tephra

iTrader: (6)

the axis is load/temp, yes.

but whether they should be swapped I am not sure.

not being an expert on Idle control I dont know what "looks right"... I will play around with the XML and see...

Eric - if you know better please let us know, as the next ver of ecuflash will have our changes... need to get it right

edit - So I have said "Flip Y" on the table (not the axis) and this is what I get.

Would this be more correct?:

biggie5252

iTrader: (47)

What is the ISC stepper table even for? Obviously it controls idle somehow, but so does ISCV initial step position with AC off Neutral table. It even has steps going down as temp goes up, but the values don't seem to coincide with actual values. At least as far as what I've seen tonight.

l2r99gst

iTrader: (2)

Tephra, I don't know if I know better, but this is what I am talking about. I looked into this a little bit in the past, but never had time to fully research and test. I was messing around with different scalings and swapping.

Anyway, the first thing that I did was I thought that the load values were going too high and they didn't seem to fit the table well. If this was truly an idle stepper table, then it probably doesn't care about load that is up to 160. This is why I asked if you verified if load/temp were the axes. If I use the ECUFlash scaling simply called 'Load', then the table looks like this:




That may make more sense why some people are affecting idle with some of their last column changes, because they may be going too 'high' vertically and touching some of the lower loads, which may be used at idle.

Then, I tried to sway and flip the axes until the table made 'sense' to me. I would expect ISC steps to increase with decreasing coolant temp and to increase with increasing load. If I swap XY and flip X, then I get the following. The two pictures are with the 'Load' scaling and the current 'Load16' scaling.




This just makes much more 'sense' to me, but I don't know if it is right. The reason I don't think this is right is because it wouldn't explain people fixing their throttle hang with the cells that they are adjusting. It may very well be correct how we have it now, but then when people make threads like this, it makes me believe that we don't have it correct the way it is.



Eric

biggie5252

iTrader: (47)

On the last table, what if you flipped it one more time as in having the 136 cells in the top right? That would have steps decrease as load went up (fixing throttle hang) and as coolant temp went up (makes sense). I'm not very smart at any of this, but looking at it, that's what I would think.

I'll load up these tables and give them a shot.

The table I posted last night, didn't take care of the idle issue, but drastically reduced the temperature required before it would idle on it's own. And as far as I can tell got rid of the throttle hang.

l2r99gst

iTrader: (2)

biggie5252, I hope you understand that I am editing the properties of the table in ECUFlash, not changing the data within the table.

So, when you said that you will load this up, I think you may have meant that you were going to copy it. But, nothing has changed. It's simply viewing the same exact table in a different setting by editing the table and choosing Swap XY and Flip X.




Eric

biggie5252

iTrader: (47)

I understand that you're not "changing" the data, but basically changing the arrangment. what I guess I don't understand is the tables do have different values in them (as far as I can tell) for a given cell in the table. Will the ecu not just use whatever value is in that cell regardless of how it got there? Does it just not look up the x axis to find the load value, then look across to find the cell that coincides with temperature and use what's there?

l2r99gst

iTrader: (2)

I remember the day that I spent with this, I did a few minute log. I started logging before cranking so that I could see the initial position of the ISCSteps (which in my case was 83) and the coolant temp, load, etc.

I used Logworks to make a 3-D chart of the small amount of data that I had. I think the scale of the load may be off, so don't worry about the exact numbers on the scale, but you can see the general shape appears that I mentioned above. There is increase ISCSteps with decreasing coolant temp and increasing ISCSteps with increasing load. Following are three graphs of my limited amount of data. One shows the 3-D graph and the other two are rotated so that you can see the relationship of ISCSteps with respect to load and coolant temp, respectively.





Then, I used ECUFlash to graph the chart I made with the swapped XY and flipped X. It seems to take the same shape. It's been a while since I messed with this, but I remember doing this one day to try to understand this table better.



I remember reading posts about this table and it's relation to the other ISC tables as well, the tables that are just 2-D coolant temp tables. I forgot most of what I read though.

I'm interested on what is truly known about these tables from disassembly. Also, if anyone has long logs available with coolant temp, iscsteps, and load, post them up. We can graph them out to get a better understanding of where the ECU is setting the ISC under varying conditions.


Eric

biggie5252

iTrader: (47)

I have no clue about anything disassembly, but I'm willing to log as much as I can to figure out what's going on. The table I posted the other day pretty much works for me as long as I'm there to maintain the throttle for a few minutes until it will maintain idle itself. I can't wait until I can figure something out so the car will idle on its own without any throttle hang. Right now it's a choice between the lesser of two evils.

mr2driver

iTrader: (11)

On mine I changed the stepper table in the same fashion and it was working great until I took it to the dealer to get the AC recharged, now it has the same symptoms yours does. I really hate the throttle hang, but I might hate the crappy idle more. Maybe when I took it to the dealer they reset my ECU, and it "relearned" the idle. I am going to try setting the stepper table back to it's original form and see if that corrects the idling problem.

biggie5252

iTrader: (47)

I've adjusted my throttle cable to about 13-14% during idle and it didn't help any. I've noticed that when the car is warm ISC steps are in the 60s and idle is pretty much dead on with target idle. When it's cold it won't idle and ISC steps drop to the 20s, but I can't figure out why.

l2r99gst

iTrader: (2)

I think it has to be something with the table and what I have been discussing.

Why don't you try adjusting less cells than you have in your screenshot on the other page and see if your low temp iscsteps are still affected.

biggie5252

iTrader: (47)

I did raise the cells back to stock until higher loads before i started the taper as shown in my tables on this page and it seems to help with the cold ilde as in it doesn't have take as long/have to be as warm before it will maintain idle on it's own. With the table I have now (the one posted on this page) it won't maintain idle until water temp gets to just under a quarter of full scale (it usually resides about half, maybe a little above) then it's fine, but I do notice a tiny bit of hang, but not so much that it bothers me. Before it had to be hot to a point that I could only reach by driving, but can't reach by idling. Does it have something to do with the cams? It seems that everyone that got the table on the other thread to work didn't have cams. At least it wasn't stated that they had cams. I didn't have this problem that I noticed, but it had been a long time since I'd driven a stick, until I got reflashed for cams.

biggie5252

iTrader: (47)

l2r99gst, I will try your table that you sourced the graphs from tomorrow and see if it works. I know you said you didn't change anything, but, to me, it's worth a shot.

mr2driver

iTrader: (11)

Adjusted the table back, idles great now, just throttle hang is back again. There has got to be another table that controls this. I think adjusting the stepper table is not the way to go.

biggie5252

iTrader: (47)

Digging this up. . .

I've noticed that since it's been warmer, my idle issues are next to none(I'm assuming due to higher initial coolant temp). It seems to be riding a fine line though. Some days (mid to high 70s) I have no issues whatsoever and other days (mid to high 80s) I have issues, including stalling when letting off throttle from a low speed when the ac is on, but they only last for a short amount of time before the car warms up as before. I believe the lack of issues with the cooler days is that I'm driving (higher load) when coolant temp goes through the temp range in which the car won't idle. I'm going to do some more logging tomorrow. I've been kinda lazy with this. But it's getting kinda embarrasing when it looks like your car runs like crap.