TBI Swap - all gremlins fixed

[AZ79K5Project]

I've completed my TBI swap, passed the inspection and sniff test at emissions.

The good:
Starts cold with no problem
Steady idle 99% of the time
IAC numbers look good
Passed emissions with real low numbers
nice throttle response (most of the time)

What's been done on the swap:
All new sensors on the TBI swap (TPS/Temp/ESC/MAP/4 wire O2/ speed sensor/IAC/knock)
5.7
Edelbrock 3701 performer intake with TBI adapter
Spacer to get above EGR valve
Rebuilt the TBI unit
7747 ECU
New fuel tank with in - tank fuel pump (12 psi before the filter up to the TBI unit.
New fuel lines
New coil / new distributor
Timing set to 0 with connector unplugged

I've been chasing a ghost that I think is related to the injectors, but I am totally lost.

Now the problem:
Intermittent loss of power / no acceleration and just plain flat

I'll be driving and the truck goes flat, no power, barely maintains 45 MPH, unable to accelerate. Unable to accelerate from a stop. I though the problem was EGR related, so I blocked the 3701's EGR port with no change. Then just as suddenly as the flat spot, full acceleration with no hesitation.

Attached is the ProTuner RT scan file that I took during one the flat runs.

The passenger side injector is not pulsing as fast and as well as the driver's side. The pattern is not as good as the driver's side. I switched the connectors on top with no change, swapped the injectors with the same result on the passenger side.

I've searched, I've read and I'm coming up blank.

The attachments have .txt at the end to post them up. They are .xdl fils.

 

[4X4HIGH]

Does the vehicle have CATS? If so one or both could be coming apart and creating a blockage which will affect power output.

 

[AZ79K5Project]

Dual cats. About 7-8 years old with less than 10k miles on them.

 

[dyeager535]

I can't see the logs, but with a noticeable spray issue that is not specific to one injector, I'd be leaning towards something in the injector pod causing an issue, or ECM.

Rebuilt TBI unit, are all the filters/strainers/o-rings/seals in known good shape? Have a spare TBI to swap? Other than an electrical issue, something affecting that side of the injector pod is the only thing I can think of that would be specific to one side and not the other.

Do you have a spare ECM you can swap? I don't have the '7747 schematic handy, but I know the ECM *can* have issues related to injector pulse if things go wonky in it. They are pretty tough units, but things do happen. I wouldn't buy another ECM as a guess, but if you have a spare, swap the ECM, see if the problem goes away. Even then it would seem odd for a wire not to just completely short to ground, or never ground, as opposed to what you are seeing, which would make me lean more towards something in the injector pod.

 

[AZ79K5Project]

I did some closer inspection with better lighting and saw that the patterns looked good on both injectors.

I was concerned about the ECM as well. I did buy an ECM from an 87 that closely matches my setup from EBay this morning. I also picked up a TBI performer intake from CL last night. I'm going to modify the center bolt holes and see what happens.

I did some work and making sure there were no vacuum leaks. I hooked up the EGR and had codes 32 an 44 thrown when driving.

What am I missing?

 

[dyeager535]

What are codes 32 and 44?

As I understand it though, this "flat spot" happens irrespective of EGR operation, correct?

 

[AZ79K5Project]

Yes - flat spot occurs regardless of EGR. Blocking plate - no real change before I did some work this weekend. EGR back in place - I got the error codes. I didn't check things after adjustments with EGR still blocked off. I may block it off again to see if I solved any potential vacuum leaks.

I think I found a split hose going to the transmission valve. It broke in two when I was checking, so I'm not sure what it was like.

32 - EGR open command did not change MAP sensor signal. Normal EGR flow should cause slight decrease in manifold vacuum which would change MAP sensor signal.

44 - Oxygen sensor voltage was under 0.2 volts for 50 seconds of closed loop operation. (Lean Exhaust Indicated)

 

[dyeager535]

Lean is potentially from vacuum leak of course, so I'd fix that (or at least plug the port), clear codes, see what happens.

 

[AZ79K5Project]

Did a lot of work on Frankenstein this weekend. Happened to find a new TBI Performer intake on CL for a good price. Modified the center bolts and bolted it on. Probably the worst thing to do when troubleshooting, but I couldn't pass up the deal.

Dropped everything on, made aluminum block spacers for the center four bolts, had to keep the TBI spacer because of the -6an fuel lines and EGR. The thing fired right up, adjusted the timing. Seems to idle better with 4 degrees of timing. Replaced the ECM while I was at it.

Went for a test drive. Acceleration was great, but then went flat momentarily. Turned it off, nothing obvious found and then I went for a drive. No problems, until I hit the freeway. 20 miles and it went flat, much less impact than before. Only happened when the the EGR should have kicked in according to the specs I looked at. Happened a few times on level ground with a steady speed. I could feel the power drop off when it happened. Continued off the freeway on surface streets, stop and go for 7 more miles no problems.

Pulled back into my neighborhood and then a complete flat response. No acceleration, backfiring through intake and total crap. Turned the truck off and back on, whatever was going on - gone, full power again.

I hate throwing parts at it. Is there a way to test the EGR solenoid? Could that be an issue? What can cause this to essentially shut down and clear just as fast?

 

[dyeager535]

Again the '7747 stuff Mark knows better than I, but I'm pretty sure the ECM checks for proper EGR operation by measuring manifold vacuum...if it doesn't decrease when the valve opens, it should trigger the CEL.

AFAIK when the EGR "malfunctions", it doesn't affect how the engine runs, even with the CEL. I would try and unplug the solenoid. If it's a problem with the EGR opening, that should make a difference. In your case, when you find the culprit, sounds like you will immediately know.

Running on flat ground, steady speed, EGR is the first thing that comes to mind that would cause a flat spot. I suppose if the knock sensor for some reason saw knock and retarded timing you'd feel it too, but that is less probable than EGR.

 

[AZ79K5Project]

The EGR should open on flat ground (no real load), above 25% throttle and one other criteria. I don't know if there is a sticking solenoid or problem there. I'm trying to find out how to check that specific system besides just disconnecting. If I disconnect it and my problems stop, at least I will know that the problem has been isolated.

I believe the knock sensor is working based on the winaldl / Tunerpro RT logging.

I've checked the voltages at the ECM and all the voltages appear in line. As I am thinking right now, I had one voltage that I am going to double check since I was getting a voltage and it should have been 0 if I remember correctly. I'm going to recheck all the wiring and double check the voltages and grounds at the sensors.

Question of the 7747 wiring. I've read a bunch of sites and a lot of people are taking wires to the block instead of the ECM. One site states A11, A12, D1, D2 and D6 can all go to ground/block. One wire of the MAP (pin A) goes to D2 and one wire of the Throttle Position Sensor (pin B) goes to A11. D1, D6 and A12 are actually grounds noted on the schematic.

 

[dyeager535]

I saw it mentioned by someone I felt knew what they were talking about that some of the grounds need to be "clean". As in if multiple things are grounded to the same point, it may cause some issues.

I'm not sure why exactly some of those things would be grounded through the ECM. It may be that way they would all have the same resistance? That (generally) might be a good question for the general forum. Fordum and Rampage(?) are the two that come to mind who might have an idea why one way might be better than the other.

Honestly though, gut feeling, a minor ground issue would not be something I'd even consider in your case.

As I recall with the 7747 and TunerPro, there is nothing telling you when EGR is commanded in the datalog? If there was, you should be able to see the MAP change when EGR was called for. Obviously a shortcoming of the 7747, due to the baud rate that a lot of time, relatively, elapses between events, so it would be harder to know if TPS and MAP were stable before seeing a MAP change when EGR might be commanded. On some of the other ECM's, probably most/all of the "faster" ones, the datalogs show when EGR is commanded, and would then make it easy to see what the MAP result was.

 

[AZ79K5Project]

ECM Overheating

I've been chasing this gremlin for a few weeks. Tonight I discovered the ECM is getting hot. As the ECM heats up, the performance drops off. After a drive, it started dogging down and having all the issues from the first post, I couldn't even drive through the parking lot and the ECM starting throwing codes. I discovered the ECM was what I would consider hot. I immediately swapped ECM and with 35's, locked rear end, squealed the tires. The 2nd ECM started warming up, and performance dropped off.

All the voltages at the pins are all in spec. What would cause the ECM's to overheat?

Looking back, each time I made an adjustment or change, likely provided time for the ECM to cool down between test drives.

 

[dyeager535]

Since you can access it, did you take/leave the PROM cover off to test, see if a bit of venting helps?

I can't think of anything that would generate excessive heat, but I don't recall ever checking to see how warm it was running either.

Do any of the wires heat up? Obviously not the ECM that is the problem if two of them do the same thing.

 

[AZ79K5Project]

Thanks for the feedback. I went through everything today, hooked it back up to the computer and found a bad 02. I replaced the 4 wire with a one wire and things smoothed out - ECM nice and cool - well as cool as any Arizona summer day.

Still chasing a lean issue. I did some logging today after everything was back together. I blocked off the EGR to remove that from the equation. The BLM readings are going crazy lean. At least that's what doing a lot of searching told me. I attached a couple of screen shots of the readings. The BLM will stay in the middle and then go lean for no apparent reason. The 169 and 172 were the highest. The 169 occurred while on constant throttle and constant speed. I think that was the time it started going a little flat while driving.

I am going to get a fuel pressure gauge from AZone on Friday and see if I can get it in a place where I can watch the fuel pressure as I drive.

 

[dyeager535]

By "in the middle" you mean BLM 128? 128 is the "ideal" for low-load cruising. If it's reaching 128, something else is going on methinks. You'll see BLM move as you change the load on the engine, obviously you don't want it maxing out though.

 

[AZ79K5Project]

BLM would momentarily max out in the 160's while cruising at a constant speed. It would then correct and things quickly smoothed back out.

 

[dyeager535]

Whats up with your battery voltage? I assume the '747 has a table built in for voltage offsets, but 15V is pretty high. Any correlation?

 

[AZ79K5Project]

I don't think the battery voltage is the issue. Everyhing I have read indicates that I am within specs. I have a 110A alternator in the thing on a serpentine belt.

I think I may have found my problem. The regulator was having problems and I replaced that with an adjustable. It was maintaining good pressure. I was watching the pressure and it dropped a few PSI. The engine sputtered and died. The pump was running, but I had low pressure. Clamped the return line and no increase in pressure.

The gas tank was a little banged up from shipping. I straightened the damaged areas. I am thinking the thing was bent or tweaked enough that the pickup area is limiting flow. Pump has already been replaced under warranty and the problem stayed. The tank is the only common thing. Everything else has been tested, checked and triple checked.

I read that the pickup screen has to be certain direction (front to back or side to side - don't remember). Any truth to that? When I replaced the pump, the sputtering was a little worse but I didn't think much of it. Can the straps be too tight and cause a restriction around the pickup?

 

[dyeager535]

I would have to look at the instructions on the pump I just installed. I CAN say that the old filter and new are completely different designs. No idea why the filter would need to be in one direction or another, unless it hit the sump or something in one direction.

What kind of tank? I can't imagine a way that the baffles would restrict flow from a few dents, but I could be wrong. I noticed my aftermarket tank only lets fuel in through two holes in the bottom of the sump (horrible design, the sump might as well be decorative) I suppose in my tank, if the tank was pushed in at both holes, it could suck the pickup dry. Then again, a full tank would completely eliminate any problem with sucking the pickup dry.