delta map pid

[5.7ute]

I am trying to make a calc pid for delta map & tps & it is returning insane values. I am using the dx() function eg: "dx({SAE.MAP},x)" x being the number of frames over which the value is derived. The values being returned are extremely inaccurate. For example a map change of 1.0 kpa can read as high as 5 or more. a TP change of 17% can read well over 30%. This value needs to be derived over as few frames as possible to test the steady state cax file that I made. Any ideas??

[mr.prick]

Is Delta_MAP/Delta_load the difference from the current frame to last/vice versa?


#Units Low High Fmt Expression
#------------ ---------- ----------- -------- --------------------------------------------------------------
*CLC-00-003
psi 0 16 .0 "({SAE.MAP.psi})-value({SAE.MAP.psi},frame()-1)"
kPa 0 105 .0 "({SAE.MAP.kPa})-value({SAE.MAP.kPa},frame()-1)"

*CLC-00-041
% -100 100 .1 "({SAE.TP})-value({SAE.TP},frame()-1)"

#Code PRN SLOT Units System Description
#------------------ ----------- ------------------- --------------- -------------------- -----------------------------------
CALC.DELTA_MAP F003 CLC-00-003 "kPa,psi" Air "Delta MAP"

CALC.DELTA_LOAD F041 CLC-00-041 % Throttle "Delta Load"

What's this .cax file for? :secret:

[5.7ute]

Thanks mr.prick. I used similar to yours & it works great.
*CLC-00-211
KPA 0.0 105.0 0.3 "({SAE.MAP})-value({SAE.MAP},frame()-1)"
*CLC-00-212
% 0.0 100.0 0.3 "({SAE.TP})-value({SAE.TP},frame()-1)"
The cax I am working on denotes the steady state conditions. It is part of the transient fuelling & injector stuff I am getting into line.
I need to double check the scalars, descriptions etc before I post it up hence the need for the delta pids to check it against.
Thanks again
The other Mick

[mr.prick]

"My Delta_MAP" might not be what you are looking for,
it is the kPa/PSI difference not percentage.
The Delta_Load may work but try
"dx({SAE.MAP},-1)"
"dx({SAE.TP},-1)"

[redhardsupra]

the dx function is rather useless on ecu's with low resolution (anything e40 and older). on the newer stuff it's much more useful. i dont use it much due to habitually exporting everything to matlab and using their smoothing functions to make sense of all the spikes and back'n'forth twitching due to lousy resolution.

what is your goal here? transition spotting?

[5.7ute]

the dx function is rather useless on ecu's with low resolution (anything e40 and older). on the newer stuff it's much more useful. i dont use it much due to habitually exporting everything to matlab and using their smoothing functions to make sense of all the spikes and back'n'forth twitching due to lousy resolution.

what is your goal here? transition spotting?

Pretty much.
I have found the parameters for when the pcm denotes either a steady state or transient condition. These I have put into a cax file which I want to double check before posting it up. This is mainly to help filtering during road tuning, but will also help in the injector offset parameters I am still working on.

[joecar]

+1 what Marcin said...

any thing less than 40 Samples/s is just too poor a resolution... and 40 S/s is not sufficient either.

[joecar]

Mick,

Are you looking for "change in MAP" or for "rateof change of MAP"...?


"change in MAP" is how you have it in post #4, simply the change in MAP... [i.e. it has units kPa].

"rate of change of MAP" is the change in MAP divided by the change in time [i.e. it has units kPa/s]... this is what dx() gives.


Example: a change in MAP of 1 kPa over 0.2 seconds (two frames at 10 frames/s) gives rate of 5 kPa/s.


Note: multiplying the rate of change in MAP by the change in time gives the change in MAP.

[joecar]

Pretty much.
I have found the parameters for when the pcm denotes either a steady state or transient condition. These I have put into a cax file which I want to double check before posting it up. This is mainly to help filtering during road tuning, but will also help in the injector offset parameters I am still working on.Mick, good job...:cheers:

does this mean we will know which of the MAF or VE the PCM is using...?

[joecar]

That some interesting calc pids you got there... ;)

[redhardsupra]

Mick, good job...:cheers:

does this mean we will know which of the MAF or VE the PCM is using...?

if you want that you can calculate different airmass values, one from MAF, one from VE, see which one is closer to DYNAIR. also, there are some really old posts (5yrs+ by now) where i think chris@hpt posted the formula with conditions at which you use SD not MAF. of course due to age of that, I don't know if/how much of it applies to the newer ecu's. but you could just mimic these conditions with a filter.

[joecar]

Marcin, do yo have a link to that formula...