I just upgraded to MSII and this MSQ is what I ended up with after a couple hours of fooling around. Currently the only real problems I cant figure out are that I'm losing a tremendous amount of power at 5k rpms under boost. It's bad enough that I dont think the car can pull past 6k rpms. AFR's are good (albeit a little rich, but shouldn't be enough to cause a problem). Timing dips to 8 degrees according to the log and I cant understand why. FWIW I've heard the car backfire through the exhaust when trying to "push" it past 6k. I have a feeling my ign. settings may be wrong.

Also, my GammaAE is always at 118 or so. It's showing up on the log that warmup enrichment is active, even with the car fully warmed up. I'm not familiar with MS2's settings yet but I don't see anything obvious that could cause these things. I'm concerned about my dwell settings. I tried to mimic what I had in MS1 but I'm not sure if it works the same way. Any help is appreciated.

 

Here's what I see.

You've told the MegaSquirt to never leave warm up enrichment. Your "warm" value is 120%.

Your rev limiter is set as a soft limiter with authority remove 12 degrees of advance at 5500 rpm. This would make it fall on its face at 5500 rpm.

 

~facepalm~ That was really stupid of me. Thanks. See anything in my MSQ that would indicate why my fan runs constantly?

 

No.

There are a lot of little changes that needed to be made to convert your MSPNP to a MS2 processor, and I did those mods to yours myself. It is a bit less straight forward than working with just a V3.0 board because the MSPNP has many critical circuits on its daughterboard. Yours was definitely a one-off, and I thought I got everything but maybe I didn't. It worked correctly on the stim, but I couldn't test everything without getting it on a car, and I don't have a basemap for that--which I know Matt made you aware of. I also asked Matt to charge you for less time than what I actually spent on your ECU because it took me a while to get my conversion game plan together, and in case anything little like this popped up later. I had previously converted a 90-93 model to MS2 for use as a shop tester. That one started up and spun the wheels on the dyno, but I never took it further than that. There are several differences between the MM9093 and MM9495, so I couldn't fully apply that experience towards yours.

I don't have specific documentation here at my home, but if memory serves the 94-95 MSPNP uses JS3 for fan output. I bet you have a jumper from JS3 to one of the spare pins on the DB37. That jumper will need to be moved to JS2, and then you can set your fan output on JS2.

Sorry, that would have been something I missed. It's little things like this that make us reluctant to perform such mods in the first place, as we want to check all assembled units (especially plug and plays) before releasing them. But I'm glad to hear that otherwise your MegaSquirt is doing well. Luckily you can rectify that pretty easily. Let me know if this is indeed what happened, and I'll add it to my notes on the conversion procedure.

 

Oh I know that this was experimental. No problems there. I expect little things like this to crop up. Just the nature of the beast. Eventually it'll be perfect. The fan running all the time is better than the fan not coming on at all right? I'll check the jumpers as you suggested.

Anyway. I removed the rev limiter, still doing it. Falls on it's face at about 5200 rpms. Trying to push it often makes it backfire through the exhaust. I bumped the timing up in that area as well, but there was no difference. I'd noticed that the tach dropped momentarily when the backfire occurred, although that doesn't show up in the log. Log and MSQ are attached.

 

Just noticed, it does it when I free-rev as well. I took a video of the tach bouncing around. The forum wont let me post it as an attachment so I uploaded it to youtube.

http://www.youtube.com/watch?v=XLxZ-w89Ecg

 

Log is clean, though your AE settings need some work and you're a little bit lean in spots. Could be some det, but looks more like a run of the mill misfire to me. Need more coils and/or less gap.

 

Yeah I know AE isnt right yet, but it's definitely not detonating. I'm 100 positive of that. The only reason I'm more inclined to think it's a problem related to the ignition circuit/settings is that the tach drops out when this is happening. A regular misfire wouldn't cause this. My gap and coils were sufficient before as well. That shouldn't change. I wish I had a scope so I could see what's happening on the coil trigger wires.

 

actually the fact that your tach drops but the log is clean is a nearly 100% sure sign of a misfire. may need more or less dwell, but otherwise points to coils/plugs/wires.

 

I'm confused. The coils give the signal to the tach right? If the spark was blowing out, how would that have an effect on that signal? It's as if the signal driving the tach itself is dropping out.

 

Don't ask me to explain it, but spark blowout will cause the tach to drop out. I got that on my car before I switched to the Toyota coils - bucking and hesitations with the tach falling to 0 every time. A new set of plug wires fixed it right away.

 

I'm going to look towards my dwell times then. The plugs and wires are fairly new (maybe 6k on them) and were perfectly adequate before the upgrade and logically still should be.

 

Finally figured it out. It was as simple as this. In ignition options, the option for "use prediction", set it to NO. That cleared it right up. Revs as high as I want to now with no misfires/breakups. The flat spot at 5k is still there however, but it seems that was an entirely different issue the whole time. Feels like it's pulling timing at 5k, and then at 6k it goes back to full power and pulls great. AFR's are a-ok the whole time. Didn't have time to play with it much more tonight so I'll deal with it another time (and when it's cooler out).

 

I popped the cover off, and it seems something is already occupying JS2. I cant tell what since it's soldered from underneath.

 

Just use whatever programmable output that isn't currently occupied:

http://www.msextra.com/doc/ms2extra/...htm#ms2options

JS0, D15, JS11--one of them will be free.

 

Haven't actually done this yet but before I go for it, I'm confused about something. The outputs on the board are named JS2, JS3, etc... in Tunerstudio, under 'output port settings', the configurable outputs are named PM2, PM3, PT6, etc... How do they correlate? What's with the different names?

 

Any progress on the flat spot issue?

 

I was having exactly the same problem with my newly installed DIYPNP ('89 eunos roadster, missing badly ~5K+) and the no prediction setting solved it for me.

Thanks! I was starting to get worried that it was going to be something expensive.