OFT - Guide To Dialling In Your OTS Base Tune

[RustySocket]

@Wayno


I am attempting to follow your guide to the letter. In regards to adjusting the DI/PI tables prior to scaling the MAF.


Is it only the total injection port ratio cold table that gets adjusted or do the same changes get made to the warm table and the hot table as well?

[JB86'd]

It's all 3 tables @RustySocket

[RustySocket]

Originally Posted by JB86'd

It's all 3 tables @RustySocket

Thanks, I'm really trying to figure this out on my own and I hate asking stupid questions. Some things seem obvious and others not so much. Hopefully someday I can contribute and help others as well.

[Wayno]

Originally Posted by RustySocket

Thanks, I'm really trying to figure this out on my own and I hate asking stupid questions. Some things seem obvious and others not so much. Hopefully someday I can contribute and help others as well.

Both threads go together and you should be able to copy and paste the injector ratios and smoothed maf scale from the example roms in the other thread. Those 2 things will fix your exact problem.

After you do that you likely won't need any other changes. All that knock above 5000 rpm will hopefully go away or at least become negligible on 100% DI.

http://datazap.me/u/rustysocket/maf-...zoom=3139-3302

[RustySocket]

Originally Posted by Wayno

Both threads go together and you should be able to copy and paste the injector ratios and smoothed maf scale from the example roms in the other thread. Those 2 things will fix your exact problem.

After you do that you likely won't need any other changes. All that knock above 5000 rpm will hopefully go away or at least become negligible on 100% DI.

http://datazap.me/u/rustysocket/maf-...zoom=3139-3302

What I have been attempting to do all along is to gain a better understanding of how the ECU interacts with the car and controls the engine. It's been something just to keep my brain occupied and to learn something new. My tuning days go back to carbs and mechanical advance distributors (VW bugs). The whole concept of computer controlled systems with adaptive learning has fascinated me and this is the first platform that I've come across tools such as RR and OFT that allow a novice like myself to really have access to the parameters that were affordable. Then I found this forum and the wealth of information that is available. There is so much information that at times I feel like I'm trying to drink from a firehose. So I decided to break it down in steps. At first it was just learning how to use RR and what definition files were, what ROM's belonged with my ECU and what didn't. I decided to try E85 and ran three tanks and loved it. It was at that point that the weather begin to change, I had a couple of cold start scares and realized that I needed to know more about modifying tables and actually using RR. It wasn't hard but I felt at that point I'd rather step back to gasoline for the winter. I began reading about actually interpreting datalogs and collecting data. I came across steve's posts regarding MAF scaling and looking at my logs it became obvious that my sensor was scaled on the lean side because my LTFT were always adding fuel. That led me to learning about open loop and closed loop data logs, gathering tons of log files and just sitting there staring at them. Reading guides like yours, steves, badnoodles, and other on RR forums just trying to figure it all out and make sense of it. About that time you posted your guide on optimizing the OTS maps. I was so focused on MAF scaling that I missed the part on DI/PI tables. Anyway, I have begun rambling but the point is that I don't feel I really have a problem with the car and was not trying to _fix_ anything, rather I'm trying to optimize it to be the best it can be. The log file I posted earlier that showed some knock was off a rom I flashed earlier today and it only had about fifteen miles on it before I posted it. All of the other data I have collected for scaling was based off the Stg1 91 2.068 OTS map with no changes made to it as I was using it to gather the data to run through the VGI scaler as part of my learning process. Here is a log with close to 300 miles on it prior to collection. I've set it to show a WOT pull in third gear and in this case don't see the knock. Other than a blip at tip in of KC.


http://datazap.me/u/rustysocket/thir...1&zoom=323-476


I understand that you have done a ton of work and that I can simply copy and paste your tables and likely end up with a perfectly running car. But part of the fun, and the learning for me is in going through the process which you took so much time to detail.


I have gone back to the 2.068 D00C OTS Stg1 91 Rom. I have modified the three DI/PI tables (I'd like to learn more about why we do this, and am willing to read if pointed in the right direction). I have left the MAF scale alone for now. I will drive the car for 50 miles or so to allow trims to settle and then begin collecting logs to process through VGI's maf scaler as I would like to see how close I can get on my own.


But just so I make sure I am getting this all correctly, I could correct my lean condition in OL above 3.0v by using the v86 Smoothed Stg1 scale from Row 8 in your spreadsheet rather than go through the entire MAF scaling process myself and likely achieve the same result even though I have a US 6MT D00C ECU.


Thanks for you patience and time. I really want to gain an understanding and learn. It gives me something to focus on and helps me escape.


If i'm on the right track I will continue to work step by step through the guide you put together and continue to learn.

[gonzo]

Can we sticky this and add it to the hall of fame? If we don't have one we need to make it just for this post!

Thanks, Wayno.

[JB86'd]

Is this a typo? I thought it would be vice versa?

Example:
If I want to lean up one maf cell by 2 percent (around 0.5 AFR), multiply the maf cell value by 1.02.
If I want to richen one maf cell by 0.8 percent (around 0.2 AFR), multiply the maf cell value by 0.992.

[Kodename47]

Originally Posted by JB86'd

Example:
If I want to lean up one maf cell by 2 percent (around 0.5 AFR), multiply the maf cell value by 1.02.
If I want to richen one maf cell by 0.8 percent (around 0.2 AFR), multiply the maf cell value by 0.992.

Yes, it's the opposite to that.

[solidONE]

How are you guys deciding how the load limit tables are scaled? I noticed on the tables at 3800 rpm the load limit is set lower than the fields before and after it. What do I need to look for in the logs and how should I be taking the logs (uphill pulls etc.) to get this dial in more accurately?

I've made adjustments to my load limits that are a bit more suitable than what Wayno posted, but I'm still not quite happy with the result. I'm not getting any knock correction there, so it's not a huge deal. But, I'd like to get this dialed in my E85 rom before I run down the rest of this tank of E85 and switch back to petrol.

Are the load limits set based on the peak engine load seen at a given rpm? If so, I'm seeing as much as 1.23 g/rev at very low RPM right when I open the throttle fully. as low as 2300rpm. Do I dial the limits to those numbers, or disregard those load numbers completely since numbers that high is only seen when going from low throttle openings to fully open?

With Wayno's load limit tables applied (lean spot at 3600) :
http://datazap.me/u/solidone/plm3-fp...6-634&mark=534

After adjustments (no lean spot but really rich at low rpms):
http://datazap.me/u/solidone/plm85c-...zoom=8867-8990

[phrosty]

Originally Posted by solidONE

How are you guys deciding how the load limit tables are scaled?

Though it is indeed a bit of a "black art" right now as Wayno describes it, he does give the process in his post. I can expand on it a little with the workflow I used.

Scale your MAF like normal. If there are any lumps or dips in the resulting table, you've probably got a load limit issue. It should be nice and smooth.

Now for those lumps, look in the log for that MAFv. Find a full throttle sample and a mid throttle sample. If the mid throttle AFR is dead on but the full throttle one is way off, look at the load and RPM of the full throttle one. Compare that to your load limits table, and adjust as necessary.

If the correction for the mid-throttle sample says 0%, showing that the MAF scale is correct, then you can apply the full-throttle sample's correction directly to the corresponding load limit bin.

You can even adjust the resolution of the load limits if you have lumps that won't go away because they're in-between the existing bins.

[solidONE]

Originally Posted by phrosty

Though it is indeed a bit of a "black art" right now as Wayno describes it, he does give the process in his post. I can expand on it a little with the workflow I used.

Scale your MAF like normal. If there are any lumps or dips in the resulting table, you've probably got a load limit issue. It should be nice and smooth.

Now for those lumps, look in the log for that MAFv. Find a full throttle sample and a mid throttle sample. If the mid throttle AFR is dead on but the full throttle one is way off, look at the load and RPM of the full throttle one. Compare that to your load limits table, and adjust as necessary.

So basically trial and error. Adjust and retest. That sort of thing? My maf scale is pretty good right now in CL and medium and low load operation up to 3.0v, so I'm pretty happy with it the way it is. Fuel trims within +- 3% for a pretty broad range of IAT's. I will adjust the load limits .05 g/rev at a time and see what happens.

[Turdinator]

Originally Posted by Wayno

1.1 DI/PI ratios

Instead of PI kicking back in at 5000 rpm, run DI until 7000 and you'll have a consistent LTFT across your entire open loop. OTS E85 tunes already do this. OTS Petrol does not. Do this before maf scaling.

98:
[IMG]

E85:
[IMG]

Is this to compensate for different flow characteristics of the DI and PI systems? If so could you tweak the injector settings to balance them?

[solidONE]

Originally Posted by Turdinator

Is this to compensate for different flow characteristics of the DI and PI systems? If so could you tweak the injector settings to balance them?

At the Di to PI changeover it will change some of the fueling requirements thus your fuel trims may have an abrupt change once the PI kicks in. Also, many have found that running more DI has more of a knock-resisting effect.

[Wayno]

Originally Posted by Turdinator

Is this to compensate for different flow characteristics of the DI and PI systems? If so could you tweak the injector settings to balance them?

That's one reason, the other is knock resistance.