The saga of the lost fuel pressure continues.
The exhaust shop had finished with my Suburban by about 1 p.m. on Monday (05-02) at which time I picked it up. The drive home was much quieter but still sluggish and full of backfiring. The computer had logged a “lean fuel condition” code when I checked it later. All signs seemed to be pointing to a fuel delivery problem.
First let's check out some of that shiny new exhaust system.
2.5” all the way back with twin Dynaflow non-chambered 3” in/out mufflers (must have been out of the 2.5” style that day) and chrome tips that exit just behind the rear tires. I could have saved some cash had I gone with the shop owner's suggestion and done a single 3” system but I requested the more traditional look. Once home I could finally hear that there was an exhaust leak which lead to a torquing of the headers. I also pulled the fuel fill hose back off to weld shut the one pinhole in it. Now the fill cap has a 'whoosh' when it is unscrewed because the system can hold pressure properly.
On Wednesday morning (05-04) the parts arrived to replace those kinked fuel lines which I hoped would alleviate some of the driveability symptoms I'd experienced. 20' of -6 braided stainless over synthetic nitrile with hose ends and adapters for the metric o-ring fittings the TBI uses.
Aaaaand a couple close-ups of the old fuel feed line:
This was the first time I'd assembled my own hose with re-usable ends. Fairly simple and straightforward if done properly.
After tightening the lines and a throwing on a new WIX fuel filter I also removed the gas pedal and bent it an inch or two. Post modification the TPS now read 0.88V closed and nearly 5V with the pedal to the floor. We'd found full throttle. Then, after recalibrating the transmission controller to accept the new TPS reading it was time for a test drive. The same symptoms were still present. Back to the manuals and research. The Fuel & Emissions supplement has some great troubleshooting charts. All signs seemed to point to low fuel pressure. The only gauge I had that registered fuel pressure in the needed p.s.i. range was an old vacuum gauge with a very small orifice. Before buying a new ornament for my gauge drawer I decided to drop the tank to inspect the fuel pump and it's attached hose instead. I should probably mention that just last year I replaced a bad in tank fuel hose on my DD, so I'd been down that road before.
The tank and pump on the Suburban had come from a ‘87 - ‘91 Blazer w/ EFI. The other pump / sending unit assembly sitting in my parts cache came from the '88 C3500 (I scrapped the tank some time ago). Speculating that perhaps the Blazer pump wasn't cutting the mustard any more, I swapped it out for the truck unit and slapped the tank back in place. When the pump didn't engage I had to remove the tank once more to diagnose a loose wiring connection. Now that everything was back in place, the engine started from cold without needing an extra boost of starter fluid, but it still backfired and wouldn't run right on the first test drive. Pooh. Soon thereafter I ordered a fuel pressure gauge and tee fitting. Section A-4 of the Fuel & Emissions supplement has a flowchart with fuel pressure diagnosis instructions which I followed precisely (pages 3-34 to 3-35 for those of you playing the home game). 9 p.s.i. is the minimum pressure the engine needs to run right.
8.5 p.s.i. was the reading on the gauge with the key turned to the 'on' position (pump runs for 2 seconds and then turns off which causes fuel pressure to return to 0 p.s.i.)
11.5 p.s.i. was the reading with the return line blocked and the pump wired to run constantly
Neither of those reading were really abnormal according to the manual. Fuel pressure should have been able to climb straight up between 13 and 18 p.s.i. with the feed line deadheaded (max pressure the pump should be able to output). With the return line deadheaded instead (fuel is able to flow through the pressure regulator) the pressure should not exceed 13 p.s.i. and if it did, then the regulator needed to be replaced. Mine was 11.5 p.s.i. which wasn't above the max. Still confused I started the engine which gave me the most worrisome reading of all, 6.5 p.s.i. at idle. It seemed to me that if the engine wasn't getting enough fuel, then it must have been a problem with the fuel pump not flowing enough. Well, that or a faulty fuel pressure regulator not letting enough pressure build up. Crap, which was it?
Turns out it probably wasn't the fuel pump. I took a gamble and ordered a nice, new Walbro 190 l/hr unit, dropped the tank once more (third time's the charm?) and installed it. That only raised fuel pressure by .5 p.s.i. at idle. Pooh again. Still, it ran much better on the test drive around the neighborhood. Now it was time to investigate the fuel pressure regulator. Instead of messing with trying to rebuild the stock unit I ordered a JET Performance adjustable FPR. I'd read that once one increases the airflow in these engines that the fuel pressure usually needed to be raised a smidgen in order to keep the A/F ratio from leaning out too much under WOT conditions when the computer operates in open loop mode.
With time to kill between bolting on parts that didn't entirely solve existing problems and ordering new parts in the hopes that they would, I tried my hand at a few other small projects. Cut a couple new holes in the dash for those side A/C vents.
Picked up the coolant recovery tank and cap I'd ordered from the Chevrolet dealership in town. I'd told the parts guy that I wanted these pieces for a 2WD '93 Astro van.
Sure, I could have traipsed around a few wrecking yards looking for junked Astro vans, but this way I got brand new parts, not 20 year old plastic with an unknown history. This also offered a much cleaner and more hidden install compared with trying to reuse the bulky '88 C3500 tank I still had in my possession.
I tried forming the needed bends in the hose using an approach I'd read about on another forum. The instructions said to run a wire through the hose then form it to the needed curves and angles. After that it was supposed to spend some time in the oven at around 400°. That didn't go over so well because the hose started emitting a faint smoke with a foul odor once the oven got up to temp. I took it outside immediately and doused the whole thing in cold water from the garden hose. It was a calm day so the house took a while to de-stinkify. The hose did still retain most of it's shape after I removed the wire though.
I also got around to tackling the transfer case shifter. It was a toss up between repairing a broken GM NP241 shifter or adapting a Dodge transfer case shifter I had on hand.
I liked the shallow packaging of the GM shifter so I decided to disassemble it and weld the broken bits back together.
The transmission hump had to be modified slightly. I offset the shifter to the passenger side in order to clear the transmission itself.
After a little trimming and a few minor adjustments it could engage all the gear positions.
It seems that the shift pattern on the stick, from front to rear is 2HI – 4HI – N – 4LO.
I killed some more time by cutting a leftover scrap of pine shelving to fill the hole where the old gas fill tube passed in the rear.
Then, yesterday, the adjustable FPR arrived.
Old (L) vs. New (R).
It's a very straightforward install, but the instructions were still quite spartan. They didn't bother to mention anything about which way to turn the Allen screw to raise or lower the pressure. I simply guesstimated and set it right in the middle of the adjustment range. After replacing the old gaskets with the ones supplied in the kit and tightening all of the little Torx screws I fired the engine up. Fuel pressure was immediately holding steady at 11.5 p.s.i. That actually caused the engine to run too rich during open loop startup mode and I had a few drips coming out the tailpipes. Once it warmed up and the computer took control of air/fuel mixture duties I took it out for a test drive. It immediately felt and sounded different. Puttering around the neighborhood below 25 mph didn't cause it to bog or backfire anymore. Getting out on the highway it seemed to do much better. But, at certain spots in the gas pedal's throw, it still bucked and popped, just not as much as it had before. I came back to the garage and grabbed my scanner. With it hooked up I tried a WOT run getting on the highway. The O2 sensor registered between 900 and 950 mV, so I know that it wasn't running lean (below 800 mV).
When I got back home and shut the engine off, I tried starting it again to see if the hot start problem was still present. A dilemma arose when the engine didn't crank. I'd guessed at the problem the other day and finally confirmed it, the alternator wasn't charging with the engine running. I'd been driving around on battery voltage only this whole time and it was now down to about 11.1 V with the engine running. That simply wouldn't do. My initial conjecture was that the alternator was a goner. I'd even gone as far as eyeballing a 140 amp Powermaster to replace it. After that I'd kept reading and researching alternator function and came across this line on page 6D3-2 of the FSM, “Either the “L” or “I” terminal (or both) is used to turn the regulator on and allow field current to flow when the switch is closed. The “L” terminal must be connected through an indicator lamp or suitable resistor.”
I'd only had the brown wire from the alternator dangling about, not connected to anything. Pulling out the wiring schematics involving the charge warning light I was able to identify the components involved. In it's stock configuration, the brown wire simply connected straight to the instrument cluster circuit board where all the appropriate connections were made, including one with a resistor.
That's the board, and in the upper left hand corner is the resistor that I pulled off of it to wire inline. The temporary wiring I crimped together in 5 minutes (two wires, a bulb and the resistor) solved the problem and the alternator started charging. Having a fully functioning charging system didn't fix the backfiring (of course), so there's still some investigation to be done.