SlowTeg

Cool way to gain some more resolution Gernby and not lose the last 4 columns. Just have to use some math which isn't too big a deal. I need to tweak mine a little more one of these days as the light throttle needs to be "desensitized" more but that's a nice way to use the whole table.

Gernby

Hard to say exactly what "half open" even means. It could mean that the plate is at a 45 degree angle, or that the plate is blocking 50% of the cross sectional area. However, my guess is that a 50 in the target throttle map (not the index) should be about half open. The 144.1 is just an intermediate reference that links the 2 throttle tables. It is basically unitless.

King Tut

That makes sense. I will try adjusting mine tonight, and if I get time I will mess with limiting the throttle plate opening above 5000 RPM again.

shind3

Have you plotted a MAP vs TPS curve from a datalog? If so, you'll see how quickly MAP goes up relative to TPS. MAP is directly proportional to Torque. With a stock mechanically linear throttle mapping, change in pedal at the top of travel results in higher torque change than intended. Adding 10% pedal might add 20% torque at low pedal angles. You also hit the max MAP way before your right foot has gone all the way down. So you're kind of only using part of the pedal to control the rear end in corners. Honda kept it mostly mechanically linear like this or people would complain about how the DBW cars weren't as 'lively' as the older ones with respect to 'throttle feel' and that they had to push the pedal more to get the power out. If you want your right foot to be linearly connected to torque output, you'll want to 'desensitize' the throttle mapping at low pedal angles. So a bigger change in pedal results in a smaller change in TPS but a more proportional change in Torque. This is great for low speeds especially where we can use the full range of the pedal to control torque like in autox situations. Everyone here probably agrees with this so far.

But why then are there increased index values at higher speed in the new "pedal vs speed" table? Because when you are driving on track exiting a corner you actually want the throttle desensitized when the pedal is near the floor. At least in a car with such amazing neutral cornering balance at speed. So you can lift a little bit and not transfer as much weight per unit of pedal movement. You don't want the torque to change much at higher speeds when you lift throttle a little bit for attitude correction especially mid corner. It's a bit more confidence inspiring to have a desensitized throttle when it matters most! High speed corner exits...

All mechanically linear throttles are easiest to vary torque output at the high end of their travel. A throttle body has a rate of crossectional area change that is high at lower TPS and low at higher TPS.

For the math nerds out there we want to think about this in terms of deltaTorque / deltaPedal. I (probably you as well) want this number to be low during high speed corner exits. So with the DBW stuff, we can have a more linear torque pedal at lower speeds but still retain the inherent advantages of a mechanically linear throttle when it matters most.

Does this make sense? I sat here for a while trying to get my thoughts in order. Hope this was a better explanation than my earlier 'because weight transfer lol'.

Shek

King Tut

So my test was a complete failure last night. I swear everytime I try and mess with this it just leads me to more frustration. I set the tables up to reduce throttle plate opening from 6000 RPM to 8000 RPM. The datalogs said that TPedal was 100% and TPlate was 60% at 8000RPM, yet my car didn't feel any slower, and still made 14 psi of boost no problem. This is the shit that really pisses me off. I can't believe what Hondata is telling me. There is no way that the actual throttle plate was at 60% open. Now I wasn't on the dyno to see if this made any actual change in power even though it was pulling 14 psi, but I can tell you the butt dyno felt no noticeable change. I don't know if it is just ignoring the TPlate value and doing it's own thing because the shift in plate value is too harsh? There isn't much I can do with 1,000 RPM resolution. I sure am glad that the table goes all the way to 10,000 RPM as well. Are there any other standalone options that might have better control over the DBW throttle?

shind3

Is this your first turbo car? This is what happens with an 'oversized' throttle body, lol.

King Tut

No. I wouldn't call the S2000 throttle body oversized. Your statement doesn't even make any sense.

Gernby

You should limit the throttle plate across the entire RPM range to something like 40% max, then datalog some full throttle pulls. Then I would do the same thing over and over again raising max throttle by a set percent increment (flat across the whole rev range). That way you can plot the response out on a graph, which will help you calculate the non-linear throttle map that you will need.

King Tut

That is the only way I have ever had success with getting Hondata to do what I command it to do. I have managed to get it to only make 5 psi at 8000 RPM with a 7 psi spring, but I had to limit throttle for every RPM. The reality is that doesn't help me with my goal because I want 100% throttle until somewhere between 5000 and 6000 RPM. I am thinking that it has some kind of failsafe where it ignores the TPlate value and reverts back to the TPedal value if certain conditions aren't met.

Gernby

Are you sure that you were setting the 5 colummns on the right to the restricted value? In previous versions, and also with the new version in its default setup, the right 4 columns are completely ignored.