TD_D mod journey. From stock to the 'bastardmobile'

[HolyCrapItsFast]

I was worried that might the case - it's had a lot of mechanical work in the last 2 weeks, so maybe a clean slate is not a bad idea. Are you worried about the latency and scaling as well (still scrunching the data in the background, so not yet clued as to the general shape of the MAF scale...)

You will find that it is doing exactly the opposite thing. Looks like latency needs to be increased along with scaling. The good news here is that the injector scaling is much closer to theoretical which is what I wanted to see. I think we need to start with a clean slate.

What I will do is set latency to the published values and set injector scale to theoretical. Then apply stock maf scale and a 1.30 multiplier. Then we should start from there.

 

[Td_d]

Ok - let's do it! Thanks for sticking with this for so long :lol:

 

[HolyCrapItsFast]

My pleasure. I was worried you would want to give up on me. I will make this one go faster.

Here it is... Series two Episode one... http://www.drift-ready.com/Files/TD_D/2-26-2011.rar

I applied new injector scale, fresh MAF calibration, reverted to stock load compensation tables and reverted to published latency.

You can load this and take a log. You don't have to give me such a long log this time, 15 minutes tops. Again smooth throttle input with both highway and town driving. Also try to vary the load and get some higher rpm's in there also

 

[Td_d]

Here it is... Series two Episode one... http://www.drift-ready.com/Files/TD_D/2-26-2011.rar

The Empire Strikes back? :lol:

Thanks - will load it up and do a a quick log.

 

[Td_d]

Question of clarity - at least you know I'm paying attention to the detail - you only used the learning's this time (not learning + correction) - any specific reason for that? You trying to base line it and see the correction from there?

 

[Td_d]

Alrighty, here's the next log - longer than I expected (couldn't find parking :lol

http://dl.dropbox.com/u/3099895/TD D 26 Feb 2011 Log A.rar

 

[HolyCrapItsFast]

Question of clarity - at least you know I'm paying attention to the detail - you only used the learning's this time (not learning + correction) - any specific reason for that? You trying to base line it and see the correction from there?

Wow you noticed that... Your good!

Yes The first calibration I usually do using the rough correction only. This usually gets the long term trims closer to zero. Then the subsequent runs I focus on both for a finer adjustment.

I will have a look a the log. Most likely you will have another map on Monday.

 

[HolyCrapItsFast]

I never thought I would say this but there is to much data :lol:

Here is another adjustment to MAF calibration. :tup:

http://www.drift-ready.com/Files/TD_D/2-27-2011.rar

 

[Td_d]

I never thought I would say this but there is to much data :lol:

http://www.drift-ready.com/Files/TD_D/2-27-2011.rar

Chuckle, sorry! What can I say, I love to drive :lol:

Thanks, will load this now, and log away.

 

[Td_d]

Ok - here's 3 logs (way to work, way to meeting, way back) - the funny thing is I crunched the morning log (17) and saw some deep negative trims - -20% and there abouts - but on crunching the mid morning one (18), the trims are still negative - but much more reasonable (around -8% to -10%).

http://dl.dropbox.com/u/3099895/TD D 28 Feb 2011 Log A.rar

 

[Td_d]

Looking at Airboy's "Engine Load Comp MRP R3.xls". My head hurts already... you'll have to take me through this!

 

[HolyCrapItsFast]

Ok - here's 3 logs (way to work, way to meeting, way back) - the funny thing is I crunched the morning log (17) and saw some deep negative trims - -20% and there abouts - but on crunching the mid morning one (18), the trims are still negative - but much more reasonable (around -8% to -10%).

http://dl.dropbox.com/u/3099895/TD D 28 Feb 2011 Log A.rar

WTF! We are not making any head way here and I don't know why. Let me make one more rough correction and we will try it and see what happens.

Looking at Airboy's "Engine Load Comp MRP R3.xls". My head hurts already... you'll have to take me through this!

I don't fully understand this my self. I just plug in the numbers and use the data it spits out. They have a big thread about it on RomRaider forums. I have to find that link though. I understand what the load compensation tables are doing but i don't know what the spreadsheet is doing to derive the results.

 

[HolyCrapItsFast]

I wonder what flights to Johannesburg cost these days

 

[Td_d]

WTF! We are not making any head way here and I don't know why. Let me make one more rough correction and we will try it and see what happens.

Yeah, I'm not getting why the trims are so wild at the bottom end. Also - here's another one I logged now, have a look at this graph (trims are not as bad as first log, but still quite deep at idle):
View attachment 815

 

[Td_d]

I wonder what flights to Johannesburg cost these days

Hahahaha... damn expensive, I'll tell you that much! At least the weather is awesome - 30 deg c today, damn hot...

 

[HolyCrapItsFast]

Yeah, I'm not getting why the trims are so wild at the bottom end. Also - here's another one I logged now, have a look at this graph (trims are not as bad as first log, but still quite deep at idle):
View attachment 815

At least it is consistently low from log to log. That"s a good sign. I'm compiling a new calibration. I'll have something soon. I'm applying only A/F Learned for a rough correction.

 

[HolyCrapItsFast]

Okay. Try this. Rough corrections to MAF calibration.

Reset the ECU prior to flashing. When you flash it let it idle for a few minute to allow the trims to settle, then start logging. You should be doing this every time. :tup: Hope this brings us closer to where we need to be.

http://www.drift-ready.com/Files/TD_D/2-28-2011.zip

 

[Td_d]

Yup, been resetting and letting it warm, then idle - will load this and log again tomorrow morning, thanks Holy.

Out of curiousity - when we use the wideband to patch up the open loop - is it much the same process? I gather what you're really trying to ascertain is whether the AFR / RPM / Load array in reality matches the table, correct?

 

[HolyCrapItsFast]

Exactly...

In open loop we can't use the trims to derive the MAF calibration because open loop does not reference the O2 sensor. What we have to do is plot out the results of the log along with the wide band readings and compare that to the open loop fuel table. Then make adjustments accordingly

So in other words, If your AFR is 13 for a given RPM/Load, you want to see a coresponding AFR for the same RPM/Load in the log. There are tools to help with this but I usually just sample each load and adjust them manually.

 

[Td_d]

Alrighty, it's already looking much better: http://dl.dropbox.com/u/3099895/TD D 29 Feb 2011 Log A.rar
View attachment 816