Turbonator SMEC codebase

[ShelGame]

Mine is still cutting out while driving.

The new .asm from above?

[Force Fed Mopar]

The new .asm from above?

Yeah.

[ShelGame]

Yeah.

Well, I confirmed that it's not the launch control doing it; and no the regular rev limiter either.

I'm starting to wonder if I've hit that mysterious processing limit again. Previously, when I had alky and NO2 as well as the CCD Bus routine all enabled, it started having a mis. The only thing that fixed it was taking out the alky routine. Since then, I've taken out others that aren't needed (Emissions maintenance reminder, CCD bus, etc.) So, I thought it would be OK now.

I'll post another test .asm later with some of those routines removed again.

[Force Fed Mopar]

I was gonna say maybe it was running out of processor speed, but then I thought I'd be dumb for saying it lol. Since I had no clue how much processor was actually being used, or how hitting the limit would affect it.

[ShelGame]

I was gonna say maybe it was running out of processor speed, but then I thought I'd be dumb for saying it lol. Since I had no clue how much processor was actually being used, or how hitting the limit would affect it.

You know, it's odd. I know I've had this issue before. But, I've been dis-assembling the Neon FCC for the past several months; it uses the same processor and does a TON more work than ours. The FCC has probably double the lines of code of even a '92 Turbo cal - almost all additional diagnostic routines. And, it doesn't run out of processor time, runnign the same main loop time of 11ms. So, I don't know what the deal is. We shouldn't be running out of time. But, it could be some other type of processor error that's setup based on the original code. I'll have to look into it some more.

In the meantime, here's another test .asm with the PTU and NO2 retard removed.

EDIT: .asm removed...

[wowzer]

rob - any way i can help using my smecstim and mpscan?

[Force Fed Mopar]

You know, it's odd. I know I've had this issue before. But, I've been dis-assembling the Neon FCC for the past several months; it uses the same processor and does a TON more work than ours. The FCC has probably double the lines of code of even a '92 Turbo cal - almost all additional diagnostic routines. And, it doesn't run out of processor time, runnign the same main loop time of 11ms. So, I don't know what the deal is. We shouldn't be running out of time. But, it could be some other type of processor error that's setup based on the original code. I'll have to look into it some more.

In the meantime, here's another test .asm with the PTU and NO2 retard removed.

Interesting. Maybe it's a RAM problem? Not sure if our computer use memory like a regular computer or not. I'll try this .asm on my way to work this morning.

[ShelGame]

rob - any way i can help using my smecstim and mpscan?

Maybe. If it's a processor error, I would think you could see something on the stim. If the misfire is a spark or fuel thing, then maybe not.

If you can try to run v19 on the stim; using the previous test .asm posted; and see what you can see when it starts misfiring. The way it runs in my car, it seems fine at low throttle openings and will rev up just fine. But, if I tip in a little too much it starts breaking up; kind of like a rev limiter, but not really.

I guess another possibility is the stack overrunning the RAM locations I'm trying to use. But, I'm still only using RAM that was used in the stock code, just re-purposed (I deleted the CCD bus, etc. and used that RAM). The top 2 RAM locations before the stack are stock locations.

[Force Fed Mopar]

Tried it out, still acts up around 4k. It not so much like a limiter as it is a really bad misfire, like the spark breaking down or something. I stayed in it this morning and it will keep accelerating, but misses bad, AFR's bounce from 10.5 to 12.9.

[wowzer]

Maybe. If it's a processor error, I would think you could see something on the stim. If the misfire is a spark or fuel thing, then maybe not.

If you can try to run v19 on the stim; using the previous test .asm posted; and see what you can see when it starts misfiring. The way it runs in my car, it seems fine at low throttle openings and will rev up just fine. But, if I tip in a little too much it starts breaking up; kind of like a rev limiter, but not really.

I guess another possibility is the stack overrunning the RAM locations I'm trying to use. But, I'm still only using RAM that was used in the stock code, just re-purposed (I deleted the CCD bus, etc. and used that RAM). The top 2 RAM locations before the stack are stock locations.

i thought i could probably try tracking the loop time and see if it is being extended. also, maybe spit out the stack pointer and see how low its going.

[ShelGame]

i thought i could probably try tracking the loop time and see if it is being extended. also, maybe spit out the stack pointer and see how low its going.

That would be helpful...

---------- Post added at 11:55 AM ---------- Previous post was at 11:53 AM ----------

Tried it out, still acts up around 4k. It not so much like a limiter as it is a really bad misfire, like the spark breaking down or something. I stayed in it this morning and it will keep accelerating, but misses bad, AFR's bounce from 10.5 to 12.9.

That's good to know. At least it's not going lean, so it's not a fuel cut issue or lack of enrichment.

[ShelGame]

OK, I found what's causing the mis. I just hope the fix doesn't f-up the staging limiter. I'll post an .asm to test tonight. Can't upload from work anymore

[turbovanman²]

OK, I found what's causing the mis. I just hope the fix doesn't f-up the staging limiter. I'll post an .asm to test tonight. Can't upload from work anymore

What did you find?

[ShelGame]

What did you find?

Something I f'd up when I was trying different things on my car. I commented out a line that I shouldn't have in a section of the distributor routine that only runs over 3776rpm...

[ShelGame]

OK, here's the next .asm to try. This has all the features I took out re-installed. If you have the chance, or if you dare, try setting the launch control both above and below 3800rpm. There's a section of code that switches dwell calculations above and below 3800. I want to make sure the launch control works both above and below that.

[Force Fed Mopar]

Okay I'll try it here shortly, or in the morning.

[Force Fed Mopar]

It works! 4200 staging limiter sounds kinda crazy lol. I'm going to have play around with the spark cut some more because 5/7 isnt holding it quite right either. Also, it doesn't sound like the anti-lag is working on WOT staging, just part-throttle. I'll try and get a vid of the difference.

[ShelGame]

It works! 4200 staging limiter sounds kinda crazy lol. I'm going to have play around with the spark cut some more because 5/7 isnt holding it quite right either.

Good, finally got it.

Also, it doesn't sound like the anti-lag is working on WOT staging, just part-throttle. I'll try and get a vid of the difference.

Hmm...

[ShelGame]

Also, it doesn't sound like the anti-lag is working on WOT staging, just part-throttle. I'll try and get a vid of the difference.

So what exactly does it do differently at WOT? Are your WOT and PT fuel tables the same in boost? Finally, what's your A/F while staging?

You've made me think I need to do something to force the fuelling at least to run the WOT tables while staging - for safety sake. P/T fuelling, while staging with 5+psi boost seems like a bad idea. At least there shouldn't be spark knock with the antilag running a low timing.

I'll have to look again, but there's a boost limit for closed loop operation. Above a set boost level, it will automatically switch to open loop, though it doesn't necessarily switch to the WOT fuelling. That is based on TPS only. So, if your PT and WOT fuel tables are the same in boost, it shouldn't make a difference. But, if you stock-ish AFR's in boot on the PT table, it might be too lean.

[Force Fed Mopar]

I'm gonna check over my settings again, I may have forgot to check the Antilag box on the 4200 cal. I did a test launch with it tonight and it did seem to hold 4200 pretty steady and sounded like a true 2-step, but no bang-bang associated with the anti-lag.