ITB tuning question

[lutorm]

Hey all,

I'm working on tuning my 400cc V-4 Microsquirt V3 setup, and I've gotten it to run pretty well using "%baro", except at wide open throttle, high rpm. In this range, more open throttle doesn't get any higher %baro, which seems to be a clear indication that I need the ITB mode. However, even at wide open throttle %baro is only ~66%. I suspect this is because the MAP sampling is picking up some intake resonance and the MAP sample is lower than the "real" value. The problem with using ITB mode is that, as I understand it, the switch to alpha-N mode is hardcoded to be 90% baro, and my setup never gets there.

If you look at the attached log you can see Lambda1 leaning out severely around 13,000 RPM at full throttle, but not at 66% throttle. The %baro for the two runs are about the same, though.

Any ideas? I don't really want to switch to alpha-N because %baro works well everywhere else, and the bike needs to run from sea level to 9000ft (which is a half-hour away). The only other option I can see would be to switch the MAP sampling to event average, but the MAP signal is so uneven at low throttle that I can't imagine that working well.

Thanks,

/Patrik

[whittlebeast]

You need to get a copy of MLVHD (like a $10 upgrade if you already have registered yours) and go to blended fuel table tuning. I can help you thru it.

I run SD below about 85 KPA and about 18% throttle and bring on AN corrections as the power comes on.

see http://www.msextra.com/forums/viewtopic ... 34&t=63074

Andy

[lutorm]

Interesting idea. Is it possible to use blended tables with dual VE tables? The secondary fuel load option is greyed out for me.

I guess I could switch to use trim tables instead of dual VE tables, my differences in VE are (barely) in range of the trim tables, but I'm not sure the VE analyze works with trim tables?

[Dennis_Zx7r]

Your log is just what I would expect of TimedMinMap.


If the V4-configuration doesn't prevent the use of Average Sample, personally I would try that and see if it fits the expected ITB behaviour better.

[lutorm]

Ok, I'm going to try event average sampling to see what it looks like.

Edit: Not going to work. Even with event average it doesn't get to 90% baro at high rpms. I guess I'll go back to the drawing board and try to use a blended table approach and handle the bank differences with trim tables.

[Dennis_Zx7r]

I just remembered there was someone here with a V4 with a Synchro-/Quadramap not too long ago. Is that you?
In that case it's no surprise to me that baro% is that low, and that ITB may not be the way to go.
Could you upload the log with the averaged Map please, anyway? I actually want to experiment with something like this and would appreciate the data.

[lutorm]

Yeah that's me. Or at least I'm also using a QuadraMap.

Here's a log with the event average.

[lutorm]

Ok, so I converted my two VE tables to an average one and put the differences into the trim tables, and added a second table using TPS load where I added in extra fuel towards as the throttle level goes up. Works beautifully!

The only issue now is that there are a few places where the trim tables don't quite have enough authority to encompass the fueling differences between the banks, but the resulting lambda offset is <0.05 so I think I can live with that.

Logs attached, in case you're interested.

[Dennis_Zx7r]

Good to hear, and thanks for the log!