ITB Mode - Megasquirt 2

[SorenDK]

Hi guys

Finally I'm ready to take my megasquirt ITB setup to the next level and try the ITB mode.
The TBs are from a Triumph. Currently I'm running alpha-n but I find it somewhat unprecise at low rpms ie. closed - or close to closed throttles. Above 15% open the car runs really well and balanced. Its a 2,7 with JE 10,1:1 pistons and a DC20 cam

I have good vacuum with closed throttles.

ECU is MS2 so ITB mode is possible - but I could really need some hands-on guidance on the practical how-to actually "convert" from alpha-n to ITB mode

So I ask you guys... have any of you done this? - where to start? - what tables to adjust? settings to alter/monitor etc. etc.

Im no mastertuner but have an allright understanding of how EFI works.

I have the paid version of tunerstudio so no digital limitations... its really just me not knowing where to start and where to end

I really appreciate your help - and practical guidance

Søren

[jpnovak]

Let me guess, your fuel table load bins (% TPS) are evenly spaced?

The trick to tuning ITBs in either alpha-n (TPS) or speed density (MAP) is having non-linear scaling. By this I mean that most of the manifold pressure changes for TPS occur below 30% and most of the manifold pressure changes in MAP reading occur greater than 85 kPa. So, from 0 - 30% TPs should be divided across approximately the first 10 bins (out of 16 rows) of your table. LIkewise the 85-100 kPa bins on MAP loading should be divided across the top 10 bins (out of 16 rows) in your MAP based load. This will give you the resolution you need to smooth out all the hesitations.

Now, ITB mode is a different method that can work well but IMO is very difficult to setup. Specifically the switchover points from MAP to TPS. Basically it divides the engine load calculation to be MAP at low throttle and TPS for high throttle. I will try to explain the process and how to select the load points for the crossover.

First you must have a very well tuned engine in speed density mode (MAP). then you have to look at the fuel table to find the steepest fuel transition on your MAP. This will generally be "around" 90kPa. In some cases it will be 89 and some cases 92. Then run a datalog and generate a graph that is TPS vs rpm. Use the data limits in Megalogviewer to isolate from +/- 2kPa along your steepest fuel transition. This will give you TPS points for your crossover that get programmed into the first ITBMode table. The Z axis can be setup as a VE value for the second transition table.

Once you have these tables programmed in you have to drive the car crossing the transition points. You will be starting from the beginning tuning the car again. You will likely have a few spots in the middle that are too abrubt as you transition tables. The datalog will show if it is a MAP value transition or a VE level transition. Adjust accordingly until smooth.

having said all that... I have setup a few cars with ITB mode. With proper scaling, speed density offers equal or better performance for about 20% of the input efforts.

Good luck tuning.

[trond]

I will follow JPs advice and try speed density too when I install my (triumph) ITBs. Have witnessed what a struggle ITB mode can be while speed density should be a question of resolution. Could use dual maps (VE1 and VE3) ? Easy to do in tuner studio. That gives 32x16 map with plenty of resolution. Any cons with using dual tables for better resolution? have not tried myself and curious if the autotune function in TS will work across maps or do one at the time only.

[jpnovak]

Yes, dual table additive should work. Set one table to a maximum value of say 80kPa and then the lowest row on the second table should be 80kPa. This second table, lowest row will be the overlap and values should be zero all the way across.

I have not tried it but no reason it can not work.

Usually dual table is to split the rpm band where turbo cars come on boost and there is a huge torque gain in narrow rpm band. Same issue with MAP readings on ITBs. Its all about putting more points in the area of the curve where rate of change is the great.

[pampadori]

I find syncing the throttles the bigger issue. Its quite important for low throttle driving. usually i'm around 3-5% throttle maintaining speed on a public road. If one cylinder isn't sync'd right, it feels like a dead miss and is super annoying.

I'm not familiar with ITB mode. Is that something specific to the MS?

[trond]

tested with dual tables today and can report that it works fine. VEAutotune works too, on both tables, but has to be started in two windows one for each table. For a car with plenum it is not required in any way only makes it harder to visualize the 3D map in the end. There is only one AFR target table and one ignition table. Only the VE table is expanded which may be useful tuning ITBs in speed-density mode

Syncing becomes more important the bigger the throttle body ID, or venturi ID. As an example old Ducatis as i had before used 40mm carbs on 300ccm cylinders which is a lot compared to cars, and syncing the carbs (and enough ignition advance) was the key to driveability. I always used Morgan Carbtune that measures vacuum on each cylinder. Only way to get a good result. Anyone tried that, or will I be the first ?