RomRaider

Bill,

First off, thank you for the new release! There is one section I was hoping to get some more clarity on:

- 'Throttle Actuator Duty' -> 'Requested Torque (Accelerator Pedal)' [reason: A more accurate representation. Unit conversion also changed to tie in with new 'Target Throttle Plate Position' table (see help text).]

I read the notes on the table properties.. but are we requesting Engine Torque or are we requesting throttle motor torque with this value?

It appears to be an arbitrary desired torque value (maybe in N-m). This is perhaps why the STi has a higher value in the WOT column than the WRX/FXT/LGT. This value is used indirectly as an input (x-axis) to the new 'Target Throttle Plate Position' table.

However, ETC logic is very hard to follow with all the redundancies built-in. What I do not know is if the throttle plate position target is further manipulated or if the desired torque target is a manipulated further in any way. From testing on an 07 sti, the logic as I have outlined and the relationship between the old table and the new has been consistent.

That's what I was looking for. I have begun to play with them on my 06 Wagon to see if anything feels better or different, no real butt-dyno feeling yet. The data in the logger doesn't seem to be much different either when it ties into the overall tune.

We know that changing the main map does increase throttle response by opening the angle to a larger value, more tweaking will be needed.. I'll let you guys know if I discover anything that helps explain it a little more or make the car drive better.

To clarify it works like this:
1. A requested torque value is determined from the 'Request Torque (Accelerator Pedal)' table (used to be named 'Throttle Actuator Duty'). So, depending on much the driver is hitting the gas and at what rpm will determine the requested torque.
2. This value is used as an input (x-axis) for the newly defined table 'Target Throttle Plate Position (Requested Torque)'. This determines the throttle plate target for which the throttle duty will be manipulated to attempt to hit the target.

So, if you look at an 06 wrx compared to an 06 sti, say, by upping the requested torque in the WOT column of the wrx in the 'Requested Torque (Accelerator Pedal)' table to match the 06 sti, you are not ending up targeting a wider throttle plate position, you are merely targeting 100% throttle plate opening sooner. So, both maps have to be taken into consideration. The torque value itself could really be anything you want as long as you are consistent across both tables (although we don't know if it is used in some other logic so I would stick with the factory range).

I see the correlation, and I believe this may help to resolve the quick stab 100% oddity that the DBW ecu does. Say you're at 20% throttle, then stab and hold around 50%, it will show 100% for a few seconds then taper itself back down towards the intended 50% area.

I will try to re-work the debug tables in line with the pedal table. The problem with some of it was the rescaling of the axis not lining up with the current tuned table. I have rescaled similar to the '06 STI now and will be testing it out shortly.

I don't think that would be related to these two tables.

Why are you labeling it "torque" versus just saying it is an arbitrary value? The way you describe it, it seems like it is just an arbitrary numeric value.

I understand where you're coming from on the complexity of the DBW system, I've had trouble making too much sense of it myself. I do see how the primary "torque" table translates to the throttle angle table, though.

Good job on the new defs, btw.

Thanks.

Arbitrary in the sense that I don't think it matters what the value is necessarily as long as both sets of tables are scaled with their relationship in mind. From all the patents I've read, the requested torque (also called desired torque) map is a consistent element in ETC. Also, the maximum torque value used is different across different models which seems to, in broad terms, mirror their relative power - STi (400), usdm WRX/FXT/LGT (320), Impreza 2.5i (250), jdm lgt (340-370). If this is N-m, that would be (295/236/184/250-272 lb-ft). I don't think it is designed to exactly match the rated torque output (otherwise the LGT should be higher), though, but it seems to follow close enough.

Any idea if this table will "make the cut" when the n/a defs become available? Seems this, along with timing adjustments could go a long way to clearing up sluggish throttle response in the 05+ DBW n/a cars, if I'm reading your description of the table correctly.

I've looked at it briefly on an 06 2.5i and it appears to be a lot more complex. The tables are huge (like 22x28) and there's something like 8 of them, all with different values, some radically different. Not as clear cut but I'll have to look into it more.