Glaser67 said: "I would suggest starting a new thread, not only for you to get views/comments about your specific question, but also would benefit anyone searching for a similar issue, just my .02,

That being said, please clarify which MAF meter you used and any other mods, such as cam and headers "

OK, new thread, all the gory details.

I started simple; just a straight conversion using ALMOST all stock configuration. The engine had been running an aftermarket set of 1.5" tubular headers for about 30 years, and a few years ago I installed a set of 1.7 Cobra rockers. This was all running on the original SD computer. I want to build a new engine to replace the 30+ year old stock engine, with enhancements like a 331 displacement, AFR 165 heads, GT40 intake, and some yet-to-be determined cam. So based on all I've read, the stock computer can not handle these changes, and most of the reprogrammers only work with the MAF computers, such as the A9L. I got the A9L and matching MAF metere from a dismantler many years ago, and did the conversion recently following the instructions found on the net.

Right after the conversion, my first problem was that the engine would not stay running for more than about a second after starting. After checking all the wiring, I found out that if I pulled the MAF signal wire from the meter, the engine would run after starting. Obviously, I couldn't run that way, so there was still a problem somewhere.

The self-test indicated low MAF voltage, and I traced that to the blown caps that everyone recommended replacing. After I replaced the caps, the low MAF voltage error code went away, but the engine still would not run for more than a second after starting. Again, pulling the MAF signal wire allowed it to run.

Eventually, I got another A9L and matching MAF meter to test with. This one allowed the engine to run, with the MAF signal connected. So the first thought is something was wrong with my first A9L computer. Since the cold engine was idling badly, it stalled while I was checking the meter.

That's when I noticed crackling sounds, and smoke coming out of the bundle of wires going into the firewall. I pulled the battery cable off as fast as I could, and tried to find where the smoke was coming from. It turned out to be from one of the wires from one of the AIR solenoids. It was frayed, and may have shorted out to its +12 wire. So I fixed that, and tested all the other wires from the EEC connector to where they should go for continuity using a test light.

After verifying there were no more shorts, and all the fusible links were still fusing, I put it all back together, but this time I installed my old A9L computer. It ran the engine as well. So I'm guessing that the shorted AIR wire must have messed up its signals or something.

Re-running the self tests still showed code 67, the NSS error, but nothing else. On my 87 car, the HEGO harness only has the wires for the sensors, and nothing for the transmission, as the later models do, so I'll have to make some kind of jumper if I want to simulate the NSS switch. As I mentioned in the other post, the engine seems to idle high while coasting, and drops to normal when I come to a full stop, which indicates that it's seeing the VSS signal that I wired in.

So for now, the only symptom left is the poor cold idling, and the occasional idling weirdness while driving. Would the lack of the NSS cause this? I notice in the wiring diagrams that the NSS is connected in parallel with the clutch switch. I haven't checked to see if that wire is connected to pin 30, but if they are in parallel, do I even need a NSS?

 

Sound like you've made some good progress troubleshooting thus far. I believe the NSS is the equivalent of the clutch sensor on a manual. I would get that on my manual (A9L) but when I did KOER with the clutch pushed in, it went away. Also, I cannot pull codes with engine on without pushing in the clutch. Anyway...

As for the poor cold idle, I'm working through that myself now too. Mine sometimes dies instantly when cold. See my setup below. My belief is that with mods like outs, the engine in not getting enough fuel on startup (cold engine required a lot more fuel to run, contributor being poor atomization), or problem with the MAF transfer. I picked up a Quarterhorse and associated software which is allowing me to make these adjustments. Its tough.

I think the other schools of thought would be to monkey with the idle set screw, fuel pressure, ect, but IMO your just tricking the ECU into doing what you want. Something to consider is you MAF setup; is it the same meter in the same housing with the same intake/airbox that is paired with the ECU? Even clocking the MAF meter in the intake can change the readout.

 

Thanks for the input. I didn't realize MAF positioning can make a difference. In my case, all the parts came from the same car; the tubes between the air box and sensor and between the sensor and throttle body, the sensor itself, the A9L computer, and the wiring harness. I cut out the 4 wire harness for the MAF sensor from the doner harness and spliced it into my stock harness. So all the parts should match. I found a couple of unused holes punched into the shock tower that perfectly matched the mounting hole on the MAF bracket, so I installed bolts through them to hold the bracket. It also positioned the MAF sensor so that it fit well in the paths of the inlet and outlet tubes.

If I swap my SD computer back in, it would run normally; smooth idle at cold start. So I'm guessing it's not reading the MAF sensor correctly when cold. I did check to see if it was actually reading by pulling the signal wire while the engine is running, and the engine does make a stumble. Maybe I need some kind of a tuner device to correct this. But this is a pretty stock engine, with stock heads, cam, and intake, so I thought it should be operating in the range that a stock EEC should be able to handle.

 

OK we need to backup a bit because a MAF converted vehicle should not run at all on an SD computer, there are a while bunch of wiring differences at the PCM that you should have made not the least of which is adding all the wiring for sequential injectors.

 

The 1987 Mustang GTs have the HO 5.0 v8, which has SEFI. The two wiring changes are moving the AIR solenoid pins on the EEC connector from where they are for the SD computer to where they need to be for the MAF computer (can't remember the exact pin numbers off the top of my head).

In addition, wires are added for the fuel pressure relay sensor, the VSS, and maybe some changes for the transmission NSS. I added the first 2, and the way the A9L behaves tells me they're working. I verified that the positions for those pins on the SD computer are not connected to anything, so it was safe to leave them there when I swapped the SD computer back in. I also swapped back the AIR control wires when I did that to make sure they still worked.

I could not do the NSS mod (pin 30), as the HEGO harness in my car has none of the wires for anything going to the transmission, just the wires for the HEGO sensors, and I think the oil level sensor.

 

No sorry... you got lot of work to do yet. Here are all the differences between the A9L and an SD truck PCM. All the wires in colors have to be moved and all those in grey must be added. You need to completely rewire all the injectors, add the MAF meter wiring, move the AC clutch to make room for the second O2, move the fuel pump monitor(there is no fuel pressure sensor), and move the TAB/TAD solenoids.

 

I'm sorry, I should have made something clear in my first post: I am converting my 1987 Mustang GT's SD computer to MAF using the ECU, MAF, and wiring harness from a 1990 v8 Mustang, both 5 speed. So the car already has the SEFI system and dual O2 sensors, and most of the wiring of the SD system already matches those of the MAF system. The stock Mustang SD computer did not have a fuel pump monitor, so I had to run a new wire. However, I didn't see any docs saying that the MAF system required running both wires from the VSS to the EEC; I ran just 1 based on the docs I read.

Thanks for the pinouts; I see a lot of pictures of the MAF Mustang computers, but almost none of the SD computers.

 

Well it makes sense now. What I would do is clean up my battery terminals and all enhine, and ECM grounds. I've had same/similar issues resolved by cleaning everything up with metal sandpaper/brillo pads and re-tightening, I mean it sure couldnt hurt.

Also, on cold start, how cold is it outside, and do you get that 1,200 rpm surge when it fires? Could you describe the rough idle? Low, stumble, surging?

 

Yeah that makes a bit of difference.

That is an SD truck computer though.. I'm not certain but I'd expect the SD Mustang computer to be pretty much the same as the MAF version except for the meter wiring.

 

I just found this PCM diagram that says it is for an '87 mustang.. https://www.justanswer.com/ford/8ij3...lectrical.html

 

Thanks, this one is a lot clearer than what I've seen. It turned out I had an old photocopy of a page from a Haynes manual that has a similar picture, but it's really hard to make out. It has dashed lines for the MAF associated wires that were added or changed in the later models, and are even more difficult to discern.

I'll try to add that second VSS wire this weekend if I have time. (still working on the 69; hoping to complete the head/intake install)

 

Looks like my brain cells are dying, as I had forgotten that I did make all the wiring changes for the conversion, including both VSS additions. I was thinking about why the EEC would respond correctly if I had not connected them correctly. So that's not the cause of the poor cold idle. The search continues...

 

Did the battery terminal connections and ecu/engine grounds make any difference?

 

I didn't change the battery and ground connections on the stock harness, which is still in place, and still works great with the SD computer.

However, I did make connectivity tests between the power and ground pins on the EEC connector and where those wires tap into the car's power and ground points, and they're solid. I used a high powered LED of about 6 watts, so it draws at least 1/2 amp, so it's more effective than just using an ohm meter.

 

Finally had a chance to get back to this.

Still struggling with very poor cold idle. It's quite obvious that it's running too lean when the cold engine is just started. If I push the throttle enough to keep the engine running for about a minute, it will stay running, though poorly. For the first few minutes of driving, it is definitely too lean, as it will buck, surge, backfire, and stumble, if I'm not careful with how to rev up the engine when trying to get the car going. Once the engine warms up, it runs much better, though idling is still a bit rough.

Does any EEC experts know if the program needs some kind of a signal to tell it to run in a special mode when first started? I checked the ECT signal, and it seems to be working. I'm not sure if the MAF signal is correct. Does anyone know what are the range of values its output supposed to be? I can try to wire a voltmeter to it to see what it shows when cold or hot.