Notes on injector / maf scalining using full time open loop

[Kodename47]

I watched that screencast last night. I need to work out how to do that on my own as I don't have any access to Excel that can use macros. I'm not bad with spreadsheets so it'll just take time to do it manually but it's a bit of a PITA.

Maybe this is a silly question, but what tables do you use to scale the DI/PI and how do you scale them based on the results? My map has very close to stock PI/DI ratios in use so I've got the OL MAF sorted I think, I spent a good few hours yesterday getting it locked down from logs I've had.

[Kodename47]

Also, as I'm wondering whether this is relevant to scaling the injectors/MAF. Any idea what the AF Learning Ranges do?

[mad_sb]

Originally Posted by Kodename47

Also I thought it might be worth tying in the ECUtek tables as they aren't labelled the same as in Romraider, just so I/others don't get wires crossed (mainly me). @jamesm is this correct:

@mad_sb's list:
GDI Pressure Multiplier A & B - DI Quantity Fuel Pressure #1 & 2
Closed loop Target Compensation A & B - Closed Loop AFR Adjustment #1 & 2
Engine Load Compensation (Manifold Pressure) - Calculated Engine Load Compensations #1 (what is #2 - it's a 2x2 table?).


When you stated to zero the Engine Load comp tables, do you mean to have 1 in every cell as I assume it's a multiplier?

I am attaching two shots of the STOCK Manifold pressure engine load comp table. How it looks will depend on how your software scales the table data. RomRaider uses actual percentage, brzedit uses decimal. THese table both represent the same values.

[mad_sb]

Originally Posted by Kodename47

Also, as I'm wondering whether this is relevant to scaling the injectors/MAF. Any idea what the AF Learning Ranges do?

Those are the ranges for fuel trims in mas air flow units (g/s) idle, low, medium, and high if you talking fuel trim language or AFR Learning A, B, C, & D in Subaru language.

[jamesm]

Originally Posted by Kodename47

I watched that screencast last night. I need to work out how to do that on my own as I don't have any access to Excel that can use macros. I'm not bad with spreadsheets so it'll just take time to do it manually but it's a bit of a PITA.

Maybe this is a silly question, but what tables do you use to scale the DI/PI and how do you scale them based on the results? My map has very close to stock PI/DI ratios in use so I've got the OL MAF sorted I think, I spent a good few hours yesterday getting it locked down from logs I've had.

it's Fuel Injection Ratio Port to Direct 1/2/3. There are three maps for different temperature ranges. 100% is port only, 0% is di only.

[Kodename47]

Originally Posted by jamesm

it's Fuel Injection Ratio Port to Direct 1/2/3. There are three maps for different temperature ranges. 100% is port only, 0% is di only.

I understand the PI/DI ratio. You guys mentioned scaling out the injector errors, how do you do that for each?

[jamesm]

Originally Posted by Kodename47

I understand the PI/DI ratio. You guys mentioned scaling out the injector errors, how do you do that for each?

Oh ok sorry I was confused. For port you use the scaler and latency, for di I believe @mad_sb is using the di fuel pressure targets but maybe he'll chime in to confirm as I'm not 100% clear on it.

[mad_sb]

Originally Posted by jamesm

Oh ok sorry I was confused. For port you use the scaler and latency, for di I believe @mad_sb is using the di fuel pressure targets but maybe he'll chime in to confirm as I'm not 100% clear on it.

For DI scaling you use the GDI Pressure Multiplier A&B tables. If you need 30% more fuel you multiply the entire table by 1.3. If you need 90% as much fuel you multiply the table by 0.9. If you need to fine tune a specific maf / gdi pressure range you just adjust those cells of the A&B.

For port scaling, you manipulate the scale number and the latency values. For port if you want 30% more fuel you reduce the injector size by 30%. Lets say your injector scaler is at 200 and you want 10% LESS fuel then you do (1.1*200)=220cc. If you want 10% more fuel, you do (.9*200)= 180cc. At low pulse with you adjust the latency and leave scaling as is.

The Pressure target table is just for fine tuning. Since doing my long tube headers 3600 rpm shows very lean readings no matter what i do with the fuel map so I bumped the target rail pressure in that area. Also dialed out a little bit of pressure up top to keep afr and error flat as you rev it out.

[ztan]

Originally Posted by Kodename47

Thanks guys, I think I'm getting my head round it. Questions:

I have not changed injectors, so is there any need to do full PI/DI sweeps? From this I can just run full open loop to dial in the MAF, correct?

Injector latency, again no need to be concerned? How do you calculate that?

Changing PI/DI ratio will have an affect on AFRs? If I do this with no injector change, is can this be overcome by pure MAF scaling?

I think it is worth doing PI/DI sweeps on stock form.
My car, stock runs ok, but was dropping IAM at part throttle cruise.
Stock DI/PI trims attached; analytic conditions: Daily driving, dMAF/dT < 0.2, ECT >85C, Closed Loop plot of total trims against MAFV.

The curves come together in the low MAF range because at warm idle, DI is used.

I tried touching latency but found that mine was fine stock (change in 0.01 increments, even 0.02 put my PI curve shape off)

I've corrected for around a 3% Port error and the car is much smoother, need to do more testing to verify it works before I post my method (many thanks to JamesM for hints).

[Kodename47]

Nice work. I now need to know how far the PI and DI systems can individually support the fueling requirement for my S/C. Is it worth scaling full pulls or just in closed loop?

I've created my own spreadsheets today to pull in the data required for this.

[Kodename47]

Another thread resurrection from me


Now I've been having a go at the OL and CL MAF scaling I've become familiar with most of the above. I still have a couple of queries.


When people are looking at DI fuelling errors, are you just looking at blanket changes over the entire map? @ztan did you ever complete that? You did day you were going to show how you did it.


As I'm doing this on stock injectors, would a good method be to scale MAF for the PI and then adjust the DI to match? Or does it not really matter which one is the base relationship? As in ztans example above, would you just scale down the PI as the DI is showing less error in the 1st place or would you scale the MAF to the PI and then correct the DI fuel to suit?


I was going through this again and I was trying to work out what the relationship is between the main map and the closed loop fuelling adjustment tables. Is there a best practice way of doing the CL table?


For example, I've included my main fuel map and the CL table. Can anyone explain how these interact in this example? I think I understand but hopefully this may clarify. The bottom map is an Eco map that I run on mode 1, I included this to see if someone can explain what effect this will have on CL fuelling since the CL table is the same for the whole ROM but obviously I can include different fuel maps on each of the 4 modes.

[ztan]

Originally Posted by Kodename47

When people are looking at DI fuelling errors, are you just looking at blanket changes over the entire map? @ztan did you ever complete that? You did day you were going to show how you did it.

Still working. I've had to play with my IAT comp maps first, had trims changing significantly with temerature, then was trying a unified adjustment method using MatLab which hasn't worked; going back to doing things one thing at a time.

[Kodename47]

I found this old thread by @mad_sb which has been a great help so I thought I'd link in here:
http://www.ft86club.com/forums/showthread.php?t=30418

@EcuTek doesn't have the Minimum Open Loop AFR table defined though so I assume we have the stock setting of 14.0.