I have had an unlucky string with the fuel system on my '87 roadster as of late.

It started with the fuel system not getting fuel when the tank got fairly low. Basically, the pump runs for 6 seconds (FAST contralled) to 42 psi and then when the pump stops it instantaneously drops to 0 psi.

I installed the pick-up tube through the bottom of the tank. That was an exercise in itself. After adding 10 gals of super unleaded, exactly the same issue. After waiting nearly 20 mins at the local Mobil, it did get fuel psi and I was able to get it home. I thought possibly the car was vapor locking.

Low and behold, one week later, I went out to start the car (totally cold) and nearly 30 prime cycles later, it does the same as always. Pumps to 42 psi then immediately drops to zero psi. The car will not even fire now. Even the manual guage on the fuel rail shows no psi. There was some residual psi in the rails when removed, but no significant volume.

The fuel system consists of a -12AN feed to the inlet fuel filter and pump through the bottom of the fuel tank. It discharges from the Weldon fuel pump through a -10AN to the front of the car. At the passenger footwell, it passes through a large inline filter and continues up the PS firewall. It feeds to the PS fuel rail. It transfers through a -8AN to the driver's side fuel rail. It then passes through an LT4 type regulator that is backed all the way out to get 42 psi. It returns through a -8AN line all the way to the rear of the car and is fed over the top into the PS corner of the fuel tank. The N2O system is totally seperate system, so this is not affecting the main system. There is no rubber line; everything is either hard aluminum line, or SS braided.

Any thoughts at this point would be most useful. The car right now is a shelf unit in the garage.

Aaron

 

I would think it is eather the fuel pump relay or the regulator.
I assume it is an L98 and fuel presure should be 40 psi.
I would check the fuses and the fuseablelink by the battery.
If you have 2 people ,take a test light and I think it is the green wire at the plug at the tank.
Turn on the key and see if you get power for 3 seconds at the tank.
One person to turn key and the other using the test light at the sending unit.
If you have good years you can also turn key on and off to see if the pump kicks on.
Each time you turn on the key the pump should run for 3 seconds.

 

So did I read right? you can hear the pump running but you have no pressure? Have you taken the pump apart yet and see if it self destructed?
P.S. c4's suck when I comes to losing fuel pressure when the tank is low. The cup at the bottom is too small.

 

The Weldon pump will not hold pressure after the pump turns off. Increase your fuel pump prime time in the FAST.

 

Let's take this from the top. And thanks to all who have responded.

REDC4,
Yes it is a L-98. The relay is fine, as the pump definitely runs for 6 secs when you first turn the ignition to the on position. You can definitely hear the Weldon when the line is empty. PSI for my system is 42 psi with no vacuum line, or with no vacuum such as when you just turn the ignition to the on position. The pump does run for a short period of time, I believe 6 seconds if the reset has timed back in. The car must sit with the ignition off for like 45 secs to allow the time out to reset.

Rick,
As mentioned above, the pump is turning on. In a static situation, the psi starts at 0 psi. When you turn the key on, the electronic guage in the cockpit, as well as the manual guage tapped on the fuel rail sweep quite rapidly to 42 psi. Sometimes, depending on how long the car has sat, it may take two prime cycles to hit the full 42 psi. As soon as the pump times out, the psi immediately backs off from 42 psi back to 0 psi. The pump is about 6 months old, probably less than 300 miles, and yes I have had it apart. It is a very stout piece and seems to be in good order. I previously wrote about loosing fuel psi, then routed the -12AN line through the bottom of the tank.

SloR,
I find this comment interesting, as previously the pump would hold the psi. It would drain off slowly, ~10 mins, but now the drain occurs immediately. I too have considered that if I ran the pump continuous (bypassed the time out function), it would probably "band-aid" the problem, but I really don't want to do this.

All,
I spoke with both TPIS and Young's briefly today. Both said it is either the fuel back-flowing through the pump, or passing through the regulator. I run a swing check valve on the pump discharge (feed side of the fuel system), which should not allow backflow, but maybe it has failed. Due to the fact, that suddenly, I seem to be going very rich, TPIS felt that the LT4 type regulator diaphragm has probably failed. This would cause excess fuel to be routed through the vacuum line back into the motor. Both companies felt that installing a -8AN cap on the regulator has a high probability of "spiking" the fuel injectors and killing them. TPIS suggested that I place a vice grip on the braided line return and gradually clamp it off. NOT!!! I think that I will instead connect an additional piece of -8AN to the regulator and see how much flow I get through the regulator following pump time out. I wil also check the vacuum line to see if fuel is passing by the diaphragm.

Any additional thoughts are welcomed.
Aaron

 

To get an idea of what this hook up is,need a little more info.
If you are using an after market fuel regulator ?I have found only Aeromotive can make one that won't fail .
Check your oil and see if you smell fuel in it.If so I would assume the regulator failed.Been there done that and locked up the motor.(Accel reg did it to me 2 times )
The L98 stock fuel regulator is to be able to hold fuel presure at all times.
If it does not hold presure I would think about the injectors?
Are they new or ?
If the gas cap does not seal tight it can add to problem.
The only 2 things I can think of that will not allow the fuel presure to stay up is an aftermarket regulator which I use and presure drops in about 5 min to anout 5 or 10 lbs.
Or leaking injectors.
You may need to check the return to the tank the way TPIS informed you .
That is how I checked mine when I was having problems.
I had to install a back flow valve in line of my barry grant pump.
The worlboro and grant pump were not working together and causing an air pocket in the tank , fuel was sloshing and causing the grant pump to suck air.Are you running the evaporation loss can.That hose also goes to the fuel tank and opens a diaphram and may not let the tank build presure.

 

Aaron, do you have your pump hooked in to the low oil pressure switch? If your oil pressure cut off switch is bad the fuel pump will not work if that is how you have it set up electrically. I wired my fuel solenoid for my primary pump into the factory wiring so that in the event I get into a wreck or find myself with low/no oil pressure the fuel pump shuts off just like the stock pump would have.

 

Sorry, for the short answer. I had about a minute last night to reply. I didn't see that you had the check valve in the pressure line. My fuel system never had the check valve. I had the pump prime time at 20 seconds and never had a problem.

Sounds like you've figured out your problem. The only other thing that I might look at is whether you might have an injector thats leaking.

 

RedC4,
Fuel psi drops instantaneously, this rules out the fairly new 83 lb/hr Siemens injectors. As mentioned prior, I am running a swing check valve that has a spring load to pull shut, plus gravity assist. The only way that I can see all of this happening is that the o-ring seal on the swing check may have failed. I plan to pull that apart and check the internals. Probably this weekend. The suction side of the fuel pump is located ~1/4" off the floor of the tank, with a taper cut on the inlet parallel to the floor. I have observed the pump cavitating, and it is not doing so with ~10 gals of fuel in the tank.

Tony,
My pump harness is wired through the factory safety switches. It will likewise shut down if oil psi drops. The pump does run, but the psi drops as soon as the fp times out.

SloR,
If I rewired my pump around the factory "time out", then I could prime for 20 secs, and it probably would start. In this case, I like the time out feature, as it serves to protect the motor in the event of loss of oil psi. I think I have narrowed the problem down to one or two items (check valve or regulator), with the issue seeming to point to slow failure of the check valve. I think the check valve finally failed completely, and that is what I am experiencing now.

All,
Anyone have a good suggestion as to where I can get a -10AN (or -12AN that I currently have installed) check valve? I will need this if the existing valve has failed. The existing valve is off of the wing of an A4 fighter jet ("swing check valve" so that when the plane banks the fuel doesn't just rush to the lower wing), and as such, I do not think that I will be able to get another.

Thanks for all of the great comments,
Aaron

 

Aaron, wire your pump directly with a bypass switch and see if you maintain pressure. I had a bad pressure sensor switch which caused my car to die right after starting. BTW, all of the experts will tell you that if your stock ECU does not see the HEI crank reference the ECU will go into limp home mode and will not operate your gauges. I converted my FAST box over to SEFI and a magnetic pickup input and my guages still work perfectly, as well as the pressure saftey sensors.

 

I pulled The swing check valve, and everything is fine. The seal is not a neoprene or Buna N type o-ring, but instead a teflon like c shaped seal. It was very soft, and sealed very well if the hinge of the check valve was located at the top.

I reinstalled the swing check ensuring that the hinge was directly at the 12 o'clock position. I reinstalled all of the fuel rails, injectors, regulator, ect. First hit of the ignition, it went to 42 psi and held. The car cranked right up. I let it shut down twice more and bleed off the psi and exactly the same easy prime and start.

The only thing I can figure is that the swing check was in possibly the 2 o'clock position before I began messing with the new feed from the bottom of the tank. I guess when I re-installed all of the plumbing after the bottom tank feed, I must have gotten the swing check installed in the 4-8 o'clock position, ensuring that the swing did not close.

Life is good. Thanks for all of you guys help.

Aaron