ULTM8Z

Just what we need... yet ANOTHER automated VE tuner...

For whatever reason, I couldn't get the readily available ones to work on my machine.... so I created my own.

Basically need a Tunerpro history table (BLM vs RPM and MAP) and your existing VE table. Copy and paste them into this tool and it'll output an updated VE table to copy and paste back into your bin.

It's up on my DIY tuning tools web page now for free download. See item #6.

http://ultm8z.com/DIY%20Tuning.htm

Feel free to e-mail me with questions, if any. ultm8z@yahoo.com

ULTM8Z

So I decided to put this thing to the test...

I inputted the VE tables from the LT1 ($DA3 mask). BLM's were pretty far off. Made one run driving the car into work with datalogging on TunerproRT.

Came home, ran the history tables through my VE tuner, got the new VE tables and put them into the calibration.

Made another run today and boom, EVERYTHING is pretty much between 127 and 132!

Now, the only caveat is there are momentary data drop outs during logging which give me false BLM's (I can see momentary ALDL disconnects while logging)... spits out BLM numbers like 29, 274, etc, in various cells. As of now, I'm manually "smoothing" data by looking at the adjacent cells and interpolating.

But want figure out how to get Tunerpro to "clamp" the BLMs readouts to between the standard 108 and 160 so as to avoid the manual post-processing work.

Other than that though, this is hugely successful!!

RBob

iTrader: (1)

The ECM checksums the data packets, Tuner Pro should check that it is correct. If not then toss the packet.

RBob.

The_Punisher454

Rather than clamp down the out of range data, maybe you might want to consider tossing out those samples completely. I think clamped corrupted readings will be bad for your averages.
Just a thought.

ULTM8Z

Yeah, that's pretty much what I do. It's a manual process, but has to be done I guess.

84Elky

Another condition to consider is time in DFCO and where BPW=0 because in those instances, BLMs = 128. IIRC, TPro does not account for this in its history file. So that leads to a set of samples with where BLM=128 where no fuel is being provided yielding misleading results. My experience is that a 10-15 minute log will have 2-4% of its samples where BPW=0 and thus BLM=128. I believe those samples should omitted when tuning. The attached application both considers and eliminates these samples so differences can be seen, in addition to providing much more tuning related information. See the "VE CORRECTION" Tab.

The_Punisher454

wow, thats impressive.