1999 Miata tuning help

[sep]

Intro story: friend gave me his bone-stock 99NB (10th anniversary) for an MS2enhanced and WBO2 install.

Install went perfectly fine after we figured the DB15 pinout. Wideband gauge (AEM failsafe 30-4900) and tunerstudio agree to the decimal on AFR, so must've done something right.

Started up on the first try, needed some small VE adjustments for a decent idle. Had idle control switched off for now since I'm lazy and wanted a decent baseline first. The very second I enable PWM CL idle control it's all over the place: revs to ~3k back to ~1.1k before I get fed up and shut it down. No amount of messing with idle settings clear this up.

More info:

bone-stock 1999 NB 10th anniversary model, no AC, just PS, OEM 240cc/min injectors, aftermarket muffler and cone filter on 90 degree elbow of the TB with the IAT sensor wired in.

A spring install delayed me a bit. Cold start is decent, starts on first crank and doesn't die on me so that's good.

Now, whatever I do I can't seem to enter CL idle. Regardless of settings the idle valve puts out 34% PWM and hovers around 1100rpm hot. I've read braineack's quick 'n dirty guide and also hornetball's explanation on entering CL idle. My settings look reasonable (to me at least) but no dice.

When I enter the idle valve test mode it reacts to PWM changes (more PWM increases RPM). Changing I term in PID from 50 all the way to 150 has zero effect (P and D are 0 for now). What am I missing here?

I seem to have some TPS jitter (-1 - 1%) so I raised the TPSdot to 3.5 or something.

Took a short drive, bucking all over the place due to lean conditions so there's lot of VE tuning to do, but I feel like I should get this idle sorted out first.

Any input?

Another update, seems to enter closed loop idle now. Slowly increasing I-term decreases RPM closer to target while lowering idle PWM so I'm getting there.
Screenshot shows while highbeams, rear window defrost and heater (setting 2) are on. This is the first time I've noticed the CL Idle indicator light up in green.

CL idle succes.jpg

Only issue I'm running into here is coming to a stop still stalls the car. Do I need more I-term here or are my dashpot settings off? I still need to find out how to tune while also driving.

Took a small drive, was still bucking like I've never driven stick before because it was lean all over the place. Pulled over and added fuel across the board (~20 points in cruise areas, fattened accel enrichments). The second I commence VEAL again it starts pulling fuel again until the bucking comes back. Why is it fighting me?

Remember what I said about entering CL idle? Forget it, didn't happen today for whatever reason. Idle PWM is static at 34%, which is the lowest bin of the cranking duty PWM table.

VEAL still pulls fuel out of the cruising and light accel area of my VE table. I fatten it up manually and after 20 mins of driving it ends up taking all the fuel out again leaving me with a bogging slow POS of a car. Do I resort to the pre-VEAL era and drive, log, analyze in MLV and import new fuel table (rinse, repeat) or set it on fire? Everytime I feel like there's progress, I'm taking two steps back the next day.


Cliffs: Entering CL idle is intermittent, works fine one moment and is all over the place the next. VEAL keeps pulling fuel even though it's already super lean (~18 AFR). Manually fattening the affected cells just leads to VEAL pulling fuel again.

NB MS2enhanced 31_10_13.msq

2013-10-29_14.40.08.msl.zip

[sep]

What did I do wrong, too much info, not enough logs? Any help is greatly appreciated

[Matt Cramer]

Just a bit of a harder question than normal. You might want to try a higher PID lockout rpmdot threshold - 90 rpm/sec is a bit low.

[nismoautoxr]

drop your pid disable RPMdot from 950 to 400 or so. Decrease your I term to about 60. increase your D term to 15. I would think that your valve frequency multiplier would be more like 3-4 in stead of 1

[slow_hemi6]

Where is VEAL pulling the fuel? If it is from idle and just off idle then yes it does that. You have to set up filters or just manually set the cells and lock them out. I am pretty sure it pulls the fuel out of these areas when you decel. If you watch the old ball and tail displays behaviour with the fuel map displayed you will see on decel that it often goes right through the idle and off idle cells. So if VEAL has a free hand it will also alter the fueling in this situation.

[sep]

drop your pid disable RPMdot from 950 to 400 or so. Decrease your I term to about 60. increase your D term to 15. I would think that your valve frequency multiplier would be more like 3-4 in stead of 1

I was told that the enhanced MS2 has an internal multiplier. I tried 16 before by manually editing the .msq file before Reverant told me to leave it at 1 because of the hardware multiplier.

Will try lowering RPMdot and PID terms.

Where is VEAL pulling the fuel? If it is from idle and just off idle then yes it does that. You have to set up filters or just manually set the cells and lock them out. I am pretty sure it pulls the fuel out of these areas when you decel. If you watch the old ball and tail displays behaviour with the fuel map displayed you will see on decel that it often goes right through the idle and off idle cells. So if VEAL has a free hand it will also alter the fueling in this situation.

It's in the light accel areas (2-3k rpm, ~60kPa ish). I have the VEAL filters to ignore anyting <1500 rpm, <35 kPa and <70 degrees Celsius watertemp.

I've followed the ball while VE was correcting and I seem to hit those areas quite a lot and despite it being waaay off target, fuel is still being pulled. I might set up some more extensive filters, Dimitris suggested I start with these: You also need to set max TPS to 40% or so for low load areas (ie less than 70kPa or so), then tps > 40 and tps < 80% for the 60-90kPa, then finally TPS > 80% for the map > 90kPa areas.

Thanks guys, much appreciated.

[nismoautoxr]

reverant and you may know better on the multiplier because I viewed your msq with an example project without having the ini present for your particular firmware but I was under the impression that the frequency multiplier was 30.5 hz so in my mind and in my past experience a frequency of around 150hz has always worked pretty good for me on Japanese cars (Toyotas and nissans) that Ive tuned. That being said ...until you get a lot of the rest of the tune under control it might be in your best interest to use PWM warmup instead so you don't have that to contend with right now, that being said, The PID disable RPMdot being set as high as you have it (according to the tool tip next to it) can cause sudden stalls and suggests values of 200-400 as a beginning point.

[sep]

Great success (moderate maybe, but success ) today. Made some changes and it's already heaps better than I had before. Started after 3 cranks and kept idling/warming up following my rpm targets. I do seem to have some intermittent TP AE kicking in (indicator lighting up) every now and then even at idle and foot completely off the throttle. I see the TPS gauge moving from -1 to 0% every now and then but it shouldn't activate TP AE right? I've even raised the TP threshold but it's still kicking in every now and then.

Another small drive and it was ok. I have no idea how to implement all the extra VEAL filters like in my previous post. VEAL did seem to pull less fuel out of the VE table and it was smoother. 2-3k rpm and ~70-80 kPa are still bucking although not as bad as before I think.

One annoying thing I noticed is that is has a tendency to stall while I'm parking. I turn in to reverse into my spot and it stalls. Fire it up, park and stall again. Stoplights etc are no problem, what could cause this?

All this driving was with low beams, rear defrost and heater on, still dips but noticeable less than before.

Log and most recent tune file in attachment.

NB MS2enhanced 05_11_13.msq

hmm, log is too big even after zipping. How do I shrink it? 7zip can compress it even further but isn't allowed here.

[nismoautoxr]

There is a utility in MLV that will edit the log down to just what you want to show. Its been a while since I did it so either someone will chime in here or you can ask in the megalog viewer section of the forum.

If you have not already done so...you should work on tuning idle advance . This has a very good affect on sudden loads such as steering and electrical loads .

[sep]

Yeah I found an option in MLV to cut before/after the cursor. I cut off some parts that were just idling and suddenly the file grew 2 MB (6 to ~ 8 MB)

I'll try dialling in some idle advance and see where that gets me.

Edit: put the log up on dropbox, let me know if this works https://www.dropbox.com/s/01bct830njxtl ... .29.01.msl

[sep]

Starting could go a little quicker, but idle was dead-on today, despite some small dips when the idle valve opens and it goes super lean for a while. Seems to recover ok and hold it's target of ~950 rpm hot. Occasional off-throttle dips to ~600rpm but no stalls. Might try a bit more dashpot and fine-tune idle advance.

I've checked the log I made today and saw some lean spikes (again) with low TP and high(er) MAP, I've added some fuel to those areas and will log again tomorrow.

NB MS2enhanced 06_11_13.msq

2013-11-06_16_modified.zip

How do I go about adding the filters for VEAL for these conditions? TPS 0-40% ; 35-70kPa, TPS 40-80%; 60-90Kpa and TPS 80-100%; 90 > kPa?

Thanks as always

[nismoautoxr]

How do I go about adding the filters for VEAL for these conditions? TPS 0-40% ; 35-70kPa, TPS 40-80%; 60-90Kpa and TPS 80-100%; 90 > kPa?


Tuner studio and MLV section is the sure fire place to get a detailed description of how to do this. Phil will be more than happy to walk you through it. Short answer is I don't remember LOL

[sep]

Filters work now, had another go today. I'm still not 100% satisfied with the VEAL robbing me of fuel even after filtering.

VEAL tuning.jpg

Cell weighting seems good to me, still there's a lot of fuel being removed from most cells. You can see in the datalog where it pegs at ~18 AFR before it finally takes off. It holds 12.5 AFR at WOT all the way to 7k rpm though. Driveability is all I'm working on now.

[sep]

Anyone? Tried some today, idle was oscillating bad. Just as I hoped at least that was fixed, this acted up.

[sep]

More progress. Starting/idle is less fucky now. Lowered cranking PW some and oscillations are waaay less. Still have some idle dipping when parking and slow driving and constantly engaging and disengaging the clutch.

Lean spots in cruise and light accel are less too, though VEAL still likes to pull out some fuel. Feels like I'm getting close now though

Clicky here for video of idle/warm-up

Log is up on dropbox since it's huge and I fail at shrinking it :loser:

Logfile

[david_syd_au]

Looking at your tune I notice that in "Basic/Load Settings -> General Settings" the "Stoichiometric AFR" is 13.0 rather than 14.7 as recommended.
Is there a reason for that change?

[imaplt]

Any chance I can get your latest msq? I saw your video and I am jealous of your idle. I cannot seem to get mine to idle that well!

Thanks