Well the 3d map is the one controlling things now i left the 2d map in there for reference.

What i meant though was which rom I have one with uncorrected load and one with corrected load as the VE load axis.

 

Huh. Didn't notice there were two variants... Was there more than one ROM in the zip ? I used the 9653xxx ROM from post # 24 .

Anyway just got back from a test drive... I didn't reset my trims that were a bit high from the previous flash so the 02 feed back went nuts for a bit then it settled down to 14.7 at about 36 load (1byte). It felt pretty good so I drove it around with no problems. It was a bit lean but if your code works I can certainly tune that out. Didn't get into too much boost, but I did see 100 load at about 1.7 PSI. after driving around for a bit it was still idling at 36/39 load before I shut it down.

The log is a bit big, over 3mb, and 400k zipped. If you want to PM me your e-mail I'll send it to you. I'll let you look at the data before I make any uninformed pronouncements, but it seems to work... Nice job

 

Ah I don't need a log that big really the goal is to see if it tunes out like its supposed to. I'll trust your analysis of whats happening rather than scrutinizing the log for myself.

Here's my original post for the second type:
https://www.evolutionm.net/forums/9040269-post99.html

I've been putting all the roms i've done on page 1

 

Cool. I'll mess with the 3D table and see if I can get it to do what I want....and if it will start my car in the AM in -5*F

 

I've tested the rom from post #99 today, only briefly though, the idle is ok and cruise seem to be fine ( even though the cruise is a little rich) but will keep you informed of how it proceeds after I've dialed in the VE a little more

Andy

 

can you explain a little more please?

from my 2D maps the rpm VE goes down the in the 3D table? and I know that the map VE is relative to the load, but not sure how to expand, I've had a go to produce this but will fine tune as I go I guess?

thanks
Andy

 

It matters how roadspike is using the load axis for the 3D map, but anyway, here's a quick example for your ROM.

See your map ve at 50kpa = 40 load. Well, that's 40/50=.80 = 80% VE. Now, to create your 3D VE for your 40 load column, just multiply that 80% for each RPM from your RPM table. Your RPM at 2500 for example is 100%, so your 3D VE is .8 * 1.0 = .80 = 80% VE. So your 3D map cell at 40 load and 2500 RPM should be 80%.

Just do that for all of your cells and use interpolation for the cells that you aren't directly calculating. You can set something up in Excel to do it easily for you, but the setup will take some time.

 

I have two versions

One uses 2byte load and the other uses the map sensor ve load

 

I've forgotten since it's been a while since John first posted all of the routines from IPW to load and how SD was implented, but I thought IIRC that load is calculated after the steps where he patches in the VE calcs. If that's true, how are you getting 2 byte load beforehand? And what 2 byte load? There is uncompensated, baro comp, baro+temp comp, etc.

I can understand using the load from the 2D map VE table, but I just don't remember off the top of my head the entire explanation of the SD patch. I'll have to go back and reread it.

 

Its pretty simple really, i don't :P.

It just uses whatever the value was before hand. Since it updates every 0.01 seconds i wasn't really worried about the load changing too much.

The 2byte load i'm using is compensated same load value in tephra's ignition and fuel tables really.

 

Driving around on the original compenstated load 9653xxx ROM for 2 days now. No worries. I've not seriously beat on it yet, but it seems to respond to tuning (3d table), idles at 14.7, shoots to 21 on lift off and after some playing, 13-11 AFR in boost.

I have MAF comp and accel enrich pretty doctored up as its freezing cold here, but no stalling and only a little stutter left on tip in, I really just have to figure how to deal with that with this new 3D map.

 

It may need adjustment in the async tip in table if thats the issue.

 

I guess I'm not entirely following what's going on. How can you use load for an axis for VE when the VE is used first to for calculations that lead to load?

For example, post #1 in Johns' thread here states:

This is what I was talking about remembering. The load is calculated by the ECU after the point where John hooks in the VE percentages. So, I'm a little confused on how you can create an axis for load for a VE table before the load is calculated?

I completely understand the scenario where you leave the map VE table in place to calculate the load, then use that load as the axis.

 

Try not to think of it as a calculation but more like a shot in time and you'll get the picture.

Essentially the load for the column is the final load from the previous cycle in the easy version of my code. I suppose load could change a little bit in a normal running environment but its not enough to worry about a in this delayed processing model.

To be frank the map ve one works about in the same fashion as the origional code doesn't leave enough room for me to do a true 3d lookup on the runtime call. So what we are doing is looking it up somewhere else saving it and using that saved value in the function.

 

Ahhh, OK. That's the info that I was missing. Thanks for the reply.