stereomandan

Hi all. It's been a while.

Been having a starting issue with my '97. It doesn't want to start on the first attempt. Cranks fine every time, just no start. If I stop for a second or two and try again it will fire right up. Been like this for a few months. Once in a while it will start on first attempt, but vast majority requires two tries.

Changed fuel filter since it's an easy attempt. No dice.

Thinking fuel pump. Doesn't build enough pressure at first try, but by the time I've let off and restart there is enough pressure at the injectors? IDK? Anyone have this same experience? If so, what was your diagnosis?

Thanks!!!

Dan

Injuneer

Most common problem is rapid loss of fuel pressure after the pump primes for 2 seconds, then shuts off. One way to check to see if this is the problem is to cycle the ignition switchback and forth from "off" to "run" quickly, several times in a row, before you try and start it. If it starts quicker, you may have found the problem.

Check the fuel pressure with a gauge. Turn key to run and watch the fuel pressure. When the pump stops priming you should have at least 40psi. The pressure should drop slowly. If it drops rapidly (like 10psi in a few seconds) you may have:

-leaking check valve in the fuel pump
-leaking fuel pressure regulator
-leaking injector(s)
-leaking fuel lines

If it's the check valve in the pump, it only affects start time, not the way the engine runs.

stereomandan

Thank you!! I'll bust out my pressure gauge and see what I find out. Will report back.

BTW, the car runs fine once it starts. I have to datalog it to completely verify that, but there seems to be plenty of fuel while driving.

Dan

Injuneer

As noted, if it's a problem with the check valve in the fuel pump, it only extends the cranking time. Once the engine starts there is no effect. A leak in the fuel lines may or may not affect the way the engine runs. Leaking FPR or injectors will affect the way it runs.

stereomandan

There is zero gas smell, dripping, or noticeable leaking.

Doesn't seem to be an injector(s). It doesn't start rough or funny, and as soon as it fires up it's smooth.

Re. the fuel pressure regulator. If it was leaking, would it leak internally in the fuel system, or are you saying it would leak external? i.e. gas smell, dripping, etc...

Thanks again Injuneer.

Dan

Injuneer

Typically, the diaphragm ruptures, and fuel flows into the vacuum compensation line, which dumps it into the intake manifold. Pull the line of and see if it's wet with fuel. But it could also be the regulator not closing down fully, allowing some fuel to leak into the return line.

Did you try cycling the key several times before turning it to "start"?

stereomandan

O.k., so here’s some more info.

Hooked up the fuel pressure gauge last night.

First, priming it. Yes, it bleeds down fairly quickly, like 10 psi every few seconds. Second, it doesn’t want to prime much above 30 psi no matter what I do. I did get it to hit close to 40 psi once, but seems like I got lucky.

Second, idle. At idle, it stays at roughly 30 psi. If I pull the vacuum line off the back of the fuel pressure regulator, it pegs to 41 psi instantly. I didn’t notice fuel in the vacuum line, but hard to see or smell all that great back there.

Third, driving. Cruising at 1500-2000 rpm, it stays at roughly 32 psi. When I do a partial throttle acceleration (25-50%), it goes up to roughly 38-40 psi. At WOT ( trying to look at the pressure gauge and not drive off the road) it looked like it was staying in the mid 30’s.

So, is it sounding more like the FPR? Thanks.

Dan

Injuneer

Sounds like the fuel pump is a bit weak. You do have mixed results in the measurements.

It should prime to at least 40psi.

At idle, vacuum line off, should be 43.5psi, with GM spec accepting anything from 41-47psi.

When you attach the vacuum line, the pressure should drop about 8psi with the stock cam. So you are looking at no less than 33psi. And that is also the pressure you should see at steady state cruise conditions, throttle held steady.

On the road, the pressure will fluctuate with throttle movement, because throttle movement affects intake manifold vacuum. Open the throttle and the pressure should rise close to the "no vacuum" reading. Close the throttle, and the pressure can drop to 30psi or even below, since manifold vacuum reaches it's peak during closed throttle decel.

Your pressure regulator sounds like it is responding to the vacuum correctly, and trying to hold the correct pressure. The odd thing is the 30psi at pump prime.

stereomandan

Thank again. Ok, I get why the fuel pressure rises with partial throttle acceleration due to less intake vacuum.

So at WOT, even with the throttle blades fully open, the intake must start to pull more vacuum the higher and higher the RPM's climb right?. (Since it's trying to pull more and more air through a defined intake size. ) This would cause fuel pressure to drop as RPM's climb.

Edit: Ignore the following question. I did some research and found my answer... This might be a stupid question, but why have the intake vacuum affect fuel pressure? Why not just leave the vacuum line disconnected and run at constant 40+psi

Dan

Injuneer

Because it would screw up the calibration of the fuel injectors in the PCM.

The flow through the fuel injector is controlled by the DIFFERENCE between the injector inlet (fuel rail) pressure and the injector outlet (intake manifold runner) pressure. The injector is basically an orifice, and the flow is proportional to the square root of the DIFFERENTIAL pressure across the injector.

The LT1 injector is flow rated at a 43.5psi DIFFERENTIAL pressure. That flow rate is programmed into the PCM. If you held 43.5psi in the rail, and the intake manifold is -10psi (~20"Hg vacuum) at idle, the DIFFERENTIAL pressure is +43.5psi - (-10psi) = 50psi. The fuel flow would increase by 10.9% above the flow at 43.5psi DIFFERENTIAL, and the engine would run rich.

So, if manifold pressure is -10psi (gauge, referenced to atmospheric pressure) and the fuel rail pressure is 43.5psi, the vacuum compensation line has to reduce the rail pressure to 33.5psi, so that the DIFFERENTIAL stays at 33.5psi - (-10psi) = 43.5psi. If the math is confusing, remember that a negative of a negative is a positive.

To answer your other question, at WOT with low air flow, manifold vacuum will be very close to "0", so rail pressure stays near 43.5psi. As flow increases, the pressure LOSS through the upstream components in the air ducting (filter, duct, MAF sensor, elbow, throttle body) all start to increase, and the vacuum level in the intake manifold increases. If you watch MAP (manifold absolute pressure) as air flow increases, MAP drops (which means vacuum is increasing). Hopefully you will not see more than about 1psi loss through the intake track at max RPM/air flow.

Now to add to the confusion, you do NOT have to have vacuum compensated fuel pressure. The LS1 engine has the fuel pressure regulator back at the tank, and the fuel supply to the rails is "dead end".... there is no return line. So, the rail pressure on an LS1 is a constant 58psi (system runs at a higher pressure than the LT1). The changes in fuel flow through the injector as vacuum in the manifold changes has to be dealt with in the programming.

I run my fuel system at 58psi, with 78#/Hr low impedance injectors. My system is not vacuum compensated. But I don't use the stock computer. The system is part of the MoTeC M48Pro engine management system, and the fuel maps have been programmed using numerous engine dyno runs, and verified by chassis dyno and track tuning. I also run speed-density, so it's even more important to develop the volumetric efficiency (VE) tables on the dyno. And I don't do this.... the guys that built and tune the engine do it. But I like to watch

stereomandan

Great info Injuneer.

Understand fully, and makes total sense now that I understand why the FPR is vacuum compensated.

I have both a new fuel pump, and FPR on hand to swap out in the near future.

Thanks.

Dan

stereomandan

Turns out it was the fuel pump. Dang that was a major PITA to replace. Pressures are now spot on where they should be.

Only problem is now the fuel level seems to be stuck at about 1/8 tank. Maddening. Hopefully that clears up. I haven't filled or driven since I put the new pump in (lousy timing, just got a drywall screw in my sidewall!), but I know the old pump was reading more than 1/8 tank in there when I replaced the pump. It was reading closer to 3/8 tank. >

Dan

stereomandan

Scratch that. The fuel level is working, but it is about a quarter of a tank lower than what I am used to. I just filled up but it looks like it doesn't even register barely to the full mark, whereas before it would be almost a quarter of a tank past full. I wonder if they recalibrated how the fuel level indicator works on the newer Delphi fuel pumps.

Weird, but at least it's working. 😀

The car feels like it's running excellent though. Thanks again for the help

Dan

Injuneer

Did you replace ONLY the pump, or did you buy a complete sending unit - pump, bracket, bucket, level sensor, wiring, etc.?

Is it possible you bent the arm for the level float? You can check the level sensor with an ohm meter. Shoebox has the procedure on his "how to" page. Resistance can be measured at the harness connector over the rear axle.

stereomandan

I replaced the entire assembly with the Delphi OEM assembly.

I'll check the ohms. Thanks for the suggestion.

Pretty sure I didn't bend anything, including the float arm. Went in easy (much easier than taking the old one out) There is a pivot point in the assembly with a spring on it. You need to bend it at that point to get the pump into the tank. Only think I can think of is that it didn't spring back fully causing the pump to sit a little higher in the tank than normal or something.

I do notice that the fuel level seems to be more sensitive to accelerating and braking than before. I used to only really notice it when the tank was low, but now I notice it even when the tank is almost full. I can get the fuel gauge to drop by 1/4 tank by braking.

Dan