mistermike

A few days ago, the engine goes into reduced power mode. Codes are P0087 and P0089. If the car idles for a while, the CEL will go out, but will return and drop into reduced power with the slightest bit of throttle.

Scanner reveals a fuel pressure reading of about 70 PSI at idle. This should be something in the 1400 PSI range. I would think that the engine wouldn't even run with the rail at 70, but obviously it does, casting some doubt on the validity of the 70 PSI number. Interestingly, at key on, engine off, the rail pressure slowly climbs up to a few hundred PSI. Alldata says look at fuel pressure sensor, HP fuel pump, or injectors for these codes. There is no evidence of any injector related problems, which aren't too hard to spot on a 4 cylinder, and the goofy readings on the scan tool superficially seemed to point to a bad sensor. Normally, I don't like to throw parts at problems, but $30 for a sensor isn't a budget buster, so that's what I did. Beats the heck out of throwing a HPFP at it.

New sensor, same exact behavior, but with slightly different values at key on, engine off. I'm not counting on a sensor designed to read pressure in the thousands being terribly accurate at the extreme low end of the scale, so I'm content to accept that the problem is not sensor related. The only thing I can measure on the pump itself is the resistance of the regulator solenoid, and it showed about half an ohm, which is within the 0.3 to 0.6 ohm acceptable range.

This would seem to leave:

1. Mechanical issue with the HP pump itself
2. Wiring issue between the ECM and the pump regulator connector
3. Issue with the 12V or ground side driver transistors in the ECM itself
4. Software issue

There are no nasty noises coming from the pump, as is sometimes the case with HPFP failures, but definitely not a conclusive diagnosis. The pump has an internal pressure relief valve, which could possibly be stuck open. Hard to say, as the HPFP is not serviceable outside of unit replacement.



I plan to check continuity between the regulator solenoid connector and Pins 5 and 73 on the ECM connector. If that checks good, I can tap into the harness and put a scope between those pins, allowing me to measure voltage and see if the PWM pulses being sent to the solenoid are present.


My only knowledge of TSB's is derived from a 2011 version of Alldata, which shows nothing, but is probably far from up to date. Nonetheless, vigorous web searches have not come up with any TSB's related to software updates related to HPFP operation on LNF vehicles. Again, not conclusive, but I don't have access to GMSI.


Any comments from people who have gone a similar route, or have up to date TSB data would be greatly appreciated. I'd like to explore the options that don't include throwing a $400 fuel pump at the problem, before going down that road.


Thanks in advance.

firemangeorge

2 things. On my cars RPD,idle pressure generally reads about 1700-1800 on a cold start, then drops to about 400 after a warm up of 1-4 minutes. Don't expect 1400 after a warm up.
Key on,engine off. The pressure you see should be from the fuel tank pump.

Other than that, sorry I don't have an answer.

JAXFL_SS

When my HPFP went out, the car wouldn't run for more than about 5 seconds. I had it replaced under warranty.

mistermike

Update. Checked wiring harness from ECM connector to fuel pump solenoid terminals, and it is fine.

firemangeorge

Hmmm. No comment on my post above?

mistermike

New fuel pump seems to have fixed the problem. I have about 250 PSI at warm idle and over 2200 PSI when I stand on it. The electrical part of the regulator solenoid in the pump tested good, and there was no leak from the piston side into the engine (another common failure), cam follower didn't show undue wear. I would suspect, at this point, that the spring that the regulator solenoid works against probably broke. It would be nice if I could get it apart and fix the spring. Doesn't hurt to have a spare lying around.

firemangeorge

And still no comment on my posts in trying to help out "mistermike". Why do I even bother.

mistermike

Sorry about that. Alzheimer's has set in. I read the pressure on the inlet side with a gauge, and of course it reads about 70 PSI. The odd behavior was that with the bad pump, the rail pressure would read 70 PSI running. Theoretically, that's not even enough pressure to overcome the 150 PSI or so you would see in the combustion chamber at the end of the compression stroke, yet the engine ran. Shutting off the engine, but leaving key on, the rail pressure would slowly climb to about 300. Really made no sense at all. I will do a complete readout of the pressure behavior tomorrow with the good pump, as my back is all screwed up from a pinched nerve or something.

firemangeorge

I'm not really up on my direct injection specs. So, do the injectors really pulse at TDC(your 150 psi reference) or do they pulse on the intake stroke to allow the fuel to mix with the air?

mistermike

They fire on the compression stroke. The very fine spray and high temperatures promote rapid mixing, along with the "swirl" produced by modern piston/head geometry. The exact timing point is variable. It's not really at TDC, because it has to occur before the spark. This might explain why the engine still ran at light throttle with wimpy fuel pressure, whereas the ECM would not permit very high load levels, where the spark and injector timing would have much less advance, and be forced to inject much closer to maximum compression. This is a big advantage for forced induction and high compression engines, as injecting closer to the ignition point reduces the likelihood of detonation and preignition. The downside is you need a boatload of fuel pressure, and the injectors and fuel system cost a lot more than injectors designed for 60-70 PSI.