Hello VEMS folks. I just swapped over my mechanical injection 190e Cosworth to a VEMS system. I had been having trouble with a hunting idle, which I think I've gotten to tone down a bit after trying some suggestions I found here and diyautotune. I am hoping someone could check my config out and see if I'm going about this stuff the right way. There's so many options and settings that I've gotten myself lost a bit. Also wondering if there's anything in particular I should read or check out? I've been going through Cliff's Calibration sticky but haven't finished it yet.
This was the config I had with the most pronounced idle problems:
http://www.vems.hu/files/SecretMachine/MB190e2.3-16-bad%20idle%20oscillation%20but%20runs%20ok.vemscfg
This is what I've changed it to as I've researched articles on here and diyautotune more:
http://www.vems.hu/files/SecretMachine/v3.3_u006712-A-2016.05.07-20.07.09.vemscfg
Am I going the right direction?
Another big question, are my injector offset settings correct? I keep finding all kinds of different info for the Siemens Deka 630cc injectors off the VEM Shop. I don't know which are the correct ones.
Thanks in advance!
Hunting idle generally comes from incorrect VE and ignition advance in the idle areas.
Also you have not attached any logs, so impossible to see what is reacting to what.
If you have sufficient air flow for stable idle via ICV or throttle body, and a smooth IGN advance (15 deg is a good starting point) and stable lambda there should be no problem with keeping a smooth idle. Start off with lambda around 0.9, it's easier to get stable idle with richer mixture.
No one can tell if the injector offset is correct except for you if you follow the instructions given by Gunni here on this forum.
That's the thing about the injectors, I saw Gunni's spreadsheet ms numbers, they are different than what I got from Siemens and what I see on diyautotune and a few other spots for the supposed same injectors. I don't know who's to use. Figured it would be common knowledge here since it is an injector offered right on the VEMS Shop.
My lambda is steady, well until it starts oscillating. I disabled ego correction and used IAC lock to eliminate the air portion while I set the idle reference curve for the temps and the ve to match my lambda targets.
Here's a log from the first linked config.
http://www.vems.hu/files/SecretMachine/v3.3_u006712-2016.05.12-21.37.05.vemslog
I did some log runs with the other config but their on the laptop I was using to tune and not this computer. I'll grab it when as soon as I get the chance.
Quote from: secretmachine on May 19, 2016, 06:21:00 PM
My lambda is steady, well until it starts oscillating.
There isn't one instance in the log you linked where you lambda is steady... it's either off the charts rich or really lean (1.2+) when you're not playing with the throttle to keep the car running.
That said there are to spots in your log where the log shows all the ECU signals going absolutely crazy... not sure if that is on the PC side or the ECU, one instance starts @ ~6:59 the other 20:51 (pictured below):
(http://i.imgur.com/rIOKYWW.png)
What ignition system are you running? What type of crank sensor (VR?) and are all the grounds and shields connected correctly? Do you have issues with the laptop/VEMSTune software skipping/freezing?
Something is up, and I don't believe it has anything to do with injector dead times. Do you have the injector flyback connected correctly?
I meant in the most current config it is more steady, my apologies. The engine had previously been running all over the place but I have gotten it to tone down. I bought the VEMS ecu and harness already made with the flyback in place. The ground it is attached to has good continuity. I triple checked that ground as I did not want to mess the ecu up and I know how important a good ground is. I am using the Hall sensor from the VEMS Shop with a 36-1 wheel. Missing tooth is 90* from the sensor at cyl 1 TDC. Ignition is the Bosch 2x2 from Vems Shop.
My members page with the setup info - http://www.vems.hu/wiki/index.php?page=MembersPage%2FSecretmachine
The laptop I have to use is an old thing with terrible battery life. The crazy moments may be related to that. It has frozen on me while the live ve tuner was running and it has forced sleep mode and has died in the middle of log runs.
Thank you for checking all this stuff Notorious, I'll grab the latest running log from the laptop once I'm home from work. This log from the 12th doesn't have any of the latest changes I have done.
Here's a log I just ran. Cold start to warm up, then at the end, a little reduction in the ve table since it was still rich after warming up.
http://www.vems.hu/files/SecretMachine/v3.3_u006712-2016.05.19-19.32.10.vemslog
two short drive log runs
https://malwarebytes.box.com/s/vlg9g7mnb01vwfrd0hoeicwfjzcjz8zs
https://malwarebytes.box.com/s/7dgaz3o734308svq5zz9ln5l4irbb8bx
Quote from: secretmachine on May 20, 2016, 03:51:27 AM
Here's a log I just ran. Cold start to warm up, then at the end, a little reduction in the ve table since it was still rich after warming up.
http://www.vems.hu/files/SecretMachine/v3.3_u006712-2016.05.19-19.32.10.vemslog
This log looks good to me... all the way until something changes @ 11:50? From what I can see it's nothing sensor or setting related... Did you change the VE table at that point? Because after 11:50 900 RPM & ~35 kpa you're requesting 4% less VE then before where the idle was stable. And I don't see any corrections changing so I figure it must have been a VE table change at point in time of the log.
Otherwise everything in your log/config looks good to me. Idle is stable as well.
it was chilling at 13:1 afr after warming up, so i let the auto tune adjust two ve table boxes leaner to get stoich. but after that when i went to drive around, that same change ended up being too lean, 15.4:1, after driving around the block. so i put it back and then it was stoich again. every now and then the car still hunts, usually after coming to a stop and it keeps doing it for a few minutes and will sometimes stop on it's own or until i go and then stop again. confusing.
First off, autotune is pretty useless... So forget about that all together.
Smooth out your VE table so there's no major differences.
2nd bump up the idle slightly 900-1000 rpm.
3rd, completely turn off over-run fueling, that will cause you big headaches trying to solve issues if you're mapping isn't complete (which it's not).
IMO it's just a matter of properly tuning the ECU. Unless you have some sort of underlying hardware issue but I will never be able to know that from here.
i'll work on the ve table. i've been doing this by myself and i haven't been able to tune that and drive, so i used the auto map. but i'm starting to see that it is pretty useless lol i'll try to get a friend to drive the car for me while i work out the table. my idle spec is supposed to be 890rpm +/-50rpm, so I guess it wouldn't hurt to bump it up, I'll try that.
how do I turn off the fuel overrun? turn up the rpm threshold or lower the kpa?
You don´t need to have another person, what you do is log your drive and drive as steadily as possible while doing so, slow pedal changes and lengty periods at all rpms and loads, you don´t need every load or rpm but enough so that a picture of known spots can be made, rest is usually interpolation between those points.
If I am by myself I log and review while parked on the side of the road somewhere safe, then update the map and repeat, doesn´t take long to get a fuel map that is 95% there.
VE maps are smooth, especially since the the VE values are representative against manifold pressure as well so the values don´t change much.
I have used the autotune but that was on a dyno and was an experiment if anything, it takes no time at all to do the same work as autotune would do.
Quote from: secretmachine on May 20, 2016, 08:02:40 PM
i'll work on the ve table. i've been doing this by myself and i haven't been able to tune that and drive, so i used the auto map. but i'm starting to see that it is pretty useless lol i'll try to get a friend to drive the car for me while i work out the table. my idle spec is supposed to be 890rpm +/-50rpm, so I guess it wouldn't hurt to bump it up, I'll try that.
Bumping up the idle is just to help things out until you're tuned... You can't expect a car to run like stock with a tune that is only 1/4 complete. You want to eliminate as many variables as possible at the beginning.
Quotehow do I turn off the fuel overrun? turn up the rpm threshold or lower the kpa?
yes high RPM (9000) and set kpa to 0.
When a car oscillates @ idle it can trigger the overrun some times and that just amplifies the issue.
Take your time, dial in the VE table as good as you can first... you want to be able to fit your VE targets for most of the table before you start to worry about other things.
i found one of the things that was effing me up, the MAT/TPS fuel enrichment table. from 20-100 MAT i set it all to 100. this thing kept changing my ratio while i was sitting there idling trying to get the ratio right and then it would be off after a short drive.
Is there anything else i should blank out while i try to get the idle to be steady and catch well on clutch in?
Gunni, do i just read the lambda on the log myself? when i try to use the statistics thing it crashes the vemstune program.
So, I’ve got it figured out finally. I turned of everything extra that modifies fuel and timing. Idle advance off, ego off, ign retard table off, I left the fuel enrichment table on because as the car heats up from idling, it started to get lean. I got the ve table as stoich as I could around the idle boxes while cold. Then used the enrich table to get back to stoich as the car heated up and the MAT number increased.
I set P=I=D=0. Then set the three pin idle motor to 20 Hz (previously at 150 Hz as that felt the smoothest on the bench but in application didn't work well). i then tuned the idle reference curve, which was 23% DC to give me just under the target warm idle rpm, i set this to 16v stock, 890rpm.
I tuned P first, looking for the flattest waveform with the least amount of oscillation. I then tuned D. Here I was trying to get the curve to slow down and elongate out. Again going for the flattest reaction and least amount of oscillation after getting near target. Finally, I tuned I. Using this to get everything back near the target quickly after the initial oscillation. If it starts to hunt, decrease this number.
My idle is now rock solid and catches beautifully on clutch in or acc on, or power steering load. Here’s some resources I found that helped me best understand this.
http://www.vems.hu/files/SecretMachine/DTN0003_Introduction_to_PID_Control%255B1%255D.pdf
http://www.hptuners.com/forum/showthread.php?33467-Messing-with-Waste-Gate-Controls
https://www.youtube.com/watch?v=SefKQb9y_B4
You typically want to set your ref dc idle value to give you an idle 150-200 rpm HIGHER than target. And let the PID correct down instead of correcting up. That gives better stall protection.
Other than that, good technique to solve the problem, and thanks for sharing your process. Very nice.
ohhh? that process does make sense. i had come across some info stating to make the car idle just slightly below target. this is without any interaction with the closed loop idle, meaning the car would idle like this with the power disconnected from the PWM idle motor. its "limp home" mode so to speak. i'll give a slight higher ref dc a try and see what the car does. thanks for the advice!
Quotethis is without any interaction with the closed loop idle, meaning the car would idle like this with the power disconnected from the PWM idle motor
Ths is not the same since additional air is given by the idle air valve refernce position even when P=I=D=0.
D action should may not be necessary. Another general quick way for PI tuning for a first huess is to set I=0, increase P up to the value where system starts to oscillate, divide this value by 2 to set P. and finally adjust I to get zero error. If the closed loop system is not stable enough then use D. Be aware that in our case there is a sliding target (IAC target RPM) which slows the rally down to the target in order to prevent too large RPM undershoot
We're talking about different things then. I set the car mechanically to idle just below target without any vems interaction. I used an air bypass built into the bosch three pin IAC while the valve had a closed window with the connector removed. I figured that way, the car cannot stall as mechanically it has enough air to keep it going without closed or open loop idle control. Then I set the ref curve, then pid=0, then tuned the pid. The tactic you are mentioning did not work for me in particular, the tuning of P until oscillation then dividing it by 2. My P number until oscillation was very small, so I instead went for the flattest waveform, then trimmed it with I and D. This three pin is a dual solenoid and reacts very fast, so I had to slow it waaaayy down with a lot of derivative gain. Here's a screenshot of the final outcome for stable idle in my application. I also noticed my idle motor, for some strange reason, liked settings that where divisible by 3.
(http://i14.photobucket.com/albums/a314/way_eerie/idle%20setting_zpstlwyht24.png)