Program to generate VE equations

[PurplePiss]

Can someone explain how to use this program? I want to tune my VE table but I don't know where to start.

[Bill@HPTuners]

Can someone explain how to use this program? I want to tune my VE table but I don't know where to start.

You already have a VE table, this program won't do much of anything for you.

[MNR-0]

Good job Bluecat! Got a Holden VE E38 which I might just give this a go.

[gofish7]

I downloaded these files but can not get them to open, getting error: "MSVBVM50.DLL was not found." Do I have to have MS Visual Basic in order to read these files?

[TiredGXP]

you can get it at the microsoft download site

http://support.microsoft.com/kb/180071

just click on the MSVBVM50.exe link to install. Bluecat's program will run fine, even on Vista, once you have the dll

[deanm11]

I used the tool last night to tweak the VE on my 08 LS3 vette. I had relatively minor changes from what I had last done in excel/xlfit with carneb's solutuion. It seems to do the job just fine. I found I needed to use it with the box unchecked that allows it to pick its own boundaries. When checked, I got some large variances in an area (~2000 rpm, light load) where I was trying to touch up). Two questions- 1) is there a way to have it solve all the cells totally independently? (I almost thought it got too aggressive in trying to keep things smooth) 2) I used 'carneb format'. Does this just sppecify the input table cell structure? So you can't go with slightly different breakpoints by just laying out in the input file?

Thanks again for the great tool.

[Bluecat]

1. It used to have another slider bar for edge smoothness, it would change the goal from max matching of the zone to best zone boundry transition. I took it out after I got it doing both jobs well. For the most part, if you feed a smooth, realistic looking VE table, it can usually approximate every point within 1.5 VE%. But that can be a fair amount of error in the smaller number regions. I can add that feature back if it would help.

The auto boundries seems to do ok most of the time, it just looks for areas where the table makes sharp turns and sets those up as the zone edges so the program has less radius to match. If you leave it unchecked it just uses the numbers that are in the current Equation.txt. So if you want to change it manually you can do it there.

2. Yes it just specifies the input/output structure. You can modify them in the Table_Dim.txt. Just don't exceed 50 elements per axis and make sure you keep the spacing and periods intact. Remeber some tables are inverted axis compared to others. The Table_Dim.txt dosen't specify that, I just have it hard coded in the program. I might change that so the end user can make as many table formats as they want in the next update.

[Bluecat]

Updated program again...

Added slider bar for zone edge weighting.

Changed Table_Dim.txt format to support user defined formats. The "N" or "I" refers to normal or invereted in respect to x and y axis (LS1 VE = normal). File now supports up to 16 formats with 50x50 resolution.

Added auto format detection for table reading. If checked it auto types the file based off the text string in the first cell. Also detects VE or RAW base off of the size of the numbers in the table.

Swaped axis on the Carneb format after I saw that he uses rpm on the x axis. It now matchs the layout on his E38_E67 spread sheet.

[gofish7]

Thanks TiredGXP!

[deanm11]

Attached is my latest file... LS3 Vette with headers and CAI, ported manifold. Tuned by wideband, not perfect, but decent. What makes it difficult to compare, however, is how the bias table is applied- I have reduced my bias values by 1/2, as with the stock values, the VE seemed way too high.. Need some warm weather to get an idea if the bias makes total sense.

[deanm11]

And much appreciate, bluecat, for the tool ! Would be lost without it.

[Bluecat]

Lookin good. I think its funny how much these new motors (even ls2's) have a relative VE thats so flat through such a large kpa spread.

I don't know if your doing this or not, but if you export your formuals to a regular VE table, say LS1 format. You can then temporarily paste the table into hptuners where you can utilize the smoothing and interpolation features to really beat down on the table and make it nice. Then rebuild the formuals with the loved on table you made in hptuners. You can still use the big table for working with it and logging, but that way you don't loose the power of the hptuners graphical interface.

[deanm11]

Lookin good. I think its funny how much these new motors (even ls2's) have a relative VE thats so flat through such a large kpa spread.

I don't know if your doing this or not, but if you export your formuals to a regular VE table, say LS1 format. You can then temporarily paste the table into hptuners where you can utilize the smoothing and interpolation features to really beat down on the table and make it nice. Then rebuild the formuals with the loved on table you made in hptuners. You can still use the big table for working with it and logging, but that way you don't loose the power of the hptuners graphical interface.

Thanks for the idea- had been doing the smoothing and interpolation in excel by hand (and some formulas in problemeatic areas).

[ramair96ws6]

Will this work on an LS7? I tried but the numbers look completely wrong


Thanks
Steve

[ramair96ws6]

Never mind i figured it out!!! Awesome tool I hope this works i get to try it tomorrow on a 454 lsx 2006 zo6

[Bluecat]

Should work fine for you.

One thing I have noticed is that after you move the numbers back to the pcm you loose a little bit of presicion. I don't know if a limitation of how the numbers are saved in the pcm, or HPtuners. But sometimes the numbers change slightly from when they are pasted into the tune compared to if you read them back out. Like on RPM^2, My program can take it 7 places to the right of the decimal (and needs to). Hptuners truncates it down to 5. That can have a big impact.

Can someone from HPtuners comment on if its a PCM or editor limitation?

[BBA]

I've seen that number shift as well, just copying a table in from excel it will change the numbers slightly. Copying out of HPT always works, but it's the other way, copying in that bothers me. I have spent a lot of time manually correcting what HPT would not copy in correctly.

[carneb]

I've also noticed that when in HPTuners copying the standard coefficents from my standard tune into the tables of my custom tune, then doing a compare between the standard tune and my custom tune with standard coefficents, HPTuners tells me there is a difference between the tables in both tunes. When you look at the numbers in individual cells and compare them they look identical.

[Bill@HPTuners]

I've also noticed that when in HPTuners copying the standard coefficents from my standard tune into the tables of my custom tune, then doing a compare between the standard tune and my custom tune with standard coefficents, HPTuners tells me there is a difference between the tables in both tunes. When you look at the numbers in individual cells and compare them they look identical.

extend the decimal precision & you'll probably see a minimal difference ie .00001 or something.

[redhardsupra]

yea, except with the parametric VE a 0.00001 difference can be huge, depends which parameter it works against