VVE Help

[etmotorsports]

New at tuning vve cannot seem to get this to be stable at idle. Here are the logs i have followed swingtan vve setup guide.
9812

9813

[Taz]

The E38 / E67 idle tuning thread below is long, but very comprehensive. Just pointing you in the right direction - I work primarily with the Gen III - definitely not a Gen IV expert.

http://forum.efilive.com/showthread.php?9740-E38-idle-tuning


Regards,
Taz

[etmotorsports]

i have seen the idle tuning but i was hoping to get it more consistent 1st with vve tuning, i can get it to idle at 18:1 or around 12:1 and maintain that a/f but no where in between. Guess i need to see if there is anything in there

[swingtan]

Some specs on the motor and vehicle would help. Auto /Maual, cam, intake, exhaust etc.

Simon.

[etmotorsports]

2010 Camaro automatic, has cam not sure what size around .58-.59 lift, single 72mm turbo, lsx intake, zr1 map sensor, was tuned off maf trying to get it to drive better and be more consistent a/f

[swingtan]

Well, that explains the VVE shape a little. try the following...


Reset the ECM between idle changes. Remember that idle and fuel trims stay between flashes so you need to clear any learned data.
Always log Inlet Valve Temp! Always!
B1845: To reduce the effects of learned idle data, set all cells from -64 to +64 RPM to "0"
B1650: reduce by 25%
B1829: set all cells under 800rpm to 9gm/Sec. set them all to the same as the 800rpm cell. You can even increase the values a little to help with the idle correction.
B5104: increase idle spark to around 21' to 23'
B5128 to B5131 look all wrong. Spark has a much bigger influence on idle control than throttle. These tables are zeroed up to 96RPM error which is bad! Actually, it's probably these tables causing most of the problem. As a quick fix, linear fill from 0 to 112 RPM with the current tables, then fine tune. try this first and see how it goes.
B1821: increase the minimum idle speed. This helps to get the idle right, then once it's stable, lower it and fine tune.


See how that helps.

Simon.

[etmotorsports]

Thank you for the advice try all of that.

I have another stupid question does the higher the # in the vve table mean more fuel? From what i can tell it does but not sure.

[ringram]

Yep basically, but the number is fairly meaningless in itself

[JAY4SPEED]

Well, that explains the VVE shape a little. try the following...


Reset the ECM between idle changes. Remember that idle and fuel trims stay between flashes so you need to clear any learned data.
Always log Inlet Valve Temp! Always!
B1845: To reduce the effects of learned idle data, set all cells from -64 to +64 RPM to "0"
B1650: reduce by 25%
B1829: set all cells under 800rpm to 9gm/Sec. set them all to the same as the 800rpm cell. You can even increase the values a little to help with the idle correction.
B5104: increase idle spark to around 21' to 23'
B5128 to B5131 look all wrong. Spark has a much bigger influence on idle control than throttle. These tables are zeroed up to 96RPM error which is bad! Actually, it's probably these tables causing most of the problem. As a quick fix, linear fill from 0 to 112 RPM with the current tables, then fine tune. try this first and see how it goes.
B1821: increase the minimum idle speed. This helps to get the idle right, then once it's stable, lower it and fine tune.


See how that helps.

Simon.

Simon, by "reset the ECM" do you mean to clear trims via DVT? Does that clear idle trims as well as regular trims? Or is there another method of resetting the ECM that you are referring to?

[etmotorsports]

I have another question if someone know the answer or can help me out. When i paste and multiply the ben map data it increase the or decreased the vve table in those spots needed but when you generate and make vve it reduces those by a drastic amount and sometimes it even makes the the opposite of what they "should be". IE if a cell is at 1566 and the data has a .91 then you get 1425 there but after generation it is only 1550 or so and i have even seen where it would go the opposite.

I know about averageing around the cell and all that does help but it seems the next log it needs another .91 or even less .85. I am wondereing if there is something i am skiping?

[etmotorsports]

It idles much better swingtan that helped alot i did everything you had up there, the only thing now is in gear it idles higher and with less vacuum, but i will try to work that out. thanks again.

[swingtan]

Simon, by "reset the ECM" do you mean to clear trims via DVT? Does that clear idle trims as well as regular trims? Or is there another method of resetting the ECM that you are referring to?

That will do the same job. I tend to use BBL/BBF all the time now so it's easier to just pull the ECM fuse.


I have another question if someone know the answer or can help me out. When i paste and multiply the ben map data it increase the or decreased the vve table in those spots needed but when you generate and make vve it reduces those by a drastic amount and sometimes it even makes the the opposite of what they "should be". IE if a cell is at 1566 and the data has a .91 then you get 1425 there but after generation it is only 1550 or so and i have even seen where it would go the opposite.

I know about averageing around the cell and all that does help but it seems the next log it needs another .91 or even less .85. I am wondereing if there is something i am skiping?

When I looked at your tune, the VVE is probably way off in the higher RPM range. It hink you need to reset the VVE zones to make tuning the VVE simpler. As it stands the zones simply won't match what the engine needs. I'm going to guess that AFM/DOD is not used, so try these settings...



B8020: set to 1300, 2700, 4100, 5500
B8021: set to 30, 60, 90, 120, 150 ( I'm guessing a bit here as I'm trying to factor in the turbo boost )
Save the file, then reopen it. You must reopen the saved file.
Open a new copy of tune tool and open the old tune.
Copy the VVE table "with lables"
Paste the values in to the VVE table of the NEW tune.
Generate coefficients and gen the VVE.
Apply changes as per normal.


This should help to stop adjacent cells from affecting each other as much.

Simon

[etmotorsports]

I saw that in the tutorial but it seemed to mess up the table even after cut and paste but i will try it again the rpm ranges are important i know but still can't wrap my tiny head around it completely. I can't believe how far off the table is above 3k rpms.

[swingtan]

In your tune, the "high speed mode" settings are at 1500 RPM, so the VVE would not be used when the engine RPM is over this value. The VVE can be any value above the settings of B8024 and B8025 and the MAF will give the info required. I've had this setting down at 600RPM to see the difference and while it works OK, I think the VVE still helps in places. Getting the car to run nicely around town is where all the work is...

Good to hear that the settings are helping. Some fine tuning of the values should help get it spot on.

Simon.

[etmotorsports]

Should i go and raise all the values off of negative and the ones under boost up?

[swingtan]

For the VVE, I try and make it "sensible" but allow the coefficients room to more if you get what I mean. It's pretty ovious that at no time will a real VE table end up with negative air flow, so I've increased all those cells to what is "likely" to be correct values. Ov course I'm never going to actually see 8000 RPM and a MAP of 15, but I know it's going to be well over "0". I just looked at the shape of the VVE and continued it on, fudging the values to try and get the VVE correct where it really counts. Altering those cells up top will alter the shape of the high RPM VVE zones, so you can fudge the value up top to alter the shape of the curve on the lower cells of the zone.

For boost, I'd again "continue" the shape of the VVE up into those cells before carefully testing with the WB. You can usually see trends happening pretty fast and you should have a fair idea of what the VVE should look like in terms of shape. This is where knowing the characteristics of the motor really help out. If you have a good idea of where peak torque is, you know that's where the VVE should also peak. If it's a wide torque curve, you know the VVE should also have a wide peak. If you see the mixtures leaning out at 3000 RPM and the VVE is fairly flat from 3000 through 6000, you know you'll need to add more to the cells all the way up through 6000. When setting up the VVE from scratch, don't be afraid to "guess" what the VVE should look like and manually set that up first. Then use logging to fine tune it, just don't run the engine lean under load.....

Check out the images of the stock VVE on page 11 of my guide, and compare it to the one on page 12. You can see that the values have changed a lot and some of those changes were me just "guessing" the values and deliberately over correcting to allow the coefficients to bring it back into line.

Simon.

[etmotorsports]

I knew the map wasn't like a regular ve map, and it made no sense to me but after seeing the wideband and what it says its alot like one except for the zones and curves. I will modify the zones like you mentioned and the modify the map slightly to help on the high rpms high loads and then go from there it seems like i wasted 2-3 hours just seeing how it acts to changes.

[etmotorsports]

I started completely over with the VVE table after the zone corrections that you suggested swingtan basically making it like i thought it should look based on the af data of previous runs, i am suprised how close it was compared to where it was before 1st is how i started and 2nd is where i am at so far, just for someone to take a look at if they want.


9827

9828

[swingtan]

That's looking much better. I'd still drop the values over the peak torque RPM, as the engine will loose efficiency past that point and require less air. as I mentioned previously, there "should" be a hump at peak torque, with the values rolling off above and below that point. A few runs up into the 6500RPM range should clear that up though. Just look at what the BEN data does to the VVE and carry on the shape of the curve where you have no data. This holds true for very low RPM cells as well as very high.

Simon

[etmotorsports]

The wideband is jumping all around i know it would be best to kinda average it but not sure what is going on look at the log under boost. I know it is rich but the lean spikes are throwing me off, i see them too at part throttle too going rich and lean for a small bit.

9831

9832

[swingtan]

Because you are running SD, the dynamics will be all out, so some rich lean swings will be expected around throttle movement areas, as well as when engine load changes causing changes in airflow. I'd start applying the BEN data from those logs though and get the VVE pretty close. Then I'd go for a good drive of at least 30 to 45 min, preferably 90+minutes with different driving conditions. I'd use the road trip to fine tune the VVE as the long duration will help to average out random data. After all that, if you want to re-enable the MAF I'd do that and dial the MAF tables in. you should get ver accurate fueling then.

Simon

[etmotorsports]

From what you said i would guess these ecm's don't do well in speed density all the time? Under load where it is going rick lean the ben data says take out .54 that seem too much?

[swingtan]

It's probably random data creeping in. As you say, it's too much to pull in one hit, thought he co-efficients will smooth it out. The trick is really to do a lot of logging to get the BEN data. Short runs can be used, but they really need some practice to weed out the bad data.

SD can work OK in the E38's, especially if it has an auto. In a manual where you are off and on the throttle all the time between gears, then dynamics get very critical. To be honest, I'm not even sure if you can really get the dynamics optimised for all conditions in SD mode. I've played a lot, but eventually went back to the MAF as it made things so much easier. If I needed more than 512gm/Sec airflow though, I'd stick with SD as my OS pegs out at 512.


Simon.

[etmotorsports]

Ok i see, i need to rig up the wideband for the road and off the dyno for a bit, i have watched the data on the short runs and see how it creeps and changes the cells i will take your advice you have been dead on so far. The last map i started over again, it was straight across at and got the curves with just a couple of short runs like you saw up there. I found it easier too just start over again knowing how the table makes the a/f act and it was good i could drive the car actually.

[ringram]

B1650: reduce by 25%

Simon.

A convert I see. Nice.

[swingtan]

Yes, I like the effect ;)

[etmotorsports]

Have you ever noticed when you save the file the vve table changes? Just saw that and that explains when i am not always seeing the change i thought i should.

[ScarabEpic22]

Yes, the software recomputes the values when you save it. Makes it impossible to get a 100% perfect VVE but the changes are so minute in the whole scheme of things its not that big of a deal.

[etmotorsports]

The numbers always go less than what you want odd. Just one of those things to help notice when you start wow. GM sure made this more difficult

[swingtan]

When I was getting my head around the VVE, I saw this. My technique was to gen the co-efficients and then save before re-genning the VVE.

Sent from my HTC Desire

[etmotorsports]

Has anyone found where and what the deal is with the FTC Cell #?

[etmotorsports]

Update,

Turned the MAF back on and enable STFT, made a full pull at 7 PSI, a/f was still rich but only had 9.5 degrees of timing and made the same power as the other tune that had 19 degrees. Car drove just like a stock one on the dyno at lease, and honestly the VVE is close but still off by 15-20 perecent in places but that has to do with where the front o2 sensors are mounted on the turbo system.

Swingtan i don't care what they say about you, you are not crazy. LOL

[etmotorsports]

When the maf is enabled and you need to pull out fuel in the higher rpms should you change the MAF or the VVE or both?

[etmotorsports]

Make small changes to the MAF and for some reason spark is being pulled out under throttle see log. I can figure it out a/f should be fine leaner than other pulls but not by much and there is no knock.

9848

9849

[swingtan]

Swingtan i don't care what they say about you, you are not crazy. LOL

Bwahahahaha, I have you fooled...... :p


When the maf is enabled and you need to pull out fuel in the higher rpms should you change the MAF or the VVE or both?

You can just use the MAF. Anything over {B8024}: High Speed Mode RPM Enter, will result in the MAF being the sole sensor used for airflow calculations. In the original tune, it was set to around 1500RPM (I think, from memory) so it would have always been used once the car was driving. The VVE would only have been used to (possibly) help with idle fueling and reverberation. I'm not sure why it was set at that level. If it's still at 1500, then it'll be used solely from there up.

Simon.

[etmotorsports]

Swingtan

Any idea as to what cause's spark to be pulled of fall off?

[swingtan]

Under "Torque Control" - "Parameters" set all [Yes/No] options to "No" apart from {B0517} and {B2523}. See if that helps.

simon

[etmotorsports]

Also i got the P1101, turned it off on mil and type but still reduces ill try the torque control

[swingtan]

RE: P1101, try bumping {C0803}, {C0804}, {C0805} and {C0806} by around 10%, maybe even 20%. If you don;t want any ECM correlation checks done you can maxt them out. Personally I like to set them in case something goes wrong with the ETC.

simon

[etmotorsports]

Ok here is a log and a/f move around so much i am not sure if it is rich, lean or dead on. Tell me what you think

9858