hr).
My HP pump is rated at 3-4 L/min. So getting towards twice the flow rating.
It is simply because the LP pumps are designed to feed as much fuel as is needed for the engine to burn, whereas HP pumps are designed to pump much more than that to continuously flush the fuel rail with cooler fuel.
And to explain my surge tank a little more :thumb ;
I used an intank pump from the same car as my V6 came from. That way I knew it was the right size. This set-up is compact and ensures mud cannot get into the HP pump. The intank HP pump also has a nice filter bag on the bottom. The arrangement is why my surge tank is square.
As for having to seal wiring through the lid, that is not an issue. The intank pump has a mounting flange which the fuel pipes and wires are already sealed through. All I needed was a surge tank with the same mounting flange as the stock V6 tank. I got a friend to make the surge tank to suit, and it cost me a carton or $40 or something 🙂
PurpleT3 ·
Sorry to intrude on your discussion re aftermarket EFI.
I'm in the process of reassembling my bog stock FI type 3. I haven't noticed any swirl pots or surge tanks, there's nothing in the tank. Are they hidden somewhere? Have I just missed them, or do they not exist? If so does this mean I'll get fuel starvation when cornering hard? Maybe a standard type 3 cann't corner hard enough to suffer this problem.
Baja Wes ·
I know nothing about EFI type 3's, so can't help 🙁
You'd only notice when cornering hard and low on fuel.
kombi_kid ·
hey
all good info here!!! its hard for me to understand it all but its still a good read!!!
cheers
rhys
tassupervee ·
Wes
That is correct to a certain extent but the flow rates specified for most EFI pumps are a theoretical flow rate based on ambient with no restriction.
As you raise the delivery pressure above ambient the flow rate of the EFI pump drops off progressively untill there is a point where the pump will not supply any fuel at all but at its maximum pressure.
This can be easily demonstrated by simply blocking off the outlet of the pump and the pressure should rise thru the roof. In fact, as EFI pumps are a positive displacement type of pump, the pump will simply stall and burn out but for the purposes of the exercise, we will presume the pump is deliovering its highest output pressure with the pump close to stalling. The pump will deliver X PSI. possibly even hundreds of PSI dependant on the power of the electric motor.
Progressively open the fuel outlet and the pump will deliver more and more fuel flow and proportionately less and less delivery pressure.
This is the relationship that must be addressed when choosing a pump to suit a particular horsepower rating of the engine setup.
Those solid state pumps will happily out perform the EFI pump in flowrates when the EFI pump is actually running at system pressure.
This is why I, and a million other aftermarket EFI users, have absolutely zero qualms in running a cheap solid state lift pump feeding the surge tank or as in my case, directly up the bum of the EFI HP pump.
Try removing the tank return line with the pump powered and the engine off and measure the time it takes to bypass a liter.
The do the same with the surge tank LP lift pump.
No brainer. You'll see what I mean.
As far as the construction of your surge tank, that is what I had figured.
I suspected you had the original pump already. It was certainly handy having the removeable top of the original tank as i reckon it would have taken quite some time to fabricate one from scratch.
hence my dissertation on constructing one from scratch.
l8tr E
Herbie ·
I thought we buried this thread months ago.
Why the sudden awakening?
Baja Wes ·
tassupervee wrote:Try removing the tank return line with the pump powered and the engine off and measure the time it takes to bypass a liter.
The do the same with the surge tank LP lift pump.
I have tried that. I put a restriction in the outlet of my LP facet pump to quieten it down, which it did. Put that reduced the flow a little. I found a happy medium where I was getting a little under 2L/min of fuel flow. I timed how long it took to fill a 2L ice cream bucket.
I did a similar thing with my front cut and HP pump when I was playing with it, and I tell you the HP EFI pump flows alot more out the return line (so pumping at rated pressure) than my LP pump. For my pump combo it is a fact the HP pump is out flowing the LP pump. Maybe your combo is different.
I'll also point out the LP facet pumps pulse, whereas the HP pumps don't, so I wouldn't feed one into the other for that reason.
And a bit of tirvia, not all HP pumps are positive displacement. I have seen the sliding vane / roller cell and similar style positive displacement pumps, but I have also seen centrifugal/axial turbine style HP pumps.
Oh, and I didn't have the flange from the EFI tank, I just got a flange made to suit my EFI pump, which has the pump side flange. My surge tank was completely fabricated from stainless steel. If I was to do it again I would use aluminium as it would disperse the heat in the surge tank faster.
Herbie, if useful bits of info keep getting added, then the replies will keep coming. We learn from our mistakes, and other peoples mistakes too. :thumb
tassupervee ·
Hmmm I gotta check out the type of HP pum U R using as I have never seen a lift pump other than any of those generic pulse style things feeding the surge tank.
Never!
On my FJ20det powered van and also on my wifes Pulsar Turbo ET the fuel flow on the return line AFTER the regulator is not nearly that high.
Pulsing into the HP pump is no issue. regulator deals with tha further down the line. Works perfectly
L8tr
E
Baja Wes ·
a few figures for the interested, fuel flow from EFI pumps, all at 40psi.
So finding a factory EFI pump that can out do the facet pump at typical fuel rail pressures isn't too hard.
Most people won't use a facet as a feed pump, and instead opted for the more expensive Carter lift pump. Since my car was taking about 45min for the fuel to get too hot, I figured I only required a low performance facet pump to do the job.
But again there's a million ways to do it, and whatever works is fine.
tassupervee ·
Yes indeed!
L8tr
E
pete wood ·
If anyone is interested, I run the same type of system in my car (EJ25 buggy) as wes does and it works absolutely fine. The only difference in mine is that the surge tank is at the back of the car, not under the front main tank.
spectreeng ·
Has anyone tried putting a pressure regulator right after the hp pump, with the hp pump underneath the car so that the "return" fuel doesn't get heated up? I'm thinking with this method you could tee it back into the hp pump feed line since you're not introducing heat into the fuel.
Baja Wes ·
spectreeng wrote:Has anyone tried putting a pressure regulator right after the hp pump, with the hp pump underneath the car so that the "return" fuel doesn't get heated up? I'm thinking with this method you could tee it back into the hp pump feed line since you're not introducing heat into the fuel.
The way normal EFI regulators are set-up is before the regulator is the regulated pressure, and after it is no pressure (only what is required to return the fuel). So if you put the regulator before the injectors then you don't have any pressure at the injectors.
GeorgeL ·
One thing that I didn't see addressed in this very informative thread is the question of why the fuel heats up.
While the fuel may pick up some external heat in the fuel rail, most of the heating is done by the pressure regulator. The fuel pump pushes the fuel along, putting energy into the fuel that it takes from the electrical system. When the fuel pressure is dropped by fluid friction in the pressure regulator the energy that went to pressurizing the fuel becomes heat, increasing the temperature of the fuel. If the fuel is recirculated directly to the inlet side of the pump it will get hot fast as it goes through the regulator again and again.
This is particularly bad when the engine is _not_ making power, as the fuel pump is constant flow and most of the fuel goes through the regulator at low power settings. At higher power settings, considerable fuel flows through the injectors and is replaced by cold fuel from the tank.
The upshot of this is that recirculating a small volume of fuel at low power settings can result in rapidly heating the fuel past the vaporization point. You can fix this by running the fuel through a finned cooler or simply by returning the fuel to the main tank, where the large skin area of the tank will cool the fuel sufficiently.
Baja Wes ·
GeorgeL wrote:While the fuel may pick up some external heat in the fuel rail, most of the heating is done by the pressure regulator. The fuel pump pushes the fuel along, putting energy into the fuel that it takes from the electrical system. When the fuel pressure is dropped by fluid friction in the pressure regulator the energy that went to pressurizing the fuel becomes heat, increasing the temperature of the fuel. If the fuel is recirculated directly to the inlet side of the pump it will get hot fast as it goes through the regulator again and again.
Hi George,
My first thought is, that ridiculous. But then I figured I better back it up with some math. 🙂
Assume a HP pump is flowing 2.5L/min of fuel through the fuel rail, and is pressurising it to 45psi (the regulator set point). It can be easily calculated that there is 12.9W of energy in this fuel, so dropping the pressure back to zero through the regulator could only possibly generate 12.9W of heat energy. In reality less than that would be converted to heat.
So how much heating will 12.9W of heat do? Not much. When your driving you'd be using about 6L/hr of fuel. If your surge tank was completely insulated then your 12.9W of heat would try to warm the 6L (and would only have 1 hour to do it before it was replaced by another 6L of fuel). Knowing the specific heat of petrol you can calculate that you'd only heat the fuel by 4 degrees in that hour (assuming perfect insulation). That isn't enough to be a problem.
Then if the surge tank wasn't perfectly insulated you could calculate how much it cools. For a 100x100x200 tank you can calculate the expected cooling due to natural convection, and what the equilibrium fuel temperature would be (with 6L/hr passing and 12.9W heating). I get the expected fuel temperature rise to be only 2.7 degrees C.
So basically the heat generated by the regulator is virtually nothing and has a very very little effect.
My fuel rail is cast aluminium, and bolts to the manifold. So there is heaps of heat conduction from the head to the manifold to the fuel rail and into the fuel.
pete wood ·
So Wes, does the fuel heat up much off the inlet manifold in your car?
I know you were haivng some dramas with something like a vapour lock a while back. Did you findout what it was and why?
Baja Wes ·
the vapour lock thing is described earlier in this thread. It was when I had the fuel rail return line going into the surge tank. So the heat from the intake would heat the fuel as it went through the rail. After 40min the fuel in the surge tank would be hot enough to cause problems and make the car stop momentarily.
Now I have a LP pump circulating fuel between the surge and main tank I don't have any problems. It keeps the temp even between the two tanks, and the surface area of the main tank is enough to keep it cool.
GeorgeL ·
Baja Wes wrote:
GeorgeL wrote:While the fuel may pick up some external heat in the fuel rail, most of the heating is done by the pressure regulator. The fuel pump pushes the fuel along, putting energy into the fuel that it takes from the electrical system. When the fuel pressure is dropped by fluid friction in the pressure regulator the energy that went to pressurizing the fuel becomes heat, increasing the temperature of the fuel. If the fuel is recirculated directly to the inlet side of the pump it will get hot fast as it goes through the regulator again and again.
Hi George,
My first thought is, that ridiculous. But then I figured I better back it up with some math. 🙂
Assume a HP pump is flowing 2.5L/min of fuel through the fuel rail, and is pressurising it to 45psi (the regulator set point). It can be easily calculated that there is 12.9W of energy in this fuel, so dropping the pressure back to zero through the regulator could only possibly generate 12.9W of heat energy. In reality less than that would be converted to heat.
So how much heating will 12.9W of heat do? Not much. When your driving you'd be using about 6L/hr of fuel. If your surge tank was completely insulated then your 12.9W of heat would try to warm the 6L (and would only have 1 hour to do it before it was replaced by another 6L of fuel). Knowing the specific heat of petrol you can calculate that you'd only heat the fuel by 4 degrees in that hour (assuming perfect insulation). That isn't enough to be a problem.
Then if the surge tank wasn't perfectly insulated you could calculate how much it cools. For a 100x100x200 tank you can calculate the expected cooling due to natural convection, and what the equilibrium fuel temperature would be (with 6L/hr passing and 12.9W heating). I get the expected fuel temperature rise to be only 2.7 degrees C.
So basically the heat generated by the regulator is virtually nothing and has a very very little effect.
My fuel rail is cast aluminium, and bolts to the manifold. So there is heaps of heat conduction from the head to the manifold to the fuel rail and into the fuel.
You calculations are correct, but you're assuming perfect efficiency from the fuel pump, which doesn't seem to be the case. I looked up the typical current draw of a fuel pump and it is in the range of 5-8 amps. At 14V, this would be a power input of 70-112W.
Now, let's consider surge tank and plumbing of volume 1 liter. one liter of fuel would have mass 0.73kg. The specific heat of gasoline is about 2020J/kg C
That means that, if we consider the 100W or so of power going into the pump to be all going into heating the fuel, the liter of fuel will warm by about .067C every second, or about 4 degrees per minute. Gasoline has a wide range of boiling points. but if we assume a fuel temperature in the tank of 20C and a boiling temperature of 60C we have about 10 minutes before the fuel in the surge tank loop will reach 60C simply from the power input of the pump.
Now, there are a lot of assumptions here.
I've neglected heat loss from the system by convection. I don't think that this would be all that much as the fuel circuit tends to be cooler than the ambient temperatures near the engine.
I've also neglected the replacement of fuel in the loop as it is burned. If the system burned a liter of fuel every 10 minutes while idling it would keep the fuel at about 40C and therefore a bit below boiling, but that would be a fairly high idle consumption!
Add to this the heat input from your fuel rail on the manifold and it is pretty easy to see how the fuel can get too hot pretty fast.
This is why I feel that the regulator and pump do make a significant contribution to fuel heating, particularly at low power settings.
George
Baja Wes ·
George,
In your first post I thought you just meant the regulator causes a lot of heating. Now I see you mean heat from the pump itself. Yes these pumps are very inefficient, so you put quite a lot of electrical energy into them and get a relatively small amount of pressure and flow energy out. That's why I was saying the heat from the pressure change across the regulator is small.
My pump doesn't run the high amps you've quoted. That seems like the peak amps with a blocked outlet. I think a typical HP pump is closer to 3-5amp. I run a HP and LP pump off the same 10amp fuse with no issues, so I doubt the HP pump is over 5 amps.
The calc should be done with the car driving, as not many people leave their engines idling for more than 10 minutes.
Also most people with small seperate surge tanks use external pumps, so the heat from the inefficiency of the electric motor and mechanicals of the pump do get dissipated as convective heat more so than my situation where I have an intank pump in the surge tank.
We should be quiet now before someone large and hairy gives us a wedgy 😀
choppa ·
Just my 2cents worth Wes.
Initially you said the fuel in your swirl pot was boiling.
Could this be cos the pot wasnt vented properly , I have heard that liquid and specially fuel will boil at room temperature if in a vacume.
I say this cos I remember when I first built my fuel system , similar to yours with no lift pump my pot boiled too.
To fix this I enlarged the gravity feed from the tank to the pot to 5/8 I think or maybe 1/2 inch. I also put a 3/8 breather line all the way up to the highest point in the boot(from the pot) and connected it to the fuel filler pipe and then down to the front of the car near the axle as a open vent to atmosphere. This sorted out the boiling. But the fuel still gets hot and my lines are external, under the sill.
These days the only time i get any stalling or stumbling is in 3rd at full noise at the strip, if the fuel level is below 1/4 full. Thats wierd cos my pot would be close to 2 litres capacity and made of 5 inch exhaust tube.
So my question is. In this thread I have seen no mention of vents. I would be interested to see any tried and tested vent systems.
Mark
Baja Wes ·
first, regarding boiling fuel. Any liquid boils when it reaches the vapour pressure, and the vapour pressure varies with temperature.
We are all familiar with water. At 25 degrees it's vapour pressure is 3.2kPa(a). [where (a) is absolute pressure] Atmospheric pressure is 101.3kPa(a). So you would need a vacuum pressure of 98.1kPa(g) (guage pressure) to boil water at 25 degrees.
The vapour pressure of water at 100 degrees is equal to atmospheric pressure, therefore that is it's boiling point.
Petrol is much different. It's vapour pressure is much closer to atmospheric, and it's boiling point is much lower. It's hard to find exact numbers for those. But anyway...
I ran a pair of 10mm lines into my surge tank, but as there was no real flow the temp in the surge tank still climbed.
Putting a vent on the surge tank would be a bad idea. As you start getting hot vapourised fuel returning to your surge tank, it raises to the top, and out your vent. The advantage is as fuel goes out the vent, new fuel will come in your other lines and prolong the length of time until you get a cavitating pump / vapour lock type problem.
The disadvantage is that fuel going up the vent line probably keeps going along the vent line and out to atmosphere. It probably isn't doing your fuel economy any good.
It is also bad for the environment, and probably illegal. Cars from somewhere around the mid 70's were fitted with carbon cannisters to catch the fuel that tries to escape out the vent line. If you fit a later model motor you also legally need to run the carbon cannister or whatever emmissions gear the donor car was fitted with.
The carbon cannisters are designed to deal with a little amount of fuel. If you are boiling off the fuel in the surge tank at a quicker rate, you will likely flood the carbon cannister, and eventually fuel will leak out the vent line. This is a fire hazard, and the carbon cannister a petrol bomb of sorts.
Basically car manufacturers try to keep the hot fuel away from the vent, to prevent the evaporation / vapourisation fuel loss. This is why normal cars have the fuel return line go to the bottom of the tank, near the swirl pot. By the time the hot fuel from the return line makes it's way through the tank and near the fuel level surface, it should have cooled down.
Turbo54 ·
This is hurting my head.... anyone else feel like this.
I have just finished designing and laying some of this out in my car. I am trying to make it as simple as possible and was even thinking of not running a looped system like my friends 1000hp LS1. I decieded to go looped in the end and the system goes like this. This is all -6 (10mm).
Tank – gravity feed through a strainer into top of small 750ml surge tank.
Surge tank gravity feed into bosch 910 pump.
Pump – filter
Filter – into fuel lines that run through tunnel to back of car.
This section hasn't been worked out yet as I have been waiting to get the injection manifolds back for fabricator.
SARD rising rate regulator – into return pipe back down tunnel and into surge tank.
I understand that anything compressed ie air from a turbo or fuel from a pump is going to get hot but from what I understand from here is that I am going to be circulating really hot fuel. Also I was hoping that the air bubbles and vapours would rise back up into the main tank from the surge tank be itself. Will this become a vapour lock. Should that peice of hose be -10 to allow for air and fuel.
It is very similar to Herbies drawing on the second post.
T54
Baja Wes ·
what you have described is very very similar to my first set-up, except I had an internal pump in the surge tank.
It looked like this;
It would be fine for about 40minutes of driving, and then the fuel in the surge tank would get too hot and the engine would cut out. I would expect you would have the same problem.
T-34 ·
Just reading through this thread as I'm about to sort out the EFI on my car and something struck me about the heat problem you were experiencing in the surge tank.
These are just my thoughts / opinions - so they could be wrong - but just a little food for thought...
Without a regulator the pump would flow the maximum rated flow (provided the fuel lines were the correct size and provided no restriction).
The regulator is added to regulate the pressure to the fuel rail such that it is a constant pressure. The regulator acts as a restrictor in the line - restricting the flow and thus maintaining the pressure on its primary side.
The flow restriction places extra load on the pump which in turn generates extra heat within the pump itself.
Normally with a submersible pump the fuel in the tank acts to cool it, with the pump sitting in such a small tank this does not happen - it simply heats the fuel and the tank up as there is not enough volume of liquid to dissipate the heat
This heated fuel in turn heats up the injector rails when it is circulated through the system.
Just my 02c
Mick.
Baja Wes ·
T-34 wrote:This heated fuel in turn heats up the injector rails when it is circulated through the system.
My intake (which the fuel rail is bolted to) is consistantly way hotter than my surge tank. So there is no way the fuel is heating up the injector rail (as the fuel is cooler than the intake that the injector rail is bolted to).
My system has been working perfect for the last (almost) 4 years 🙂 . I would like to build a bigger capacity fuel tank with an in-built surge tank though. It would certainly clean up the front of the car. I am also a little concerned that on big hits offroad the long tie-rod will flex enough to take out the surge tank (I have found evidence that it has flexed and rubbed on the surge tanks fuel lines). Ah, there's always something to do to the car...
kombi_kid ·
so the low pressure pump just pumps the fuel between the the surge tank and the fuel tank and the high pressure pump pulls the fuel from the surge tank into the fuel rail and back to the surge tank?
going to put a surge tank in my bus i think on hard cornering with no surge tank it suffers a little
cheers
rhys
1303Steve ·
Hi Rhys
If a Kombi tank is the same as yours, try and get a factory EFI one.
Steve
kombi_kid ·
i was thinking that they would be the go!
cheers
rhys
leweydafly ·
T-34 wrote:Just reading through this thread as I'm about to sort out the EFI on my car and something struck me about the heat problem you were experiencing in the surge tank.
These are just my thoughts / opinions - so they could be wrong - but just a little food for thought...
Without a regulator the pump would flow the maximum rated flow (provided the fuel lines were the correct size and provided no restriction).
The regulator is added to regulate the pressure to the fuel rail such that it is a constant pressure. The regulator acts as a restrictor in the line - restricting the flow and thus maintaining the pressure on its primary side.
The flow restriction places extra load on the pump which in turn generates extra heat within the pump itself.
Normally with a submersible pump the fuel in the tank acts to cool it, with the pump sitting in such a small tank this does not happen - it simply heats the fuel and the tank up as there is not enough volume of liquid to dissipate the heat
This heated fuel in turn heats up the injector rails when it is circulated through the system.
Just my 02c
Mick.
i think you hit the nail on the head there. better cooling having the pump submersed in the main tank or out in the open, not in the surge. my setup on my one tonner runs from te standard pickup in the tank through a low pressurelift pump, which fills the surge tank, surge tank overflows back into the main tank. the high pressure pump is mounted to the chassis where it gets good airflow and doesnt heat up much. runs through a filter then the rail and returns back to the surge tank. the return to the surge tank goes into the top of the surge tank along with the return to the main tank so the warm fuel is pushed prety quickly back to the main tank. never had a problem with any of it overheating or problems with power. same setup on a lot of comp off roaders 've seen (bought the kit from a 4wd conversion mob) this system should work with any setup
sorry if i've doubled up on what others have said 🤪
cheers
mick😎
Baja Wes ·
Mick, the set-up you've described is basically exactly the same as what I am running, but I have an internal pump in the surge tank, and you have an external pump. There is basically no difference there.
As I've already stated in this post, I disagree with the theory that the pump does any significant heating of the fuel. My cast aluminium fuel rail is bolted directly to the aluminium intake which in turn is bolted to the alumium head. They conduct heat to each other very well, so all the heat from the head and cooling water goes straight to the fuel rail. This heat is the problem.