Help me understand the tuning logic

mass-hole · May 14, 2023

I have both HPTuners and VFTuner but I am having a hard time bridging a gap in the tuning logic on the UR motors.

It looks to me that the Torque request in the ECM just appears out of thin air and then is sent through the Torque to airload conversion table to determine how much air should be allowed into the engine. It seems that tuners are using the Torque to airload table to "spoof" the engine into flowing more air by adjusting how much air load is demanded for a given torque value. This is how you get snappier throttle response.

Is this correct or am I missing something? I find it very confusing that there is no "drivers demand" table where it uses pedal position and rpm to determine a torque value. This is how every other engine i've tuned works. My Jeep Liberty diesel and my F150 Ecoboost both follow this logic.

snivilous · May 16, 2023

mass-hole[OP] initial post: ↑

I have both HPTuners and VFTuner but I am having a hard time bridging a gap in the tuning logic on the UR motors.

It looks to me that the Torque request in the ECM just appears out of thin air and then is sent through the Torque to airload conversion table to determine how much air should be allowed into the engine. It seems that tuners are using the Torque to airload table to "spoof" the engine into flowing more air by adjusting how much air load is demanded for a given torque value. This is how you get snappier throttle response.

Is this correct or am I missing something? I find it very confusing that there is no "drivers demand" table where it uses pedal position and rpm to determine a torque value. This is how every other engine i've tuned works. My Jeep Liberty diesel and my F150 Ecoboost both follow this logic.
Click to expand...

I have seen a couple tunes where desired air load or optimum torque was adjusted, and had asked a similar question to "SlowNStock" on the HP Tuners forum. This was his response:

Optimum Torque

The torque maps are a little tricky; they aren't used like a "true" torque based setup, where driver demand governs everything like, say, Gen V GM stuff, but they are referenced in conjunction with the more traditional tables (throttle angle vs TPS, etc.), and the interplay isn't straightforward.

The Optimum Torque table can be reasonably estimated for cases where the only change is boost, since torque at a given RPM is (reasonably) proportionate to load - very similar to a VE table in that regard. In the same vein, changes that dramatically affect VE across a given load column are much more difficult to handle, because we can't always predict what the behavior is; the most obvious example would be a significantly more aggressive set of cams and cam timing. Fortunately, the latter is rare since very few people are doing cam/head work.

With some setup and massaging, you can build an optimum torque table from scratch with street logs. It requires a very limited set of channels (Engine Speed, Vehicle Speed, Absolute Load, Actual Spark, Current Gear) set up for the fastest polling interval possible, careful logging, and you have to parse out the data to correct for current gear and ignition efficiency. In essence, you determine the net force (Rolling Resistance + Aero Drag + Acceleration) at any given time from the logged speed, translate that to Input Shaft Torque (Net Force / (Tire Radius * Current Gear Ratio * Final Drive)), then correct that figure for converter stall (if unlocked - requires torque converter characterization) and ignition efficiency from timing vs MBT. Not a trivial task, as you can imagine.

Pragmatically, it doesn't really make much difference if you're off by a little either way, so it's not worth trying to whittle it down to an exact figure.

Desired Air Load

Assuming the Optimum Torque table is correct, the Desired Air Load table being the inverse (which is how the non-TRD and TRD tunes are both set up) is a good starting point. When referenced, Desired Air Load can add/remove throttle angle to achieve the targeted air load for a given torque request; if the value in the table is too low, the PCM will limit air load by closing the throttle, if it's too high, it can try to increase air load by opening the throttle further. As a result, the inverse of a properly calibrated Optimum Torque table will generally be close to ideal (won't hit the limit, won't be overly sensitive) in most cases. From there, we can make small adjustments - if it's still too sensitive, we can decrease the values, if it's too dull or hits a throttle limit, we can increase. As far as exactly when it references the table, I haven't fully mapped that out - my very cursory answer is "not always, but enough to always matter".

Like the Optimum Torque table, leaving it as-is for common setups (S/C +/- pulley, no scaling) isn't much of an issue because it mostly dictates part throttle behavior, and those elements don't come into play as a result. With a turbo setup it's a bit more important because the turbo acts more independently of the throttle, under/overshoot is exacerbated and it's much easier to get something jerky or sluggish as a result (same with cams, to a lesser extent).

I will generally start with the inverse, then iterate in smaller steps (+/-5%) to see if it behaves better one way or the other. It doesn't have to match Optimum Torque forever, but it's much better to start with the inverse for the aforementioned reasons, then refine for subjective drivability than shoot in the dark hoping it works.

Caveat: If the tune is scaled, it Desired Air Load behavior can be screwy depending on how scaling is done. At that point, even the inverse is often just a rough starting point; I usually have to spend a bit of time refining it, but that may be (at least in part) a consequence of the types of setups that need scaling.

Optimum Toque Trans

I'm fairly certain this table plays into driver demand and trans behavior, but I've done fairly little work sussing it out. From some trial/error, I typically get rid of the dip around 2600-4000 RPM up to 45 deg throttle (which is an artifact of the factory ACIS activation point being @ 50 deg, which we either lower for N/A or turbo setups or remove for S/C), add a little at WOT (since we are making more power), then smooth it a little (the stock table is a bit messy). I haven't done enough testing in terms of adding significantly more torque for boosted setups, but my suspicion is it would be beneficial to do so.

Summary

Overall, what people do with regard to those three tables - and the two load limiting tables - is a bit of the Wild West, and I wouldn't take any of it as gospel because I don't trust that many have spent time to really understand what's going on. Same goes for using the values from someone else's tune without having a lot of detail on how they arrived at those figures.

The big challenge with precisely defining all of the behavior of the torque tables accurately is we can't log driver demand/calculated torque, so it's all done indirectly and with inferences - not exactly quick, easy, or precise work. If we could log those, it would be much faster and easier to back out the logic. Of course, having a BIN to decompile would be best, but more effort than it's worth...

I know VF had a brief writeup on it at one point, but none of it was really any further along than the above if memory serves. Worse yet, based on some logs, exactly how some of those tables are referenced by the PCM changed over the years, so there may not be just a singular answer.

Sorry, not the most helpful info on these...
Click to expand...

And as far as snappier throttle, from my understanding and comparing tunes that's primarily just modifiying the "Desired Throttle Angle" tables (Engine -> Airflow -> Electronic Throttle -> Desired Throttle Angle).

mass-hole · May 23, 2023

snivilous said: ↑

I have seen a couple tunes where desired air load or optimum torque was adjusted, and had asked a similar question to "SlowNStock" on the HP Tuners forum. This was his response:

And as far as snappier throttle, from my understanding and comparing tunes that's primarily just modifiying the "Desired Throttle Angle" tables (Engine -> Airflow -> Electronic Throttle -> Desired Throttle Angle).
Click to expand...

Awesome, thank you!

mass-hole · May 24, 2023

snivilous said: ↑

I have seen a couple tunes where desired air load or optimum torque was adjusted, and had asked a similar question to "SlowNStock" on the HP Tuners forum. This was his response:

And as far as snappier throttle, from my understanding and comparing tunes that's primarily just modifiying the "Desired Throttle Angle" tables (Engine -> Airflow -> Electronic Throttle -> Desired Throttle Angle).
Click to expand...

This makes me wonder how the PCM asks for more torque when you add a supercharger. Like the desired load and optimum tables even on a non-supercharged truck exceed loads of 1.0 already, but where is it coming up with the desire to exceed 1.0? Why wouldnt it just close the throttle to limit torque? Is that simply from the desired throttle angle table and how much air mass it gets?

Log in or Sign up

Welcome to Tundras.com!

Help me understand the tuning logic

#1
mass-hole [OP] New Member

#2
snivilous snivspeedshop.com

#3
mass-hole [OP] New Member

#4
mass-hole [OP] New Member

Products Discussed in

Log in or Sign up

Useful Searches

Welcome to Tundras.com!

Help me understand the tuning logic

#1 mass-hole [OP] New Member

#2 snivilous snivspeedshop.com

#3 mass-hole [OP] New Member

#4 mass-hole [OP] New Member

Products Discussed in Entire Site Forum: Performance and Tuning

#1
mass-hole [OP] New Member

#2
snivilous snivspeedshop.com

#3
mass-hole [OP] New Member

#4
mass-hole [OP] New Member

Products Discussed in