ITB vs Alpha-N

[billjam]

I am trying to tune my 3.6 Porsche engine with ITBs and looking at using settings for ITB rather than Alpha-N because MAP readings are a bit erratic at low revs.
Initially, I set the main control algorithm at Engine & Sequential Settings screen to "ITB". No problems, engine runs and responds nicely.

My question is ... what load parameter settings should I use in the General Settings screen?
It seems that Primary Fuel Load changes to "ITB" when the main control algorithm is changed to "ITB", but Primary Ignition Load remains set to "Alpha-N".
Should I change Ignition Load to "ITB" ?

BJ003 2 general settings.jpg

[elutionsdesign]

I would say yes if your fueling seems to be working out for you. Realize that Alpha-N is purely TPS driven and the ITB load uses MAP below a switchpoint you set so if your MAP is noisy at low load you may want get better results with Alpha-N. Use what gives the best response!

[billjam]

I would say yes if your fueling seems to be working out for you. Realize that Alpha-N is purely TPS driven and the ITB load uses MAP below a switchpoint you set so if your MAP is noisy at low load you may want get better results with Alpha-N. Use what gives the best response!

I need to read up on this again. I was under the impression that ITB was TPS driven and that Alpha-N was MAP based.

[hardline]

I'd suggest using 2 tables, an alpha n for the area where there is poor map stability, and then run speed density at higher rpms and load where it is more predictable and repeatable.

You can define the switch point by many different factors. Also, please post up your MSQ. ITB mode alone may work great for you if you can get map sampling setup correctly.

[billjam]

MSQ file attached.
Still a lot to learn, so any and all comments appreciated.

[hardline]

Yup, your map sampling is just an average. Use the 'Use Timed Min' option, and follow the instructions here to set it up. http://www.msextra.com/doc/ms3/MAP_Sampling.html

[billjam]

I haven't had a chance to figure out the MAP sampling yet. I have been trying to resolve more pressing problems related to fluctuating AFR readings which I think are affecting VEAL.
To try and calm down the AFR signals, I have changed the Lambda Averaging Lag Factor to 30 (default was 60) and I have changed the LC-1 response speed to 1/6 second.
Engine starts and runs but is very rough with little power, so something in the tune is causing significant mis-firing. This may be the cause of the fluctuating AFRs.
I have two separate exhaust systems and EGO sensors and I am getting the same signals from both sides.
MSQ and logs attached.

[hardline]

Look into that MAP sampling soon. I did an ITB build on a 4 cylinder and MAP sampling took it from near untunable to predictable and very tunable in just a few mouse clicks.

[billjam]

Made some reasonable progress tonight.
Someone on Pelican forum noticed that ignition timing was set to "fixed" at 10*, so now it is running on the ignition table.
Lowered idle VE settings significantly. Idle is now low and smooth.
Manually tuned out violent surging and misfire in the zone around 3500rpm/50% by increasing fuel in that area.
Killed off serious exhaust popping by dropping VE to zero along the bottom of the table.
Did some tuning with VEAL.
Ran some logs, one attached.
Smoothed out the VE table after VEAL roughed it up a bit.
MSQ attached.
Comments welcome.

[billjam]

Got some spare time today so looked into MAP log and sampling today.
Not really sure what I am looking at here, but it seems to me that MAP is varying between 60 and 66 during a cycle at idle.
To me, that seems reasonably steady, so not sure that I need to go any further. Correct?
What I don't understand is the concept of "MAP window" and why MAP readings are very high and almost instantaneously very low at some point in the cycle.
I'll attach the log which covers idle at about 900rpm, 2000rpm and 2800rpm.
FYI, there is one small vacuum port on each TB and they are all connected to a small manifold which supplies the MAP signal to the ECU.

map log.png

[muythaibxr]

You should switch to the cycle average method of MAP sampling. Then the window won't matter.

[billjam]

You should switch to the cycle average method of MAP sampling. Then the window won't matter.

Thanks, I am pretty sure that is the setting that I am using, however hardline suggested I might benefit from using a window. I guess that if the signal is reasonably steady, the window option becomes redundant.

[hardline]

I am just suggesting what netted better results for me. I tried using a vacuum manifold like you are, and it was too noisy to utilize. I ended up referencing the itb on the same cylinder the wideband is installed in. The difference was night and day. However that was a 4 cylinder which does not have overlapping intake events, where a 6 cylinder will. That alone might change how the map sensor behaves.

[kaeman]

I have itb's and found that the sampling rate Lag factors needed to be as close to 100 as you can be without running into signal noise. my tps, map, rpm, all 100 and my lamda and clt,mat, bat, are at 95... down at the 50 range where initial setting is engine would hardly run. itb's make large changes quickly, you need the ecu to see the changes as quickly as possible. so that it may start making corrections sooner.

my 2 cents worth.

[billjam]

Thanks kaeman, your two cents worth is noted. I will look into that.

At present, the engine is running quite well using the ITB configuration.
I am conscious of the fact that I haven't yet played with the tables that define the changeover between TPS and MAP, but while it is working fine, I ain't gonna mess with it.

The only areas where I am looking for improvement now are:
- cold starting ... takes about two seconds of cranking, hot start is instant
- idle ... not smooth enough
- exhaust popping ... can't lean off low load areas without inducing surging on light throttle
- AFR ... gauge readings are still continually drifting up and down even on constant throttle
- IACV wiring ... isn't making sense yet

Tried tuning with VEAL and got it running reasonably well, then manually edited fuel table to tune out light load surging and to improve throttle response.

Latest msq attached for those interested.