Virtual Torque Table Adjustment

[[email protected]]

2006 Corvette LS2, 6L80E Auto, Yank 2800 Stall, E85 Flex Fuel, E-Force Supercharger, 10 lbs. of Boost, PRC heads, 227/234/114 600 lift cam, Johnson 2116LSR lifters, Kooks Long Tubes, Corsa Sport Exhaust, 7.5 gal PNR ice tank. Just completed a Dyno session which provided me with accurate torque data/RPM for both Airmass and MAP. My Virtual Torque Tables have never been changed from stock. The stock Virtual Torque Table values are 25% higher than the torque values obtained from the Dyno. I am conflicted as to weather I should update the tables, lowering the torque values by 25% or just leave the tables as they are. After adding the CAM, Heads and Lifters, and realizing a 60HP gain (currently 691hp at 6200 rpm) the transmission will not always shift as scheduled, hitting the rev limiter. After some research and reading, it appears that not adjusting the Virtual Torque tables to accurately reflect actual torque being realized, may impact the 6L80E's ability to shift as demanded. Some tuners say to leave the tables alone while others say these tables should be changed any time your tune changes. Should the Virtual Torque Tables be changed to accurately reflect actual torque or is this unnecessary.

[hjtrbo]

25% higher than dyno is fine. VT is engine torque not rear wheel.

[[email protected]]

Thanks hjtrbo, that helps clear things up for me.

[hjtrbo]

no problems. At 25% high plus / minus some dyno variation it sounds like you're spot on with the stock torque values considering we would normally add 10% for the trans. In your case you're good to leave alone.

For shifting, its easiest to back out the speed and rpm a fair bit then walk it back up. After a 2 logs of shifts you will get an idea of the lead time the trans needs, that 3rd adjustment should put you very close. +/-50rpm of setpoint is good enough. +/-10rpm consistently is mega.

[[email protected]]

hjtrbo, Thanks again, will follow your advice and dial back shift points (speed and rpm) and sneak-up on something that will work well without hitting the rev limiter.

[Ben Charles]

2006 Corvette LS2, 6L80E Auto, Yank 2800 Stall, E85 Flex Fuel, E-Force Supercharger, 10 lbs. of Boost, PRC heads, 227/234/114 600 lift cam, Johnson 2116LSR lifters, Kooks Long Tubes, Corsa Sport Exhaust, 7.5 gal PNR ice tank. Just completed a Dyno session which provided me with accurate torque data/RPM for both Airmass and MAP. My Virtual Torque Tables have never been changed from stock. The stock Virtual Torque Table values are 25% higher than the torque values obtained from the Dyno. I am conflicted as to weather I should update the tables, lowering the torque values by 25% or just leave the tables as they are. After adding the CAM, Heads and Lifters, and realizing a 60HP gain (currently 691hp at 6200 rpm) the transmission will not always shift as scheduled, hitting the rev limiter. After some research and reading, it appears that not adjusting the Virtual Torque tables to accurately reflect actual torque being realized, may impact the 6L80E's ability to shift as demanded. Some tuners say to leave the tables alone while others say these tables should be changed any time your tune changes. Should the Virtual Torque Tables be changed to accurately reflect actual torque or is this unnecessary.

If a tuner tells you to leave the VT alone on a build like this, get another tuner.. a lot of 6 speeds tore up bc of this

[hjtrbo]

Hey Ben, to me based on the OP's WOT dyno hits sounds like they are well in the realm of a good spot with there virtual torque. Good injector data (and MAF placement if applicable) should reflect in accurate airflow measurements to the ECM which therefore would translate to good VT values with minimal adjustment no? Certainly sounds like to me at +25% rear wheel the OP sure is spot on. Keen to hear your take on where you think this person should be at.

Granted, in other cases I've found with my own experiences that shit injector data results in significant changes to virtual torque to bring back into reality. Then along she comes with DW or ID data and the virtual torque is near on spot on to an Australian Mainline dyno with estimated drive line loss.

In kind, HJ

[GHuggins]

The problem with the torque model isn't so much wide open throttle torque calculation because in reality the wide open throttle torque calculation from Factory is very close in boosted scenarios. The problem with the torque model is the low and mid throttle areas that have to be increased for the added torque. Unfortunately with the older torque model equation setups if you increase this area a lot of the time it also increases the wide open throttle areas. This unfortunately is just a necessary evil that you have to tune the trans around for, but if you don't increase the mid and light throttle areas it'll cook your trans due to slipping. The newer engine calibrations actually add in more quadratic terms into the equations and will allow better 3D modeling. I believe this may be what Ben was referring to

[hjtrbo]

Is he referring to gen 5?
Those mid areas can't be done without a dyno so hows the guy on the street supposed to do it accurately?
I've always figured if you've got 20nm at idle, -110nm on DFCO decel, 150-200nm cruising on flat road at 65mph, and your good up top at WOT then the tables are close enough.

[GHuggins]

Gen 5's do have the better quadratic terms for better 3d modeling - but only on the 16ish and newer yr models and ONLY on the airmass side of the equations which makes things interesting to say the least... You can get the airmass side of the equations really close and dialed in, but then the map side is a whole other story on it's own with LOTS of experimenting - I can almost tell you what the torque is going to be and what timing it will roughly idle at now just based on shape and peaks. Yes, I dial them in for 0 cam up to max cam on both airmass and map. Personally remote tuning I go by what the customer is telling me - whether it's shifting too hard here, too soft there, not deceling right, idling funky or whatever - very time consuming, but this is how you make stock trans cals last when you can't tune them - talking gen 5's again - if it's a gen 4 you can pretty much work both sides of the equation.......

I personally like the add 25 to 50 in the lighter throttle areas and possibly 100 at peak boost - this all depends on boost levels too. Also highly depends on - as you pointed out - injectors and how they're scaled in the tune... Personally I've seen the majority of trans deaths caused simply from this - more so here in the mountains where your pulling steep hills.

Honestly Nathan, you're doing yours better than most I've seen. About 99.some percent of them won't even bother to dial in the different sections or bother to touch things at all. Personally - especially on the gen 4's - I've found the OE to be off even stock. Don't know if it was just particular calibrators or the fact that it was early in the torque modeling, but on the ones that are noticeably off I typically skew them as much to keep things more in line from the OE cal if that makes sense. I suck at typing things out But, I've seen some OE's 100 under what should be actual and some that much over - I think just shifting the torque cal for the additional adders is more important than making the numbers match actual - at least on the gen 4's.......

[hjtrbo]

Hey Greg, to be honest this is the first time I've heard of virtual torque causing dramas when you're just cruising around or in the mid torque areas such as towing etc. I've always figured if idle, cruise, WOT and decel is reporting good then the rest is gold.

I've been meaning for a long time to get into an in depth discussion about this. Its always been one area I've been interested in. I wish I had a dyno...

Take the following, this is a -10 to 50 degree torque sweep at fixed 2800 rpm and 1.1gram cylinder airmass. Guys that drop all that spark out in the light to mid boost areas will be sitting much lower on the reported torque curve than they should be if they had just left the OEM timing in there.
The same for cylinder airmass. If their air model is jacked up or down then the cylinder air mass will put you into areas that you shouldn't be in as well. Obviously good injector data and correct MAF placement are key.
Untitled.png

Here's a boring picture showing the effect of adjusting each of the individual torque co-efficients +20%.

Untitled.png


The million dollar questions:

So is it really a virtual torque problem or is it because guys are doing any combination of: incorrect spark for load, shit injector data, shit MAF data.

Without detailed light boost dyno data what are we supposed to recommend to the next guy that comes on here asking a question about his virtual torque? Just blanketly say add x%? Surely not???

[Alvin]

IMHO worrying about Virtual torque in Gen 4 or older applications is completely over rated. GEN 5 yes... important. Gen 4 no...

[GHuggins]

As previously stated - most of the time it's the incorrect "other" "bad" tuning methods that will cause this kind of damage. I try to look for something from 130 to 180 lbft in or around 40ish percent throttle on built setups and less on minor things like cams. Biggest issue I see is here in the mountains where you can change a 4000 ft elevation with 7% grades in around 20 to 30 minutes. Boosted setups in particular will be told to "build" their transmissions and when those burn up they start thinking tune - usually cheap inj and inadvertently incorrect torque model.

Tuned a truck for the military last week - 14000lb fully armored tahoe - had to play with the torque model fine amounts even with a stock engine to get the trans holding and shifting pressures how I and they wanted it - this is an example of loaded vehicle and hills I'm talking about.....

Now to "blanketly" say do this or that - no - you're 100% right - this isn't correct and every application and it's intended use will be different. I also think "blanketly" saying don't touch it and it's useless is also 100% wrong, but that's me

[GHuggins]

This is a prime example of what I'm talking about for "load" on a vehicle and OE torque models. Both vehicles have 6l's - different transmissions - engine's otherwise the same with possibly different intakes - even so the map side of the models are different too. One is a freightliner (17500lb) gvwr with a 5sp Allison - the other a g body van - much lower gvwr with different trans... You'll find the freightliner has a noticeably higher torque model - moving heavier loads and commanding higher fluid pressures... One would think this would be done in the transmission side, but instead the "OE" seems to put it into the engine side of the calibration........ ECM only stock is the freightliner...


So again, it was said it shouldn't be touched - the OE changes them vehicle to vehicle with the same exact engine setups, soooo

[dhoagland]

This is a prime example of what I'm talking about for "load" on a vehicle and OE torque models. Both vehicles have 6l's - different transmissions - engine's otherwise the same with possibly different intakes - even so the map side of the models is different too. One is a freightliner (17500lb) gvwr with a 5sp Allison - the other a g body van - much lower gvwr with different trans... You'll find the freightliner has a noticeably higher torque model - moving heavier loads and commanding higher fluid pressures... One would think this would be done in the transmission side, but instead the "OE" seems to put it into the engine side of the calibration........ ECM only stock is the freightliner...


So again, it was said it shouldn't be touched - the OE changes them vehicle to vehicle with the same exact engine setups, soooo

Great info!
That freightliner would greatly benefit from a Diesel

[hjtrbo]

Thanks Greg and others.

Obviously the trans slip pid is extremely important, especially for us back yard guys without access to load bearing dyno data.

I'm still confused that a tune can be done and given to a customer and the trans slip goes unnoticed and toasts the trans days or weeks later. People can't be that stupid, surely?

[GHuggins]

I've seen them burnt up in as little as a week per what customers have told me. Their "tuner's" would then tell them they need a "built" trans... Usually power adders like blowers that require inj and MAf changes as you pointed out I know you've seen some very questionable tunes from "tuner's" out there They don't pay attention to anything and tell people everything is good - one in Texas is real bad for this - or at least I think that's the one I see the most on here getting complaints on...…

I still believe trans tuning is primarily done via the torque models on 4th gens as pointed out by how the OE did it and is why I say on these - real world numbers don't necessarily work.

[Alvin]

So again, it was said it shouldn't be touched - the OE changes them vehicle to vehicle with the same exact engine setups, soooo

The original calibrations are not consistent in any way. Just look at the injector data between years, models or whatever with the exact same part number injector.

[GHuggins]

The original calibrations are not consistent in any way. Just look at the injector data between years, models or whatever with the exact same part number injector.

That's kind of what I'm talking about - 2 identical engines will have completely different setups in the cals for the vehicles they're going into and due to whatever group may have calibrated them.

[HawkZ28]

Is the correct way to log and adjust virtual torque to use a PID like Engine Torque (that isn't clamped), and set up a histo in the scanner to mirror the VT tables, and then filter for each "Spark = X" with a filter?

I've used the following filters for "Spark = 10*" for instance:

[14.161] <15 and [14.161] >5

[14.161] =10

Then click extrapolate, and then calculate? Should any smoothing or interpolation be done to reduce peaks?