shawnk

My supra TT pump died after 700 miles. I bought it from rotormotorsports so it looks like I have no recourse. I was thinking about going to 2 pump setup anyway so now I am headed in that direction. The question is one pump to each rail or bring both pumps together and then split back out so if one fails there is still fuel to both rails? This is what I was thinking:

Drill top tank plate and run a 6 AN bulkhead union there. Line from original pump and line from new pump (through new fitting above) to a 2x 3/8" - 1 x 1/2" Y, then to a 8 AN filter and then back into Y (like above) and split back out to 6 AN to each rail. I would use the hard lines up to the engine (my vapor one has been capped for a while so I'll just use that for the additional). That way should one of the pumps fail a single pump could still supply both lines. Thing I am worried about is - are the pumps going to stress each other? My initial thought is if I have a fuel pressure regulator then there is no way pressure can build higher no matter how many pumps on it and therefore backpressure on the pump will be the same no matter what and it shouldnt stress the pumps. But when I ordered the parts the guy at summit seemed pretty against the idea.

Thanks for the replies,

Shawn

kabooski

I'm going to run 2 walbro 255's
($90) each
I had the same argument with my self as to running
a separate line to each fuel rail
But you know what? the primary rail is way smaller then the secondary which means it has more fuel pressure

I going to run about 8" to 12" of individuel -6 or -8 lines to a NoS Y
then feed the rails regulated by a SX FPR on the outlet of the secondary rail (one port used)
and a "maybe" keep using the stock return line

The very important point here is not to wire both fuel pumps together (link the power from pump to pump)

Run 2 independent + and run them each with there own relays

Carl Byck

One consideration with two pumps vs one, is that with one big one, if it dies, you have no fuel, and the motor does not run. With two, one pump dies, the other supplies enough fuel for you to blow your motor. i realize that is an oversimplification, but it is a recorded problem on two pump Supras. Just take it into consideration in your design. In a Supra (1 rail) a pump supplies each end of the rail, with a return in the middle. Carl

fatboy7

I'm sure if you're innovative enough, you could set up some sort of warning system, to tell you when a pump has gone bad. If I thought about it, I'm sure I could think of somthing....

fd3st04r

best way to tell is with fuel pressure gauge. if the fuel pressure drop during wot, you know something is wrong. 2 255lph pumps will be much better with even one pump it should support 400hp. no worries

Marcel Burkett

A drop in fuel pressure would be a bit difficult to spot if one pump dies at w.o.t. , I am sure we are all quite familiar with the rapid and sometimes violent change in the cars behavior at mid to wide throttle , I for one would much rather be looking at the traffic and the road ahead !! , as Carl said ........., I would prefer the car stop running if a single pump dies and pay a tow truck than have to pay " rebuild bills " and suffer through the extended down time .

IRPerformance

iTrader: (41)

If a single pump dies while at wot you will most likely pop the motor. This is the whole reason why I want to run 2 pumps

fatboy7

I, for one, have never seen a fuel pump just suddenly die. I'm sure it happens, but most of the time they get sick before keeling over.

in2twins

I run the WB O2 all the time, so hopefully I will see it coming. My setup is a large in tank pump, with a smaller external pump supplyng a surge tank(TBI). Carl Byck

jspecracer7

If that was the case, I think that a fuel pressure gauge would fix that issue.

I have two stock FD pumps running to the front, one to the primary, one to the secondary. I used the aeroquip regulator as the "Y" and then the return line goes to the charcoal canister line back. Once it ends, I have hose going to a bulkhead fitting on the fuel tank.

shawnk

Well my pump was fine as far as I could tell then one hour later I tried to start my car and it was dead - no pressure at all. Switched back to stock pump and started right up. It was very weird but overall a pretty painless stranding as I was 5 min walk from my house and had the stock one in my garage to put back in.

I have been thinking about how to tell both pumps were working and this is what I came up with.

You can't prevent one or two pumps from failing at WOT no matter what you do so I am not going to worry about this instance - if it happens it happens and I get that port I would like to have.

There is the possibilty that if you have 2 pumps at WOT and one drops you could survive it where if you only have 1 you probably wont. But the danger of having 2 pumps is one could fail and you wouldn't know it failed and then next WOT you would loose your engine so this is the real problem to solve.

The first way I thought was 2 fuel gauges - sure $150 more to spend and gauge space is not unlimited but not hard decision to make. So I use up my center dual gauge pod for 2 fuel gauges. But the problem here is that no matter where you put your sensors it will be difficult to tell a failure because the other pump will provide pressure to whole system (and that is the idea right?).

So the first thing you would have to isolate the areas where the sensors are. I found a Fuel Check Valve at summit for 60 bucks that could go between the pump and the joint of the lines that would not allow backflow in the event that one pump dies. Put the sensor between pump and check valve and you would know for sure what each pump was doing. This would add about $120 to the total cost (not including gauge or fittings). I am actually considering this because I have done my fuel system 3 times so far and I would like to get it so I dont have to worry about it anymore. From where I am right now I would have to spend $300 and I would be done. (dont get me started on how much I have spent on this fuel system so far).

What do you think? I am going to post this and then add up the whole fuel system if I did it this way the first time to see what I get for a total cost.

Shawn

Marcel Burkett

If you put a check valve after the discharge of the pump and the pump dies you still wouldn't be able to tell 'cause the checkvalve would close and prevent backflow through the pump but the dead pump will also hold a pressure between itself and the checkvalve as most (if not all ) pumps also utalize an internal checkvalve , ever wonder why F.I . systems remain pressured when switched of ? . I know that you are thinking that your system maybe depressures when the pump stops , most times this is because the aftermarket FPR is normally open and closes when boost is applied , MAZDA's stock regulators work opposite to this , they are normally closed (at atmospheris pressure and above) and are "pulled " open when vacuum is applied . This is why you need to depressure a stock system before working on it , the pressure is maintained between the pump and the fuel rail / regulator .

shawnk

Thanks for the reply - I was afraid of that. So check valves and a second guage are not going to work.

Now the goal is to make sure the second pump doesnt fail without knowing so what about this....

Putting a separate switch (one that is very difficult to turn off or requires constant pressure to turn it off so that a mistake is not made) on the power to each pump and then at startup do a quick check of each pump by turning 1 pump off then ignition to on position and watching fuel gauge, then the other.

Takes a second, might add a chance of making a mistake by switching off one by accident, and not the best solution I am sure but a bit better than going blind.

Anyone else have another idea to monitor 2 pumps running in parallel? Splitting them completely would make it possible to put 2 guages on them but it does double your chance for failure. So I'd like to keep both pumps feeding both rails.


Shawn

rotorbrain

well, could you hook a switch up to the pump itself to see if it is doing anything. i dont know if the ecu reads what the pump is doing or not, but i figure a light or something could work. . . like if the switch/sensor reads that the pump isnt cycling it would then do its thing and close a circuit to a light. . . if that can be done of course.. . im not fluent on how the fuel pump system works.

paul

maxcooper

My fuel system is one in-tank A'PEXi pump (not cheap, but the Bosch sounds like it would work for many applications, including mine, for less money), stock lines wherever possible (adequate for my power levels, cheaper, lighter, and easier to work with than SS), and only recently did I add an Aeromotive FPR since I think the stock one doesn't flow enough to dump all the extra fuel when the pump goes high-speed at low RPM/load and seems like it might be slow to react under boost sometimes. Both phenomenon seemed to be evident on my fp gauge, but I'm not 100% sure as they were just casual observations. And I still need to remove the stock FPR to complete the change.

I generally prefer a single pump since installing two pumps seems to double the chances of having fuel pump failure trouble, and may be harder to detect. Having two pumps that each flow enough to support the full power of the engine would probably be the safest setup, as long as they were plumbed such that all the flow will still get to the engine when one fails, and you can detect when one has failed.

I think the dual pump---fp_gauge---check_valve---union should work for detecting a failure. If one pump totally fails, the check valve will close and the pressure in the line between the failed pump and the check valve will be lower than the gauge reading for the good pump.

Some kind of light wires in series with the pump would indicate continuity, but you'd have to be very careful how it was wired so you don't lower the voltage available to the pump too much and the light can handle all the current that the pump is going to want. The more I think about it, it seems like a light in series wont work. Perhaps someone with some electronics knowledge can offer a solution for monitoring fuel pump continuity.

-Max

Heath

iTrader: (11)

Max, which Bosch pump are you referring to and where can I locate it? Will it support a t-78 at 20 psi?

maxcooper

I'm not sure of the part number, but it is often referred to as the "Bosch Motorsports" pump or something like that. I think it was used for a bunch of high-HP Porsche race cars. But that might all be rumors and hearsay rather than fact. I'm not totally sure about it.

Anyway, Wade (http://www.wvinter.net/~flanham/wlanham/fuel/pumps.html) has some info on it, and Wael (http://www.efini.net/fuel.htm) has some info on his install of the one with the nut on the bottom.

You can use my fuel calcs page (http://www.maxcooper.com/rx7/how-to/...tem/calcs.html) with the data from Wade's site to help determine if it will be adequate for your needs. I punched in some numbers and my guess is that it might not work with a T78 at 20 psi. If you can keep the pressure (effective + boost + drop) below 75 psi, the A'PEXi pump (http://www.maxcooper.com/rx7/parts_i...exi_bnr32.html) might work for you. Don't go by the HP figures from my RC data sheet, though -- punch the numbers in my fuel calcs sheet and see if the pump flows enough at the pressure you need.

HP ratings for fuel pumps are rough estimates for certain applications at best, flow at some pressure are also worthless unless you plan to run at the rated pressure. You really need a graph or table of flows at different pressures (including a test voltage) to determine if a pump is suitable for your application. RC will flow test a pump for you at a single voltage for about $50 if you want to check out your pump before installation. It just has to flow more than you need, as the regulator will dump the rest. Having a grossly over-acheiving pump won't get you any more fuel than a merely adequate one. But it is a good idea to have some headroom for safety -- the pump voltage changes under real-life conditions (radio, rad fans, etc.) and I am sure flow changes with temp as well.

-Max

shawnk

I was thinking about an electrical solution. But that is one area I dont know enough to even guess what a good way would be. Maybe just a small LED/bulb for each pump mounted in the pod below the guage. Or some sort of LCD for each with a voltage reading mounted somewhere. Anyone?

Shawn

Heath

iTrader: (11)

Thank you for the very informative response Max. Is the A'PEXi pump actually a different brand pump rebadged?

Carlos Iglesias

iTrader: (6)

Hey guys,

Here's a hint on THE Bosch pump:

'79 930 "front" pump

So now the secret is out officially, so somebody owes me a Guiness on this one...

setzep

part #'s for bosch fuel pumps:
100l/h, Y 580 700 118
165l/h, 0 580 254 979
200l/h, 0 580 254 044

all are rated at 5 bar (72.5psi)
not sure on the voltage though

I think I have a solution to the "how do you know if one of the pumps has gone bad" thing.
plumb each like this: suction - pump - normaly closed pressure switch hooked to a highly visual light - check valve- fuel line going to rails.
If you did that the light on the bad pump would turn on when the pressure went below whatever the switch was set to. Then you could see the pump wasn't holding X amount of pressure.

setzep

Here is what I'm talking about.
Not sure what setting the pressure switchs should be set at, what do you guys think?

shawnk

The thing about that is if the pump stops the check valve will hold pressure in that segment. Now when you turn off the car (pressure loss in system allows check valve to open and depressurize) and then turn it back on you would be able to detect the loss. Where can you find a pressure switch like you are talking about? 6AN (3/8" ID) maybe 25psi rated. I am still leaning toward a completely electrical monitor but I am starting to realize how much of my college physics I have lost over the years. I started a very similar drawing and will post tonight.

I also think I have decided to run the pumps on high all the time to simplify the system.

Welcome to the discussion setzep.

Shawn

maxcooper

It does seem that it would hold some pressure, but it also seems that it would leak down pretty quickly. If you were comparing two gauges, for instance, one of them would not rise with the boost pressure if the pump was dead.

-Max

Badog

The Walbro check valve is on the outlet nipple and can be removed. It looks like a Schrader <sp> valve.