Not-quite resets
 

I'll be driving, and it's like I clicked the "burn to ECU" button, but I don't even have the laptop hooked up. When it does it, it'll do it a number of times in a drive, but other times it won't do it at all. It'll do it when megatune is hooked up, and the reset counter doesn't change, no resets show up in the log. Actually, nothing unusual shows up in the log.

Any ideas what this might be, or how I could fix it? It doesn't do it all the time, but for some reason it likes to do just after I get on the interstate.

I'd search, but I don't even know what I'd search for, since it doesn't seem to be a reset problem.

This sounds very much like a missed input on the CAS. Do you see any flutter or noise in the RPM trace on the log when this happens?

Yes, actually. There is just a little flutter when it happens, but I barely noticed it. The only other indication in the log that it even happened (besides the big red line where I put the mark) is a slight spike in the AFR.

So if it's the CAS input, how can I fix it?

Well, my solution was to throw the CAS into the trash and lathe an adapter to mount a missing-tooth wheel on the crank. But that's a tad extreme.

Have you performed the "cap mod" on your input circuit?

If it's only a small flutter (and not a huge spike) then it could also be a simple misfire. What does your ignition system look like, what's your plug gap, and what value resistors did you use for the pullups in the Spark Out mod?

My ignition system is stock, I honestly don't remember what I gapped my plugs to, but I read on here what to use before I did it.

If by cap mod, you mean the 0.1uF cap to run HiRes, then yes, I did that. The resistors I used were the ones specified in Brain's DIY writeup.
My board looks like this:
http://i29.tinypic.com/10mn9eq.jpg

Probably a dying CAS. But try new plugs BKR7E's or equivalent (they're changing the stock numbers on these as I recently found out when I bought 12 of them...) and gap them to .020". Might fix your problem. I'm also running a custom trigger wheel setup and never actually used a CAS, so I'm not too familiar with how they fail other than knowing what I've read on this site about others' problems.

Do you have a data log of this happening? If so, could you please post it?

I'll add a +1 to Pat's suggestion that you try some new plugs at .020".

Additionally, you're probably running 1k pullup resistors in your spark output circuit. I personally never had an operation issue with this config, but I did observe it with a scope to produce ugly trigger rising edges. Someone else here actually observed this to result in a net loss of dwell, which can cause the problems you're seeing. With regard to the following illustration, inspect the two resistors on your board which on the drawing are yellow rectangles marked as 1k. If yours are 1k (brown black red [silver/gold]) replace them with a set of 220 ohms units (red red brown [silver/gold]) which may strengthen your spark a tad by giving the coils full dwell.

http://diyautotune.com/images/cartec...t_mods_med.jpg

In my experience, CAS input problems are generally represented as a large spike in RPM, whereas actual misfires will produce a more subdued ripple.

Ok, attached is a datalog of the event. It happens just before the marks, I'm figuring where the afr spikes, at T=10412.905

So it looks like I have a more subdued ripple, which means spark. I'll give some new plugs a try, and if that doesn't work I'll try changing out those resistors for smaller ones.

FWIW, the dash tach does the same twitch thing it does when I burn something to MS.

do you have a constant Barometric Sensor?

I don't think so. I think it just takes the reading before startup. Really messes with "boost" readings though.

Actually, it looks like an intermittant CAS signal to me rather than a misfire. The RPM trace itself has just a small ripple, but there's a big one in the RPM / 100 log.

I don't see anything in that log that points to a CAS issue. In fact, I don't see any tach ripple at all- your whole boosted run there looks clean.

Try new plugs gapped to .020, and if that doesn't work, replace the resistors with 220s.

Also note the plug gap and heat range of your old plugs and a couple pics of them please.

Good observation, Matt. You're referring to the dip at 10412.779? That does look like a CAS glitch.

I can do that, but it might be a couple days before I get 10 minutes to do it (busy week. I don't even get to drive my car more than twice a week, and that's just to get groceries!). Should I do anything specific before I take the pics, like shut the car off at some specific situation, or just whatever, idle's fine?

can you post your MSQ? (no need to zip it)

Sure. For some reason I thought I remembered that I had to zip stuff before I could post it here, but as I was posting that last one I saw that .xls and .msq files could be posted.

Yes, that is what I meant.