Newbie trying to fix drivability problem, please help!

[Wingnut87]

:help2:
Hello everyone, I'm new to both the forum and Flash scan. I have been using Autotap for a while and just bought the $250 Flashscan scan only tool because I couldn't figure out what my problem is with Autotap.

I have a 1987 RX7 GXL with a 2000 Camaro SS engine swap. From the time I got the new engine running, it has run poorly. It runs great for 2-5 minutes and then it almost dies at idle and feels like it is missing under acceleration. I have checked and rechecked the O2 sensors. I even replaced them eventually and the problem persists. Autotap reads reasonable values for the two front ones and 450-900mV for the rear two (I have eliminators connected). The MAP and TP sensors check out. The short term fuel trims eventually drift towards a -40% condition. I really have no idea what is wrong. One clue may be the VSS reads 0 mph all the time, but the wiring from PCM to sensor connectors checks out. Also, my aftermarket gauge reads the speed off the sensor just fine. Can the PCM have internal problems and still run the engine? I've been racking my brain about this for months.

My Flashscan with arrive tomorrow or Monday. What do you guys recommend I do first? I have been reading the FAQ's and tutorials and have a decent feeling about the system and am excited to try it out. I'm an engineer and learn quickly. Any advise is appreciated!:master:

thanks,
Tom

[emarkay]

Did the engine perform 100% before the transplant?

It runs great for 2 minutes ONLY WHILE from a cold start? If so, that would be while it's in open loop and using the "average/best guess" fixed tables - you could pull 2 plugs and stuff a sock in the intake and it would still run fine then!

Any possibility of pinched wires or something mechaical like that or some bad grounding issue, fuel pressure, vacuum leak, exhaust backpressure, engine misalignment or other unintended loads?? With a swap like that you need to be an engineer as well as a mechanic!

Yea PCM's can be partially dead - ask any mechanic. They can also be intermittant. Wonder if you can get a "test" PCM to - um - test? Any possibility the PCM was "LSEdited"? They are not "bulletproof", and can be damaged by mishandling, moisture, vibration and other things when not bolted to and grounded in a vehicle.

Keep us informed.

[Wingnut87]

Thanks for your response.

I bought the engine, trans and computer from a yard on Ebay. I don't know how is ran before. I have done my best to check all vacuum lines. I ran shop air into the engine and found a few clamps to tighten. I ran big ass ground straps and am getting good signals from the sensors. Fuel pressure is a steady 62PSI.

The PCM has had some airconditioning codes removed, but the problem predated any modifications. I don't know anyone who can program and "lend" me a PCM. I was hoping Flashscan might be able to diagnose a partially dead PCM. Is this possible? I have dumped enough money down this hole and don't want to buy a PCM (although I will if I need to).

I am an automotive engineer (non engine stuff mostly) and have always been a gearhead. I built three Formula SAE race cars in college. I hope I know enough not to make silly mistakes and figure this problem out, but you never know. I have found so many problems and have fixed them all thinking I solved THE problem, only to find it must be something else. I am out of ideas now. Is there anything I should log first with flashscan that might give me a clue?

thanks,
Tom

[ringram]

wait4me will sort you a pcm for peanuts if you need one. Pity you dont have the tuning version you could run open loop. You dont know what injectors and so on you have do you? if the pcm doesnt match the engine you might have an early pcm with a later model engine. The injectors were larger on later engines. But closed loop should sort that out. I think with the scanner version you might be able to download the pcm tune, if thats the case then compare it against a stock tune to make sure its stock. Also check injector flows etc. Exhaust leaks will cause wrong 02 readings. But my vote is for a 99 pcm with 01 injectors or something. Basically 26lb vs 28lb. Any chance you can borrow a wideband as well?

[Darkness]

He said the fuel pressure is at 62 psi, I dont know if that would affect it but mine is about 58psi.

[joecar]

Can you capture and post a log (include as many parameters as you can) of it running and dying...?

[Wingnut87]

I got my Flashscan today. I will try to log a run over the weekend.

thanks!
Tom

[Wingnut87]

Ok, I used Flashscan for the first time today. It seems there is a lot to learn, but the more info the better! I am attaching a log file. Let me know if there are any other PID's I need to include. The car got pretty bad around frame 52. I was mostly just idling or reving the engine. I drove around the parking lot from frame 112-120. I'm also including a picture of the car.

thanks for your help!
Tom

[ringram]

Nice. A couple of suggestions. If you keep under I think 28 or 26 PIDs you will run at 10fps, over that your sample speed drops. You are logging too much stuff. There is a default pid selection file, use that as a good startpoint. If you have specific issues like idle tuning you can make a new selection up with the relevant pids and save it.

One thing thats often used is wideband input so you may want/need to add that.

Usually map, tps, rpm, ect, iat, kr, g/cyl, g/sec which should all be in the default list should form your base as you will build maps with them to track spark and fueling.

[TAQuickness]

Given the unknown past of the engine, I would start with a compression check, then systematically check each sensor with a meter (might want to send Joecar a pm for help on this one).

I would cross reference the engine, injectors, and PCM part numbers to verify compatibility.

Verify all injector connectors are secure, verify all spark plug wires "clicked twice", verify the grounds.

[joecar]

Tom,

Yes, like Ringram said eliminate some of the PIDs, try to keep the number of channels no greater than 24 (some PIDs take 2 channels, the calculated PIDs take zero channels); include SAE.TP for throttle position.

Looking at your log, it seems to me that it's running rich and the PCM is removing fuel (and you said the LTFT's drift off to -40%);

in the first 3 minutes of your log, the HO2S11/HO2S21 voltages barely switch, and then after than they peg high, indicating way too rich.

with engine off, prime up fuel pressure (key on several times or use bi-dir mode or use fused jumper wire in relay socket), and measure pressure at fuel rail (tell us what it is) and see if it holds steady for a minute;

Edit: run pump for 10 seconds, shut it off, and then see if rail pressure holds steady for a minute.

then, with engine running, what is fuel pressure...?

The powertrain in your photo comes from an F-body (you got that to fit in your stealth RX-7, good job :cheers:), but what fuel pump are you using...?

Like TAQuickness said, check that the plug wires are properly engaged on the plugs and the coils.

Edit: and check all the ground connections.

If you want more info on testing any sensor, just let me know, I can find the test procedure for you (some sensors you just ohm them out over their range).

Edit: Got anymore pics (engine bay, underneath, etc)...?

Joe

[Wingnut87]

First of all, thanks for all your replies so far. At the moment, I am pretty confident that the MAP, TP and O2 Sensors are all working correctly. I verified the TP using autotap rotating the throttle blade by hand. I put a hand held vacuum pump on the MAP with Autotap as well. I have changed the O2's and checked the wiring.

Joecar - I get a constant 62 PSI of fuel. I am using a stock, but brand new rx7 fuel pump which is supposed to be good for 400hp or so (I'm not the first to do this conversion). The rotary was a fuel hog. I will have to borrow the pressure gauge again to check leak down after I shut the engine off. I have about 100 pictures of the conversion and have been waiting to post them to my website until I finished the car. I'll post a few here just for kicks when I get back to work (I'm at a friends house now). Thanks for the compliment!

About compression tesing. The car runs fine sometimes. I would be totally happy if it ran that way all the time. Doesn't this point to the problem being electrical or computer?

I changed the plugs and wires and made sure I got a good click each time. There was no change in performance.

My PCM part number matches the one listed for my engine in alldata.

One question that I have. I have checked wiring for the VSS and it is fine. My autometer gauge taps off of the circuit and reads fine, but I get no signal in Autotap for EFIlive. Any ideas? Is it likely, there is a problem with my PCM?

Also, what is the best way to check the injectors? What about the Flashscan injector balance test?

Also, I have verified all grounds from the PCM connector to the chassis and battery. I ran some heavy cable around the car to make sure because the battery is in the back. Anywhere else I should check?

Ok, here is my ToDo list. Please add if I'm missing something.
1. Fuel pressure leak down.
2. Re log a drive using flashscan and less PID's at high rate
3. Check injectors
4. Try to find a PCM I can borrow?
5. Post pictures

thanks again!
Tom

[ringram]

Ah ok, you might want to get a copy of redhotsupra's injector flow table and recalculate yours. Seems with that pressure you might be like 8% over GM stock pressure (58psi or so) If you run a nice flat 62, then recalc the injector flow, it might help your rich condition.

I think there are 2 output wires for VSS, GMPX will be able to tell you exactly which one matters.

[joecar]

Tom,

Question: is your fuel pressure regulator MAP referenced (has a vacuum line going to it)...?

If not, then proceed...

as Ringram said, use this Excel spreadsheet to recalculate your B4001 IFR table:
http://www.allmod.net/hpt/injectors.xls


Your IFR will vary with manifold vacuum because the FPR is not MAP referenced (i.e. B4001 will be sloped).

You already know/measured your rail pressure (62 psi), you need to lookup the rated flow rate and pressure for your 2000 F-body injectors (...someone may care to post these numbers...).

[Wingnut87]

My pressure regulator is not MAP referenced. So let me understand what is going on. Alldata says fuel pressure at the test port should be 55-60PSI. Either because of the rx7 fuel pump pumping too hard, or a restriction in the return I plumbed, I'm running at 62PSI. Since this is higher than the range, the engine can't compensate for it.

The way I see it, the proper way to fix it would be to regulate fuel pressure down to 58 PSI or so. What you are saying is that I can recalculate one of the tables in the PCM and reflash it to compensate. Do I have the right idea? Since I have EFILive scan only, I'll need to find someone to flash my computer. Or I can buy an adjustable pressure regulator. Either way, it sounds to me like you guys have helped me find a problem that very well could be the source of my misery. Let me know if I am understanding correctly.

thanks!

[Wingnut87]

Here are some more pictures. Please excuse the exposed wires, I am still troubleshooting. The exhaust is finished, but I don't have a recent picture. The picture I am including shows it 2/3 complete.

[joecar]

My pressure regulator is not MAP referenced. So let me understand what is going on. Alldata says fuel pressure at the test port should be 55-60PSI. Either because of the rx7 fuel pump pumping too hard, or a restriction in the return I plumbed, I'm running at 62PSI. Since this is higher than the range, the engine can't compensate for it.

The way I see it, the proper way to fix it would be to regulate fuel pressure down to 58 PSI or so. What you are saying is that I can recalculate one of the tables in the PCM and reflash it to compensate. Do I have the right idea? Since I have EFILive scan only, I'll need to find someone to flash my computer. Or I can buy an adjustable pressure regulator. Either way, it sounds to me like you guys have helped me find a problem that very well could be the source of my misery. Let me know if I am understanding correctly.

thanks!Tom, yes that's correct. :cheers:

[joecar]

Here are some more pictures. Please excuse the exposed wires, I am still troubleshooting. The exhaust is finished, but I don't have a recent picture. The picture I am including shows it 2/3 complete.Ha... it seems you have more space in the engine bay of your RX7 than I do in my Trans Am. :beer:

[Wingnut87]

Well guys, I think you helped me solve it. I found a major problem with the way I initially plumbed the fuel. It runs great now! Thanks!

[joecar]

Well guys, I think you helped me solve it. I found a major problem with the way I initially plumbed the fuel. It runs great now! Thanks!Tom, good work... :cheers:

What was the problem...?

Joe

[Wingnut87]

This 2000 LS1 engine is actually the newest engine I've had in the project car. I didn't realize the thing was a returnless system with the regulator in the tank. Of course, I had the rx7 tank, so I was running with no regulator. Anyway, I used a corvette fuel filter with a built in regulator and return. Problem solved.

I'd driven the car sucessfully for a few days until today when I got on the throttle for the first time on an on-ramp. Since then, I have had a knocking sound. I can't believe this never ending project! The last thing I need is an internal problem. I'm going to start a new thread for this. Please help!