asx wrote:I did mean dash pot in the MS terminology, yes. Personally, I think the terminology fits
Until I searched, JUST before I posted, I didn't know that a dashpot was something other than a potentiometer on your dash. I had no idea that it was a term for a damper. Which likely explains my disdain for that aspect of their terminology, though I still think it's unclear.
the main purpose is t hold the rpms high when returning to idle to prevent stalls, but i can see the lack of clarity is an issue.
This seems like a weird thing to have to do. I've driven quite a few different cars, and as far as I can recall, none of them held the RPM artificially high post revving just to prevent stalling out. They just naturally fell back down to their steady state. The closer they get to that, the less they are decelerating anyway.
I'll cleanup my post tonight (3-6hrs) when I can get on something with a real keyboard instead of my phone.
Cool, I grabbed a copy as-is in tiny format in my post for reference, go nuts, if you want :-)
I both agree and disagree about closing the valve.
This paragraph made me think about some behaviour that I experienced where tip-in was ruthless (negative torque to considerable positive torque in a step), I wonder if that's what you mean?
My miata (99) is effectively completely reliant on the idle valve.
As are many modern engines.
With the throttle closed, valve closed, and clutch disengaged (from the flywheel), there is too much engine braking.
I'm not sure there is such a thing. Driven an older diesel? They engine break enough to swing the arse out in a corner, and I love it :-)
It is great for shifting quickly as the rpms fall quickly, but far too fast when I'm leisurely shifting. With the clutch engaged (locked to the flywheel) It is good for coasting down steep hills, but too "jumpy" when freeway driving and it sheds far too much speed when going down small hills.
Clearly this trailing throttle air flow would be user settable, not necessarily zero by default. I can see it being a problem, maybe, but even if you leave the shifting for 5 minutes and it's sitting at idle, it's just going to come back up to match with a tiny bit of clutch slip.
What about an option to close the valve / overwrite valve duty cycle in the engine braking / fuel cut on overrun code?
I'm not sure that it's necessary. Motec engineers call it "feed forward" and I call it "essential" to have an open loop tune of what you want before PID takes effect at all. This is certainly how it'll work. For the given conditions, TPS, RPM, CHT, ? there will be a particular duty as a base and the PID, if even used, will act with that as a foundation adding to or subtracting from this. You can't have a PID loop work properly without it, period. If some clowns elsewhere tell you otherwise, ignore them, their credentials are shady at best.
Personally, i like using the idle control to lug me around in heavy traffic and parking lots.
Sure, that's a personal choice. I prefer the natural torque level, though, not one that increases with sustained RPM differential.
There can definitely be a safety issue if the idle code is badly tuned and the valve "runs away." Not sure how to avoid that beyond being able to set a max RPM after which you are kicked out of the control loop and a new initial duty cycle is pulled on reentry. This is one of the reasons why storing/reusing the last duty cycle used is the devil.
What sort of idiots do that? Wait, don't answer that.
I don't really have any comments on idle specific fuel/spark controls. Most ECUs I've dealt with either have more than enough table resolution to dedicate a couple columns to idle rpms, or they had small idle trim tables, or I took advantage of map switching to give one table a lot of low rpm low map resolution and then switch at boost onset to a table with a lot of high map high rpm resolution.
What do you mean by "small idle trim tables"? This is likely what I'm referring to and trying to come up with a rough design for.
On a semi related note, how do you feel about fuzzy logic vs pid? I think that for novice users its a lot easier to tune the fuzzy logic values (i.e. O.E. 16bit subaru ecu boost control) than understand PID
I have no opinion, educate me! :-)
, BUT that could be fixed by good documentation, which I sure as hell haven't seen from a certain s12c based ecu.
LOL, I have no idea what you mean.
Fred.
