As the subject line indicates, what would the draw backs be if one increased the ->Fuel->Transient->Delta TPS during steady state tuning to prevent transient fuel from affecting the steady state data logging?
One caveat is that you would drive really smoothly.
Would the car even drive well enough to collect data?
TIA,
Chris
Transients are natural, you cant really get rid of them. So what are you trying to achieve by increasing it? Going from one wrong value to another makes things different, but not better.
I'm trying to reduce the effect of transient fueling during steady state logging in the hopes to get 'cleaner' data that isn't biased by transient fuelingOriginally Posted by redhardsupra
if all you're interested is steady state, then just use filters, and ignore all values in transitory states. Or are you concerned with the residual effects of transient fueling on the steady state right after transients?
Would you happen to know the best PID to monitor for transitory state? The dx/dy filters don't seem to work filtering transient data out.
I'm mostly concerned with any transient during or after that would affect the lambda error I'm recording while compiling a VE histogram.
I don't, I do all such data mangling in Matlab. Does HPT even have deltas? I haven't used their scanner in years.
If you're already watching transient fueling, that's the piece of data to use as the transient indicator. If it's not big enough, crank up the values for the transient modifier, and then you will see transients as rich spikesprobably not the best idea for driving, but if all else fails, might as well try to observe it that way.
MAP/TPS spikes are the quickest ones to move. Same goes for airmass. The real problem is the scanning frequency. Older (LS1b) computers aren't very fast, so catching a transient in progress is actually kinda lucky.
All these things are just filtering out unwanted data. The best way is to simply not have much of it, or if you do, have it stick out badly. So either drive super smooth (my preferred method) or snap that throttle open/shut like a pissed off teenager.
I create separate histograms to record lambda error for a specific table configuration for "driving" where TPS>0.4 and Wheel Accel/Decel <10 (this is PID.8010)
and "idle-cruise" conditions where TPS<0.5 and then drive really smooth for both situations. You can add the PID.8010 filter to idle-cruise histogram too, since
the less than 10 means the rate of change is less than 10 whether it is an increase or decrease (absolute value type parameter).
For calibrating the transient Impact Factor Gain table, I use the filter [PID.17]>5AND[PID.11]>60 so that we are measuring lambda error during the time that a transient occurs upon acceleration. This allows you to see how much this gain at a given mass airflow rate has to be decreased to make up for the decrease in tau (wall film) on a modified motor. The histogram for this is just MAF LB/HR on the rows to match the gain factor table in the tune file.
Reply With Quote