From one extreme to the other. AFR's

[jmsmotorsport]

Tuning a L76 VZ Thunder and I will go out and log and get a good amount of data and spread across my map, but when I apply the data to the VVE and adjust around the dips and spikes when I hit Create Co-effs then Generate VVE the parts that should have come up have now gone lower then what they were and the bits that needed to go lower slightly have gone way to far, in turn my AFR's go from one to another. What am I missing here? if I multiply the areas that need to go up by the EQ Ratio when I generate VVE area's that don't need to go up but in fact should go down also go up? Tried to attach the log but its to big.

[joecar]

Can you post some sceenshots...?

[joecar]

Can you apply the transient filter and resave the log (so it gets shorter)...?

[jmsmotorsport]

Ok transient filter what's that one joe?

[joecar]

Look at this pic, the transient filter may or may not be already defined:

19765


but, in the filter, you need to change GM.TP to GM.ETCTP.

[joecar]

( you should be applying the transient filter before you paste multiply into the VVE table )

[jmsmotorsport]

ah well this could be my problem, I have only been using the filter that I had setup for AutoVE, I will apply this one and resave

[jmsmotorsport]

The log is still 2.39mb even filtered. so here are some screen shots of EQ1976719766

So my process was to copy and fill with lables, past and multiply with lables, smooth around humps and dips by 2% then hit generate co-effs then generate VVE is that correct?

[jmsmotorsport]

So after do it that way I would copy paste and multiply in parts so from say 15kpa up to 35kpa from 400rpm up to 7000rpm then generate coefs and vve then move onto the next part 37kpa up to 65kpa and repeat, but what I noticed is when I got to the parts that needed to be adjusted by EQ 1.03 when I generate my coefs and VVE the stuff in the lower kpa areas would also increase there fore undoing what I had just reduced them by?

[jmsmotorsport]

1977019771Ok well after playing around I am now here. As you can see high rpm mid-low throttle way to lean. Mid rpm Mid throttle pretty good. WOT varies from good to rich to lean.

[jmsmotorsport]

I really cant get this E38 VVE setup, it needs to be richer at WOT and when I only multiply this area that I want to work on it creates a huge hump in the MAP I generate coeffs and VVE and the numbers go from 2488 in one of the cells after multiplying 2135 by 1.17 from my BEN map, I hit generate it goes back to 2155 its barely higher then when I started? and a cell that shows on my BEN map as 1.00 goes from 2400 to 2424 so now that will be leaner. I feel as thou every time I apply my BEN map it alters areas that are good and I'm back at square 1?

[joecar]

The log is still 2.39mb even filtered. so here are some screen shots of EQ1976719766

So my process was to copy and fill with lables, past and multiply with lables, smooth around humps and dips by 2% then hit generate co-effs then generate VVE is that correct?Yes.

Could try doing copy-with-labels (i.e. without fill) and see how it goes.

[jmsmotorsport]

Yep tried that. I decided to mash to tunes together one that was good at WOT and High Rpm Mod throttle and one that was good mid rpm mid throttle and idle. Haven't been out to log but just had it idling. Idle is at EQ .92 so I changed just the area around the idle to improve that area a bit hit generate coeffs and VVE and this changed everywhere from 35kpa up to 105kpa through out the rev range so now my WOT area numbers are lower then where they need to be. What's the reason it changes everything so far away so drastically? I find this makes it impossible to get every area dialled in to where it should be.

[joecar]

So after do it that way I would copy paste and multiply in parts so from say 15kpa up to 35kpa from 400rpm up to 7000rpm then generate coefs and vve then move onto the next part 37kpa up to 65kpa and repeat, but what I noticed is when I got to the parts that needed to be adjusted by EQ 1.03 when I generate my coefs and VVE the stuff in the lower kpa areas would also increase there fore undoing what I had just reduced them by?

Yep tried that. I decided to mash to tunes together one that was good at WOT and High Rpm Mod throttle and one that was good mid rpm mid throttle and idle. Haven't been out to log but just had it idling. Idle is at EQ .92 so I changed just the area around the idle to improve that area a bit hit generate coeffs and VVE and this changed everywhere from 35kpa up to 105kpa through out the rev range so now my WOT area numbers are lower then where they need to be. What's the reason it changes everything so far away so drastically? I find this makes it impossible to get every area dialled in to where it should be.

Try this: first do all the parts, and then at the end do generate coeffs once.

[jmsmotorsport]

I have tried both ways. Change all the areas then generate or do bits at a time and generate after each change. Still alters areas that were good to much making them out of whack again.

[joecar]

Hmmm, so the generate coeffs changes to much of the surrounding area...

[jmsmotorsport]

Seems that way, are you able to apply my logged data to the tune and upload it so I can see how you would have done it? Then I can compare what numbers I'm getting to yours.

[joecar]

When you generate coeffs, do the surround areas in the VVE table change immediately...?

Post some screenshots...

[jmsmotorsport]

I will take some screen shots of what I'm doing as well. Pretty much trying to just dial in WOT for now as everywhere else is good enough that the trims will take care of it during OL cruise.

[joecar]

Greenshot may help with the screenshot (lets you add arrows and notations).

[jmsmotorsport]

Is that a freeware program?

[joecar]

Yes.

Google getgreenshot

[jmsmotorsport]

19773197751977619774Ok great program, that will come in handy. I have tried to explain as good as I can what I'm trying to show.

[joecar]

Ok great program, that will come in handy. I have tried to explain as good as I can what I'm trying to show.Greenshot does help to quickly take a screenshot and mark it up... your marked up screenshots show me exactly what you did...

I don't know the answer yet for preventing those other cells from changing when you generate coeffs.

I think they have to change due to the polynomial for that zone mapping to all the cells in that zone...

I'm not sure, let me get some more opinions.

[joecar]

You could try to redefine the zones.

[jmsmotorsport]

Thanks joe, my zones are based off swingtans tutorial, I read that some people use the paste with corrections method rather then multiplying this seemed to keep the VVE much smoother on pasting but I'm guessing this makes very minor corrections. The weather here has just turned to shit so won't be able to get out logging.

[joecar]

Paste-with-corrections: ok, I'll look into this.

[jmsmotorsport]

Well joe I think I have found my new method, by using paste corrections 50% it gives a much smoother alteration to the VVE rather then multiplying and trying to correct a big hump therefore when you generate coeffs and VVE it doesn't want to alter everywhere else to keep its shape. If I had a dyno and was able to hit every cell across the board you could get it perfect, I tried 25% but this didn't seem to alter it enough even thou with the time you could go this path so it doesn't upset other areas to much. I will do some screen grabs.

[jmsmotorsport]

here we go some actual comparison, the final result leaves altered numbers similar one generated but hasn't altered other areas as drastically. I think my VVE needs the high RPM lifted up to improve its overall shape also. Not sure the best way to go about this yet as in what sort of numbers it would need, do I just copy the last row at around 7000 rpm and paste this over all rpms after that area making it just flat?

After Paste Multiply: 1979719798

Then Generate: 19799

Now using paste corrections 50%: 1980019801

Then Generate: 19802

[joecar]

Ok, thanks, paste-corrections-50% seems to be the way to go.

Yes, paste the the 7000 rpm row to flatten the rest of the table.

[jmsmotorsport]

Yep I will be doing that before I start next time to get the shape nicer from the start, this might also help stop the coeffs generation going wild across the map.