i2bme

Hello All,
<?xml:namespace prefix = o ns = "urn:schemas-microsoft-comfficeffice" /><o> M</o>y daughter's 2004 Tahoe, 4.8L, 178K miles, was idling andrunning very rough and stalling often. She was told that she needed to replace the coils, plugs & wireswhich she did, but it did not help.

<o></o>I towed it home to try to fix it. It was showing a P0300 code: random misfire.
<o></o>Compression test showed 2 cylinders with 160psi, 2 with 190& the rest 200-215. All the plugswere white except #2 which was all black (it also was on the highest ratedcompression @ 215psi). All the coilswere tested for spark with the respective plug wire attached to a sparktestor set for a .040 gap plug (thetestor's free air gap was over 1" at that setting). All the new coils & wires provedfine. The fuel pressure tested at 55psi.

<o></o>This vehicle has no external fuel filter, no external fuelpressure regulator, no idle control valve and no egr valve, and no mechanicallinkage to the throttle body.
I cleaned the MAF sensors which appeared clean, and thethrottle body, which was quite dirty. Iput it all back together to check results. The idle was markedly improved, but it still had a miss. I swapped the #2 spark plug, which was stillblack, with another plug, and took ashort test drive which showed no tendency to stall. I checked the #2 plug again & it was wellon the way to becoming all black, then I cleared the P0300 code.

In hindsight, I should have run a cylinder balance check,but I did not. Instead, I assummed #2injector was hanging up & running rich, so I ran a can of SeaFoam with 4ozof Lucas fuel injector cleaner directly into the fuel rail with 50psi pushingit. I figured since this was 5psi lowerthan what the internal regulator was putting out, none of the cleaner would go back into the fuel tank. This is where I made mistake #2: To be absolutelysure that all the cleaner was goingthrough the injectors, I decided to pull the fuel pump relay, plus the 20A fuelpump fuse, and gently crimp the rubber hose coming from the tank. I crimped the hose, exposed the fuses &relay, got distracted, and did not remove either. With the engine at running temperature, I ranthe cleaner though the injectors.
The miss was unchanged and the P0300 was back. So, I pulled the 2 fuel rails off with theinjectors, filled my injector cleaner tool with gasoline, set it to 60psi andlooked to see if #2 injector was leaking with the engine turned off. The injector was dry, as were the other7. I then ran the rest of the gasthrough by cranking the enging while observing the spray pattern of all 8injectors. They all appeared to bedispersing in an even pattern.

Running out of ideas, I decided to measure the output ofeach injector over the same given amount of time. I removed all the plugs, disconnected thecoils, released the clamped fuel input from the gas tank, and went to reinstallthe pump's fuse & solenoid. This iswhere I realised the pump was running against the clamped hose all the while Ihad the engine running. Upset withmyself, I continued on with my attempt to measure/compare each fuel injector output,but now the fuel injectors would not spay ... the fuel pump had quit workingsomewhere in the cleaning process.
Thinking that, in working against a blocked hose, the pumpeither overheated and burned out, or it blew the 20A fuse. The fuse was fine. Just in case, I swapped the relay withanother identical one, but the pump still did not come on.

I do not have manuals to tell which relay pin goes where,but the relay has a diagram on it's side showing it's 4 pins: #30 and # 87 are the poles of its switch, soone of them must go to the pump & the other to the 12V bus. The other 2 pins (86 & 85) are shown asthe 2 leads to the solenoid coil, so one of them must go to ground & theother is activated by the computer to turn on the pump.
#30 has 12V with switch in run or off, Pin 87 had no voltage, butit showed less than 1 ohm resistance to ground, so it had to be the pin alsogoing to the fuel pump. I immediatelyapplied 12V through a fused meter to it and the pump turned on and drew 7.5amps. Therefore, the problem has to bewith the circuit that is activating the pump.

I checked out pin #85 & it was less than 2 ohms, so Iguessed it was the relay's coil side to ground, which would make pin 86 as theone with the voltage from the controler. With the car swith off, it showed 0 volts and infinity in ohms. When I turned the switch on, it showed12V. If this were true then the relaywould have to swithed on the pump, so both relays that I swapped must have beenboth bad.
I tested both relays, as per its diagram, and both had coilsthat activated when 12V was suppied at pin 86 and pin 85 was grounded, and bothlit up my test light when it was connected to pin 87 with 12V applied to itsmated pin #30 and then the coil activated. Confused as to why a pump that was proven to be good, would not workwith a proven relay, and with the relay's pins proven to have the requiredvoltage and grounding, I put my test light on pin 86 (the activating sourcewhich prior had 12V in run position and 0 Volts in off) now did not indicateanything with my light testor.

I figured that since my meter showed proper voltage, but thelight testor would not come on, that there was not enough current available toregister the light. Must be a poorconnection. But when I put my meter backon pin 86, it now showed no voltage, on or off, and has stayed that way.
The conclusion I now draw is that I have a "on andoff" connection for the voltage source which activates the fuel pumprelay. Is it just a coincidence that theproblem occured right after blocking the output of the pump? I do not know what if anything else isassociated with the relay/fuel pump and the computer controler somewhereupstream of the relay, without the schematic details. Is this problemindependent of the misfire condition or related? I got hungry, ran out of daylight and calledit a day. I am a shade tree mechanicoutside of my comfort zone. Any advisewould be most welcome. Richard

mikeatpriestlake

just throwin it out there, how about the relay for the computer?

in2pro

Welcome to the forum...
at 50 psi is it stable and does it hold when the engine is turned off?
50 is really too low, I suspect the fuel pump is your problem....

i2bme

Thanks all for feedback so far.

I have not checked computer relays, but will do that next.

The engine was putting out 55psi, I set my air pressure on my injector cleaning tool at 50psi. The engine had no noticable difference between 55 & 50psi, it ran equally rough. The engine only ran at a stable rpm after cleaning the MAF sensors and Throttle Body valve. I didn't get to the holding pressure before the pump quit, but the rails had lots of pressure released when I opened them.

i2bme

Random misfire code was due to broken #2 intake spring.

Still no clue as to why the PCM is not putting out a signal to the Fuel pump relay. Pin 9 from PCM which activates relay coil, is fine & when 12v is applied to that associated pin in the PCM connector (removed from PCM) the relay responds & pump turns on. no theft detection/security lights are showing & engine starter engages fine.

mikeatpriestlake

I would really look at the fuel pump, like in2pro said. not sure what the spec is for your vehicle, but i think 50 is low. Find out for sure, then test the pressure again. It should always read at spec. I went back and checked and edited this post. if you have a VIN Z (E-85 ethanol) then the pressure is 48-54. all others are 55-62 lbs. hope this helps