rmoore45

Hello forum members, I’m having a problem with an efi swap that I did and I need your help to solve this problem. I’m going to list everything that I have done from the beginning so bear with me. This is going to be one heck of a read.

I swapped a 1991 C2500 350 tbi fuel injection system to a 1986 C10. This is not on my truck, I did this on my friends 1986 C10 and he wants his truck running properly again. We have been fighting this on and off for about 6 weeks.

Details of the swap: Bought a 1991 C2500 700r4 auto work truck from a buddy, the truck was rusted beyond repair but the motor and trans worked. Odometer stopped working at 300k miles. Drove truck home and it ran fine and shifted fine. We did a few burnouts and general goofing off before pulling the motor. I pulled the motor and put it on a home built test stand. I then removed all the wiring from the truck, yes really, ALL THE WIRING. This was done so the truck could be hauled off and I would have everything I needed for the fuel injection. Here’s the link I used as my bible for removing excess wiring and overall conversion.

http://www.truckforum.org/forums/chevy-truck-forum/37470-tbi-swap-non-injection-vehicles-all-borrowd-diff.html

I cut away all excess wiring and eventually got it narrowed down to just a few wires

• Constant power for the computer: I used the original fuseable link that was in the 91
• Ground: all the grounds were tied into 1 main ground that is on the thermostat housing
• Switched power: 1 wire to turn on the computer that activates the fuel pump
• Starter crank signal: This wire goes to the solenoid on the starter to tell the computer that you’re trying to start the engine.
• Coil power: The white wire from the ignition coil goes to a switched ignition source
• Ac signal wire: tells the computer to bump up the idle when the ac is turned on

These are just the loose wires, all the other wires go to their appropriate sensors. I shouldn't have to say that but I did anyway. Also checked continuity from each terminal from the computer connector to its appropriate location, I did this 3 times.

After the wire diet I put the harness on the test motor from the truck that is on the test stand. I put the gas tank from the truck next to the engine and ran some fuel lines to the throttle body. Lo and behold the motor fired right up. Noisy as hell because all I have on there for an exhaust is the downpipe from the truck. The motor ran great with my modified harness, it would go into closed loop and idle perfectly.

Now that I had a complete fuel injection harness I needed to do a few other things for it to work on the 1986 C10. The motor in the truck is not original, the PO had put in a 1980-85 4 bolt main 350 block casting number 14010207. This means that the intake from the 91 would not be a direct swap so instead of swapping intakes a simple adapter plate was made from ¼ inch thick aluminum plate. I designed it so that the bores of the throttle body were over top the larger rear bores of the Q-jet iron intake on the 86. I also needed an inline fuel pump for the truck, I chose a Precision E16116 fuel pump rated for 9-22 psi. The o2 sensor needed to be mounted as well, a weld in o2 bung was purchased and welded in right after the exhaust manifold.

THE SWAP: I took my time and did it right, totaled about 5 days of work from the time the truck drove into the garage to the time it was all put back together. I won’t go into the details of doing all that work. The only thing that we could not transfer was the Egr valve. It would not work on the iron intake so we just put a block off plate over the hole for now. This should not affect how it runs.

THE PROBLEMS: Initially, it ran fine. However once it went into closed loop the idle began to surge from 900-1600 rpm. It also did not want to accelerate at all and seemed to be lacking power overall. During heavy acceleration the motor seemed to just ‘fall on its face’. A severe lack of power was present.

We soon determined that 1 problem was a vacuum leak at the throttle body, specifically the lower gasket that was for the Q-jet iron intake. One was purchased and aviation permatex was used to ensure that it didn't leak again.

The next problem was the knock sensor, the esc module from the 91 was for an automatic but we were running a knock sensor for a manual trans. After some research I found out that it does make a difference. We put the knock sensor for the automatic on and the motor ran much better when accelerating.

Exhaust leak at collector gasket, replaced both donut gaskets and put copper coat on them to ensure they didn’t leak again. This could have been throwing the O2 sensor off.

The idle was still surging from about 700-1100 rpm so we thought that it could still be a vacuum leak. While revving the motor in neutral and looking at the intake we could see exhaust leaking out of the intake at the mating surface between the head and intake. This is where the crossover port is on the intake. We figured that if its leaking there it must also be leaking vacuum. I sprayed carb cleaner all around the intake but I could not hear any change in idle. We replaced the intake gaskets anyway. The intake was milled flat and .125 inch thick gaskets were used.

After the intake gaskets were replaced a new problem developed. Once the motor would warm up and go into closed loop it would still surge at idle but this time from about 500-800 rpm and would occasionally die at idle. It did not die at idle before, only after the intake gaskets were changed did this problem pop up. I think that the vacuum leak is solved but now the unsolved issue is causing the engine to die.

After the intake was changed and it still had problems I broke out the big guns and downloaded winALDL and made a connector to go from the ObdI port to the serial port on my laptop. WinALDL is a computer software program that allows you to monitor what the engine is doing while running. You can also datalog and keep the information for later however I have not figured this feature out yet. I thought this would be the solution and point out what the problem is but it just provides raw data that you have to determine if its good or bad.

winALDL info link http://winaldl.joby.se/aldlscreen.htm

CURRENT PROBLEMS 2-11-15 : Runs very rich when in open loop, Idle surging and dieing when in closed loop, severe lack of power under acceleration.

Well that is the story so far, a not so happy ending for a supposedly simple fuel injection swap.

Here’s all the data that I have collected

Computer number : 1227747
Tps voltage at idle: .69 volts , within spec.
Fuel pressure: 40 psi, this is the pressure before the regulator. I have not found a way to check the pressure after the regulator. I’m not sure why a fuel pump rated for 9-22 psi is putting out 40 but thats what the gauge in the kit said.
Rebuilt throttle body with BWD 10912 tbi kit. All gaskets and regulator components replaced.
Code 24: VSS code. I eliminated the Vss so it will always have this code. It will not affect how it runs.
No other codes currently
Timing set at EXACTLY 0 degrees with the bypass connector unplugged.
No loose hoses to cause vacuum leaks
MAP sensor vacuum hose connects at back of throttle body like it originally did

We have replaced all these components with good BWD units purchased at O'reilly Auto Parts. Coolant temp, TPS, MAP, O2 sensor, ignition module, oil pressure, ignition coil, IAC motor, knock sensor.

The only original components are the following: Distributor and pickup coil, Esc module, egr solenoid, throttle body and injectors, computer.

Please read through this all and comment as to what your ideas as to what the problem could be. If you need specific data just ask and I’ll see if I can get it for you. I’m going to college in San Marcos and the truck is in Houston so I’ll have to contact the owner and see if I can get the data. Sorry for the super long post but I wanted to be as thorough as possible so that you have all the info.

Thanks, Ranger

Allan In NE

Hmmmm,

Fuel trim way too rich? I think you're constantly in limp mode due to that VSS issue.

Allan

rmoore45

Thanks for the ideas and suggestions. A quick Google search seems to have confirmed your comments about the vss. The vss controls the egr, idle, lean cruise conditions, and also can prevent stalling under deceleration. I was in the junkyard the other day and happened upon a 1987 c30 with a 350 and sm465. That's exactly the combination that the 86 has so I'm going to look at it and see how the vss works on with that trans. I know the speedometer is mechanical so I'm betting there is an inline vss with the correct control module somewhere on the truck.

I was thinking today and had the same thought that you said. Too much fuel pressure.

When cold, in open loop, the computer is sending an injector pulse based off a stored map. This combined with the high fuel pressure is making it run super rich (black smoke).

When the motor warms up and in closed loop it starts taking readings from the 02 sensor. These readings indicate that its running super rich so the computer leans back the a/f ratio as much as possible, however the extra fuel pressure is too much and the computer cannot compensate enough. The computer eventually just cuts fuel and the causes the motor to surge at idle and occasionally die.

That's my current theory that I have come up with today. My ideas to test this are to
-look at the spark plugs to get an indication if its running rich or lean
-look at the 02 sensor voltage on the computer to see if its running rich (.9 volts for rich, .1 volts for lean, .5 volts for 14.7 a/f )
-I'll also try to lower the fuel pressure some and see if that helps.

Allan In NE

On those old square bodies, the VSS uses a buffer at the speedometer head. A lot of vans of that period did that also.

Anytime the ECM sees a 'fault', it throws the check engine light and just plain throws fuel at the the engine just to keep it running.

You don't have to pull the plugs, you already know it's overfueling because of the black smoke.

Yeah, that's way too much pressure from the pump.

Allan

rmoore45

Did a little research this morning about the throttle body and how fuel flows through it, specifically how the fuel pressure is regulated. This picture explains it all


The regulator regulates the amount fuel being bypassed after it reaches the injectors. I did not know that. This means whatever the pressure is on the feed line is what the injectors are getting. Based on my previous fuel pressure check that means that the pressure going to injectors is about 40 psi. WAY too much pressure. Roughly 3x the amount it should be getting. Furthermore my research revealed that the fuel pump I am using is fine for this application and is not just blowing by the regulator. I have found multiple people on various forums that have used a similar pump with no problems. That leaves only 2 possible causes as to why its getting too much pressure: Faulty regulator or restriction on the return line. The regulator is brand new so I'm leaning toward the restriction somewhere along the return line. The reason I'm saying this is because I accidentally tested the fuel pressure with the return line off the throttle body. Fuel was sprayed all over the intake and it made a big mess but I seem to remember getting a pressure reading of about 10-12 psi, right where it should be.

I'm going to have the owner of the truck disconnect the return line where it connects at the frame and just have the original GM return line on the throttle body. From there just run a rubber hose to a 5 gallon gas can and check the fuel pressure.

I have a feeling that this will solve the problem. Thanks, Ranger