EFI V8 retrofit into 70s TVR

[slowM]

Hello all,

All of the items listed with a part number, have been purchased. This still leaves me with a good number of questions.

Fuel System:
Fuel Pickup (In-Tank) ......................... Paxton, part # 8002122
Fuel Filter: .................................. Paxton, part # 8001689
LP Pump: .........?
Surge Tank: .....?
HP Pump: ...................................... Paxton 1000 HP, part # 8001688
Filter Element: spec? ... part # ?
Injectors...................................... Bosch, part #0280155968, type - EV1 "Green Giant" with connectors, Impedance - 11-18 ohms, Flow Rate - 42 lbs/hr, 440cc,
Fuel Rails: .....................5/8" ID billet Paxton, part # 5007026-2
Fuel pressure Regulator: ........... Adjustable Paxton, part # 8001690

This is essentially what I have in mind: (or see attachment)
Pickup (8AN out) → Filter → Low Pressure Pump → Surge Tank/Swirl Pot ^ top → Return, bottom → High Pressure Pump → Rails → Pressure Regulator → Return.

Having thought this through, I'm left with plenty of questions:

What filter element (or filtration size in microns) should I use?
When feeding the high pressure pump from an external surge tank, I suspect that the flow capacity of the LP pump should slightly exceed the calculated maximum fuel usage, at any point in the rev range. Is there a standard correlation/factor, or a ratio to HP, that determines the delivery volume of the LP pump?
I will probably run an Aeromotive speed controller*, for the HP pump. Would I want to speed control the LP pump in a similar way, to maintain some form of parity?
For what reason would one want a filter between the pumps? I have read about one design where there was a filter after the LP pump, and another, after the HP pump. What is the benefit of this design?
What factors are usually considered, for sizing the surge tank/swirl pot? I have read, that the base line should be the time it would take to negotiate the longest corner, multiplied by the maximum flow rate of the injectors. I'll avoid the skidpad.
I like the Walbro pick-ups, but am not certain they will fit in the tank. Do any of you know what diameter they are? The dimension I am finding on Amazon is +5”, but that could be for the box. I seroiously doubt that I can fit anything much over 1.75” diameter. Is there a similar product, that meets my diminutive sizing restrictions?
Am I making any errors, that I'm not seeing?

*

http://aeromotiveinc.com/products-page/accessories-electronics/fuel-pump-speed-controller/

Many thanks, in advance, to any and all who can give me a little more insight, and best regards,
B.

 

[slowM]

Oops.

|

http://thumbsnap.com/Sbfy4mgn

[/url]]

 

[JayMaechtlen]

Factory vehicles usually have a filter sock in the tank to keep rocks and dirt out of the pump.
Then a good filter after the pump to protect the injectors.
In your case, you want a good filter between the HP pumps and the injectors.
If your LP pump is very sensitive to contamination or you expect to run on really dirty fuel, you might put a better filter on its inlet side.
If you don't expect the LP pump to throw chunks of itself into the fuel, you might forgo the suction-side sock for the HP pump.

I see you're wanting to size the swirl pot. If you're wondering about corners, a question might be: How much power will you be making in a corner that's steep enough to cause issues in the main tank? And how long in that cornering mode? That would suggest much fuel you'd need.

Under full power in a straight line, can the stock fuel pump keep up with the intended engine?
If the engine can draw much more fuel than the LP pump can deliver, does the swirl pot return line allow air from the tank back into the pot as the pot's level is drawn down?
(or is there a suction bleed intended?)

cheers
Jay


Jay Maechtlen

http://www.laserpubs.com/techcomm

 

[slowM]

Jay,

Thank you. Better safe than sorry, to protect the components makes good sense. I'll just put filters before and after each pump. The engine represents too much of an investment, to me, to destroy on being cheap. I will admit that I do resent every ounce of added weight.

As for the swirl pot, corner 8 at Istanbul park is renowned as being amongst the longest corners in racing, and it might take 14 seconds at 80% power.

Therefore, if we use 6cc/sec x .8 max power x 8 injectors x 14 seconds = ±540cc meaning, if I had anything more than a 1 1/2 liter surge tank, or swirl pot, It would be overkill, in my view.

I'm not sure what you are addressing with the question regarding the stock pump. That was a cam driven mechanical pump, on a TR6 motor. The High Pressure Pump is an older version of the Aeromotive A1000 pump, with the Granatelli name still on it.
• -10AN inlet and outlet
• Flows 700 lbs/hour @ 13.5 volts/45 psi
• Fuel injected engines:
Supports up to 1300 HP for naturally aspirated engines.
By my calculation, 6cc/sec x .8 max power x 8cyl / 1000 x 60 x 60 = 138 liters per hour. I was planning to use a 255lph in-line pump, is that too much? I am concerned about heating the fuel, and constantly flowing it through the swirl pot vent pipe, and the electrical load being a continuous one, on that circuit, that is why I wonder about some form of speed control, on the lp pump.

I don't know what a suction bleed is, please edify me. I did think the vent pipe between the top of the swirl pot should be continuously open.

Best,
B.

 

[JayMaechtlen]

I was confused on the LP pump. Was thinking of an oem pump in the tank, which isn't what you have.
I don't know what really happens in the system in a hard corner - if you are depending on the volume of fuel in the swirl pot, that means that the volume of fuel in it is decreasing.
Which means that either it pulls a vacuum, or the return line from pot to tank must change direction of flow and pull some mix of fuel and air back into the pot from the tank.
I am theorizing - I haven't done this myself.
I was/am considering such an arrangement on my project car, to compensate for fuel starvation in hard corners and to provide additional capacity under wot.
(OEM tank and fuel feed was limited. I should upgrade everything...)

As far as fuel heating and electrical load - if it's a street machine, I'd think the controller would be a great idea.
Not cheap, but outside of DIY, nothing is.

Re size of swirl pot - prob want to allow enough reserve depth and/or baffles so its pickup stays well submerged under the conditions that make it necessary in the first place!
Regards
Jay

Jay Maechtlen

http://www.laserpubs.com/techcomm

 

[crerus75]

slowM,
I skimmed your posts and have a few thoughts off the top of my head. I worked in the aftermarket industry as a fuel injection test technician and used to spec fuel systems, so I might be of help. The rule of thumb I always used was to utilize a 100-150 micron filter before the pump or pumps, and a 10-25 micron filter after the pump(s). I like the billet "rebuildable" filters that have either cleanable or replaceable filter elements. The coarse filter protects the pumps without being so restrictive or easily plugged as to cause cavitation. The fine filter protects everything else.

Be aware that one of the biggest sources of contamination that I saw when dealing with aftermarket systems was debris from assembly. I can't tell you how many injectors I cleaned that had bits of rubber in them from the owner/builder/tech cutting and assembling the AN lines and then forgetting to flush them clean. Make sure everything in the fuel system is spotless, because if you don't find it, the injectors will.

Note that, according to your diagram, you need to build a pressure-tight surge tank that will either gravity-bleed or pressure-bleed back to the main tank. The disadvantage to this is that the surge tank will collect all the hot fuel returning from the engine, which then goes back into the inlet of the HP fuel pump. A better solution IMO is to sump the factory tank if you can and run a single high pressure external pump. There are companies that turn your factory tank into a fuel cell, filling it with anti-slosh foam and sumping it to feed an external pump. Eliminating the extra pump and lines would get rid of a few possible leak points as well as another pump that could fail.

I will probably be branded a heretic for saying this, but I don't like those Aeromotive pumps. They are (or were, at least) noisy and didn't last long. I REALLY like the Bosch inline pumps. The number I recall off the top of my head is 0580254044, which (going from memory-- don't sue me) was something like 67 gph at 3 BAR. It is part of, or is based on, Bosch's OEM pump architecture and is extremely robust. If it meets your requirements, I strongly suggest using it if possible. Make sure that the pump inlet is mounted as low as possible for any pump that you use, since most inline pumps do not like to prime "uphill" very much. Also make sure that you mount the pump(s) in some sort of cushioned mount unless you want to hear it resonating throughout your car's interior.

Don't worry about the pump heating the fuel. The engine will heat it just fine, no matter what you do. I suspect that you could probably get rid of the variable pump controller and not see much of a difference in fuel heating, but I've never used one so I don't want to go too far out on a limb. If anyone has any numbers on fuel heating, I'd love to see them. AFAIK, variable-speed pump control circuits are mostly used in OEM applications that use returnless fuel systems, where pump heating of the fuel might very well be significant.

 

[JayMaechtlen]

I think the OEM returnless systems eliminate the pressure regulator and control the fuel pressure at the injectors with the pump.
In concept you could do that, but...

Jay Maechtlen

http://www.laserpubs.com/techcomm

 

[crerus75]

JayMaechtlen,
The returnless systems I've seen regulate the pressure with a small, non-vacuum-compensated regulator built into the pump module inside the fuel tank. The bleed-off fuel is recirculated back without ever leaving the tank. I'm not aware of any systems that regulate system pressure exclusively by varying pump speed, but I'm not current in EFI any more. I stopped doing it for a living a few years ago, and by that time I was focusing mostly on aftermarket injectors and fuel systems, and very little on OEM systems.

For the record, I wasn't recommending controlling pressure this way-- I think it would be cumbersome to implement, as pressure would have to be controlled by the PCM or by some black box with (I assume) some sensor input. I don't know if the factory systems that vary fuel pump speed try to maintain a constant differential pressure through the injector (as vacuum-compensated systems do), but I can imagine how, umm, interesting it would be to have another tuning variable to work out on the dyno. No thanks. Sorry for any ambiguity on my part.

 

[slowM]

Thanks for the pointers.

I will likely stick with the existing components, until I can afford to have a safety cage, built, and purchase a fire system, fresh restraints, and a proper fuel cell. Right now, budget is dictating a lot of decisions. When the Paxton/Aeromotive unit fails, I will replace it with a Bosch unit, like the one you mentioned. It sounds like it is an industry standard specification, flowing 255lph @ 42.5psi. Is there a standard, or a calculation, that people use, for sizing fuel pumps? The expected power should be below 550 crank hp.

I have some confusion as to how an R.P.M. based speed controller will not affect line pressure, and have been reading, that some people run 43 lbs normal, to 70 lbs @ full throttle. I have no understanding of how or why, and will probably get in touch with Aeromotive, to see if they can shed a little bit of light on speed control, for me. One thing I do understand, is that the speed controller is supposed to extend pump life, by not letting it run at full speed, for every moment of its operation.

As far as temperature control is concerned, I will probably use insulated heat shielding for all of the braided fuel lines, leaving the rails as the most exposed heat sink. I hope to have a cool place for the surge tank, which will be designed to have a tapered, thin part at the bottom.

Best,
B.

 

[slowM]

By the way, do you have contact info, for the folks who modify tanks? Somebody, on another forum, alerted me to the fact that Aeromotive have a product out, that they call Phantom, which is essentially a retrofit kit that sounds similar.

Best,
B.

 

[crerus75]

SlowM, I apologize in advance for the brain dump. Post back if anything is unclear.

The variable speed control certainly does affect fuel pressure. However, most aftermarket EFI and most older OEM systems are designed to work with a constant DIFFERENTIAL pressure through the injector. You can hook a fuel pressure gauge up and see that at idle, system pressure is 36 PSI, and at WOT the pressure rises to 45 PSI. This is because the pressure regulator is vacuum compensated, and as the pressure in the intake manifold rises (vacuum drops), the gauge pressure also rises. However, the pressure at the injector OUTLET also rises, for a net pressure FROM INJECTOR INLET TO OUTLET that is the same at both throttle positions, and everywhere in between (or at least close enough). This eliminates a tuning variable, namely compensating for fuel pressure vs. throttle angle. Were this not the case, the tuner and the engine management system would have to compensate for a lower differential pressure across the injector as throttle opened, the opposite of what you usually want.

This was the way EFI was done for many years. Returnless systems no longer (to my knowledge) vacuum compensate their pressure regulators, because the regulator is in the fuel tank. Variable pump speed can be used to replace the vacuum compensation role of the regulator. I don't know if that's what the OEMs are doing because I haven't looked at any OEM code or tuning strategies for late model cars, but they certainly could do it. I don't know what they're doing with cars that have both a returnless system and a fixed-speed pump, but I imagine it involves more PCM code to compensate for varying differential pressure across the injector.

As for extending pump life, most of the failed pumps I've seen in an aftermarket setting have died from contamination or running dry. Sometimes the fuel you get from the 55 gallon drum has silt in it, or you forget to clean out an AN line after you put it together and you end up pushing all that junk into the pump, or you open up the fuel cell without cleaning the debris off the top and some falls in. Sometimes you get the bright idea to run methanol and then forget to purge the fuel system. Sometimes you can't pit to fuel up, or the fuel gauge is off, or you just plain forget, and you suck the tank dry AGAIN, dammit. Stuff happens, but most quality aftermarket pumps do not wear out very quickly unless they're asked to push ungodly pressures, they run dry A LOT, or they suck in a load of debris. I've seen many cars run multiple seasons in all kinds of racing with no fuel pump problems at all. My gut tells me the fuel pump controller doesn't extend pump life in a limited-use "fun" car enough to matter. YMMV.

The guys who run very high pressure wide open are usually doing it because they need injectors with a very big variation between idle and WOT flow rates. This usually indicates either boost or very high RPM or both. Both circumstances create a condition where the injector is either too big to idle cleanly or too small to make power at the top end. The old way to get around this was to get a fuel pressure regulator that was non-linear, IE a fuel pressure vs. manifold pressure plot would have a steeper slope. The newer breeds of injectors have made this approach somewhat obsolete because they have higher dynamic range. Engine management systems that allow staged injection (a second set of injectors becomes active at a predetermined point) are also a common approach. This problem does not usually arise in most car engines unless they're heavily boosted or have a very high redline.

There are several formulas to calculate fuel flow requirements based on engine characteristics. The quick and dirty one I use is to multiply horsepower by BSFC to get lb/hr of fuel needed, then multiply by fuel density. If your engine makes 600bhp at 0.5 lb/hp hr, you require roughly 300 lb/hr of fuel. The density of pump gasoline in the US is roughly 6 lb/gal, so you require roughly 50 gal/hr of fuel, or about 190 liters per hour. In reality we end up using 255lph pumps an awful lot even when 255 is overkill because they're cheap, commonly available, and do a good job for most of the projects that people build. I really, really like the Bosch pumps from a quality and durability standpoint, but most people will use either an in-tank or inline Walbro 255 lph pump for half the price and never look back. I can't say I blame them. BTW, that same rough calculation works for injectors-- just divide by the number of injectors to get the theoretical power they will support. Assuming your engine BSFC is 0.5, each injector is good for (42 lb/hr) / (0.5 lb/hp hr) = 84 bhp per injector (roughly).

I don't know who modifies the tanks, I just know that there is an outfit that does it. You can purchase a fuel tank sump and modify your own tank (or have it done if you're uncomfortable welding on what amounts to a fuel-air bomb). Googling "fuel tank sump" brings up a lot of hits. I've seen some that have baffles in them to keep the pickup covered even in turns. You can also install the return line such that fuel from the regulator and/or the surge tank is returned to the sump. Chunks of anti-slosh foam are available from companies like JAZ and Fuel Safe. You might be able to stuff the tank yourself, though I've only done it with fuel cells and don't know if the foam blocks would pack in well enough to really reduce slosh.

The Aeromotive Phantom looks interesting, even if it is a bit pricy. If you decided to use it, you could get rid of the surge tank, low pressure pump, and coarse filter entirely. The Phantom has a sump, a filter, and a high pressure pump, and the return line sends fuel back into the sump to keep the pump pick up covered. If you did this, you also wouldn't need to sump the tank. I'd be very tempted to get one and simplify the fuel system. The only unknown is the manufacturer of the pump.

 

[slowM]

crerus75,

Thank you so much for your comprehensive answer. I've just learned a great deal.

I will have to look at my pressure regulator, to see if there is a vacuum port, on it.

On the next round of revisions, your suggestions will doubtless be utilized to the best of my ability. In this iteration, your cautions will be heeded.

Best,
B.

 

[crerus75]

Best of luck to you. This sounds like a great project. Please post back periodically with questions and results.

 

[slowM]

Thank you. It's only a seventeen year project, so far, but I hope this will be the final phase. I will be glad to be able to ask further questions, and will definitely report results.

By the way, here's a link to the build log:

http://www.pistonheads.com/gassing/topic.asp?h=0&f=34&t=1080028&mid=0&nmt=3428TM+Build+Log

Best,
B.