Jiggerachi

I've been revisiting this issue the last month and a half, debating starting a new thread since its been so long but i'll keep it here for now.

Things i've done/confirmed since spring classes let out.

Added copper heat sink to MTX-L - got rid of the E8 overheat error. Recalibrated wideband while I had it out.
Had a friend who knows MS very well tune the VE table, timing, and work on idle. Runs great on moderate temp days.
New spark plugs/wires.
Confirmed baseline and max fuel pressure is good, using the ignition on jumper trick in the FAQs.
Confirmed FPR is operating correctly by disconnecting vac line.
Confirmed the fan circuits arent causing any issues by disconnecting them briefly and triggering the issue.
Attempted to rule out AC idleup settings by incrementally lowering the idleup duty% down to 0.0 with no change(power cycles each step), engine still cut out at whatever RPM. I guess turning off idleup disables the AC completely because that's what it did and the issue went away since the AC was no longer cycling (assuming that's normal but really not sure).

Each step of the way I was still able to reproduce the 'engine cut' issue, whatever it might be. At one point I was able to get it to happen at idle and revving in the driveway. I have this on a log i'll post. My friend and I are narrowing it down to something in the MS3 wiring possibly, but we are stumped.

To confirm its the MS3 and not anything mechanical, I waited for a hot day, took the car around the block and easily recreated the engine cut at AC on around 2 - 3kRPM reliably. Parked it and immediately swapped back to stock ECU, took the car back on the same route, was able to bounce off the rev limiter multiple times. Could not get the engine to cut out no matter how hard I flogged the car. Came back swapped the MS3 back in, drove around and couldn't make it past 2k with the AC on. Its definitely something in the MS wiring or tune, not mechanical.

This week I plan to flash to the latest firmware since i've ruled out mechanical issues it's the last thing left to rule out before narrowing down to software/wiring issue.



I have done a ton of reading/learning/tuning, the only glaring issue that I can't get past is this damn engine cut with the AC on. It makes the car undrivable, and now that i've isolated the event it's causing half my idle problems. The rest is just dialing in the MAT corrections for hot days which I can handle...as long as I can have AC, lol.

Logs and msq shortly.

Baseline ign-on fuel pressure:




MAX fuel pressure




Idle




Disconnect vac line from FPR

Jiggerachi

MSQ and datalog attached.

This log is the best I could isolate the issue. Sitting in the driveway, revving up to around 3k, hit the AC button. A second or two after the compressor kicked in the engine would cut out and RPMs would drop. If I could switch the AC off before RPMs hit zero, the engine would spring back to life and recover like nothing happened.





It would do this at idle as well, but not enough time to turn off the AC before the engine died.

Braineack

iTrader: (62)

I'm looking at your log in excel and I see a lot of glaring issues:

1. your AFRs look incorrectly calibrated--youre not actually idling at 22:1 AFR like it's suggesting.
2. your TPS calibration looks wrong--max is 69%, min is -4%.
3. your enrichments are crazy-- your MAT corrections are pulling 10% fuel, warmup adding 5% (at 160°F CLT), and baro adding in another 1%; for a grand total of 97% fueling.

Judging by your PW before the rpms drop to 0, I'm guessing you flooded the motor. your accel enrichments add in a ******* **** **** ton of fuel just before it dies. up to 150% at one point when you blip the throttle and then your AFRs show full lean because the sensor doesn't work when it's satruated in raw fuel.

4. your VE table doesn't seem to adjust for the load while the a/c is running. your idle at 60% VE without a/c @ 15.5-16:1 AFR*. but when the a/c is running your at ony 63% VE and your AFRs show 17.5-18:1 AFR*.



You just need to tune the car. In the least you need to tune your IDLE VE table to add fuel when the load increases with the a/c running.



*supposedly.

Jiggerachi

Where are you seeing this? The car idles between 13.5 to 15, never seen anywhere near that number in MS or the wideband??

Oh I see it in excel, after each start the MTX-l goes through warmup even if the engine just died, in the logs it has junk readings like this. I think that's what you're seeing. After warmup it idles near 14.7 with exception of what I need to fix in MAT correction.

Strange, i've calibrated TPS a handful of times, and recently, last I looked at it in MS it was -0.2 - 98 or something. I'll recalibrate but its odd this is different than what MS reads.

Edit, I never went full throttle for this log, that's why MAX is 69. Min is between -0.2 & -0.6. Easy enough to recalibrate, at I will, but it's not a problem.

Edit: This may be a result of the issue rather than the cause(or I may be misunderstanding), a sec or two after teh ac clutch engages, the engine just dies, assuming its still pumping in fuel with dead ignition or something, it would definitely flood.

3 and 4 noted, these are the things i'm still working on. The problem is something is turning on like a switch, completely killing the engine when AC on and RPM reaches whatever threshold it doesn't like that particular day, and suddenly spring back when either are no longer true. I can see MAT and accel enrichments exacerbating the issue, but something is still going on.

Edit: I believe I had set MAT corrections back to default at the time of this log, Let me double check.

gjsmith66

This is a shot in the dark, however I had a similar issue, although not tied into A/C usage. Take a look at your launch control settings and make sure launch control is turned off.

Braineack

iTrader: (62)

that entire log the car is averaging idle AFRs well above 14.7. I dunno how you can miss that.


looking again i see you may have been holding the throttle open (records 105-109sec) where it's holding at 3000rpm but 18-22 AFR before it dies.

were you pretty much always holding the throttle open? I see no clean areas on the log where the TPS was at rest and a/c engaged. It's always around 3000RPM and 5% TPS or so.


ah okay, it was only that it flat lined at 69% that i thought you when WOT and held it.



A/C shouldn't turn on like a switch, there should be idle up delay set (i used 250ms) in so the engine has time to open up the idle valve and get ready for the load to hit.

Jiggerachi

I was referring to the usual. That was a hot day and MAT is not tuned so it was running lean, but nowhere near 22AFR. I think that's the MTS-l throwing bogus numbers while still in 'warmup' mode.

Right, there wasn't much idling on this log, I got it to where I could reliably make it happen then grabbed it on a quick log.

My ac idleup settings could be wack, I'll admit that. I've been googling for months and I can't seem to find anyone who explains it. I've tried to do what made sense to me /emulate other settings i've found.

Jiggerachi

Yea I have all of the launch control settings turned off.

Braineack

iTrader: (62)

trying to figure out why idle target and max decel load are disabled... are you not using CL idle control???!!!

stefanst

iTrader: (1)

Every time it stall it goes waaaaay lean first. You're running very lean already, like 16-17 AFR and then it goes even leaner. MAT is going up too. MAT correction is pulling even more fuel and then it dies.
Tune properly and reduce MAT correction.

Jiggerachi

I'm using PWM warmup idle control, keeping it simple while I get everything hashed out as per the MSExtra setup guide, it grays out those sections.

As for the MAT corrections, it's not tuned yet, before I posted the thread I reverted back to default map, with zero change. I know that's still not ideal but its what I had to work with at this point, thats where that MAT correction is from, the default.

I'm going through building a new project on the newest firmware. I'm using the basemap from the MSPNP Pro, its the closest to the MS3xbasemap that we could find.
MegaSquirtPNP by DIYAutoTune.com

Brain, this is your old MS3x out of your 1.6 turbo car, when I bought it from you years ago you rewired it for 1.8 and I could have swore you loaded the basemaps, but i'm betting some map somewhere still has your settings/tune that's throwing things off. I've gone through most of everything, but probably missed it. Starting out fresh from a complete new project with base maps is probably the best way to take care of that, if its the root of that problem.

Jiggerachi

Going through the settings building the new project I came across these. I could have sworn I went through and turned this off before.

Probably not the issue, but i'm better off without it for now.

Braineack

iTrader: (62)

I'm looking at your msq and log again at home.

you have VE IDLE Table turned off, and your fuel map looks completely untuned.

it really looks like you simply dont have enough fuel to idle with a/c on. you pretty much as throwing the idle valve completely open, but without extra fuel, on a system that's already idling VERY lean, you simply cant do it.

what size injectors do you have? your msq is setup for stock, but your VE values suggest maybe 600-1000cc?




Here you idle at 60% VE and 16.0:1 AFR-- RED BOX<br/><br/>then with a/c you idle at 65% VE and 17.1:1 AFR - PURPLE.




This is why I think your TPS is messed up--each time you stall, the TPS rises and holds until you crank the motor again.




Looking at your fuel map, you should rescale it for n/a, starting from the top row I'd put in 110, and work down 5 until you hit the 48kPa row and then rework your map to fit. Then I'd turn on idle VE back on and tune idle AFR with that map. The way I sent it there basically 2 rows for when a/c is running and 2 rows for when it's not and there's a clear jump in VE table values between the two for the difference in load.

but you need to get your idle AFRs back down to 14s in both conditions.

Jiggerachi

Good stuff.

I had tuned with idle VE, but we turned it off when we sort of tuned it, for simplicity.

Injectors are stock.

I think the issues at idle here are probably MAT related. When we tuned it on a relatively mild weather day, we had it idling at 14.7, didn't use AC at the time. Weather heated up and we have what we see here.

We tuned the VE map on some highway pulls, giving the car what it wanted, it ran very good, AFRs were where they needed to be. Before this VEAL screwed the map up, probably the AFR table is ****.
Again after that the weather heated up, so MAT is now screwing with things. I'm beginning to understand this phenomenon.

I see the little TPS spike. Interesting. I have a spare throttle body in the garage, I'm pretty sure it still has the TPS installed so I could swap it out.


Your last point I agree. I'm throwing my maps out the window basically and running the base maps from DIYPNP as a starting point. Map for 94-95 here. http://www.megasquirtpnp.com/mspnpp_maps.php

Base maps i'm going to start with after I flash to new firmware.