ive attached most of the stock roms including K01I to this post at end


crab the K01I stock rom and an OFT K01I definition from the openflash support web site, that will let you open the K01I stock rom then copy in values into your OFT K00D which is rearly a K01I calid hacked

Thanks, but I don't see any attachments?

arrh forgot link

http://www.ft86club.com/forums/showthread.php?t=62332


direct link to K01I rom its zipped , so unzip first

http://www.ft86club.com/forums/attac...0&d=1504331012

Really appreciate the help here Steve. I got your ROM and went to OFP's site to get the def and couldn't find it on the download page for the MY'17 EUDM AT. They only have the ROMs.

https://www.openflashtablet.com/shop...-eu-6at-v4-03/

If you happen to know the address for the table in the US or EU stock ROMs, I can just pull the values out of the binaries. Thanks.

If you download the tune pack im pretty sure the def is in the zip file
If not i have thte def

Surprisingly it doesn't. The US AT files had a def. Is it region locked?

Anyway I searched the stock binaries and found the y-axis RPM values and was able to verify the table entries are the same in the OTS ROM, so that isn't the issue.

Have a look at you closed loop fuel tables ( they are an offset from 14.7) and compare the afr in samecrpm load range in the open loop tables. Note the large differences.


Now the v4 tunes have zeroed the ol\cl timers and reduced the throttle transition for ol\cl.


The ol tavle looks a bit rich to me a lower rpms and loads


Id graduate the closed loop afr tables both of them to more closely resemble the open loop after avout load 0.6 and rpms over 3000. This will make the teansition from cl to ol smoother

There are 3 tables that play together with transient ignition timing retard. It's active all the time and not just high rpm shifts.

https://i.postimg.cc/SsD2hHrY/Screen...t-13-55-31.png

Subaru changed the strategy with MY17. The large table "Cylinder fill percentage" is the same for both MY13 and 17.

If we take an example, 2400 rpm and you drive with 5% throttle and increase the throttle to 10%, it's a difference (or delta) of 21.3 to the cylinder fill percentage.

For the MY13 a delta of "4" is required to activate transient ignition timing (small table left) and it will take out timing according to the load. E.g. load 0.30 and 2400 rpm = -14.844 deg timing.

For the MY13 a delta of "8" is required to activate transient ignition timing (small table right) and it will take out timing according to the load. E.g. load 0.30 and 2400 rpm = -29.661 deg timing.

Obviously, this is an extreme example and the result is the same as the delta is so high. But imagine you drive with 7.9% throttle and increase it to 9.4% (made up numbers didn't check what that would result in). But you get the idea, one table with just 4 minimum CFP delta might activate where with 10 it won't.

So the new strategy is for it to activate at a higher delta, but then take out significantly more timing.

If you use the new transient table with the old Minimum CFP delta table, you will have bucking, especially around 2400 rpm and low throttle (causal) driving.

If it takes out too much timing, it will have the same effect as hitting the soft rev limiter (works the same by taking out a lot of timing). It will feel like misfires and you get the bucking. It's not just a hypothesis, this is logged and tested on several cars that had bucking issue due to mismatch of those tables.

Here an example of bad bucking which went away after using the new Minimum CFP delta. Ignition timing drops from 30 down to about 8 deg a couple of times. Of course, it gets more complicated than just reading straight down the tables as we did in the examples above as load is going to jump around too:

https://i.postimg.cc/HxGdRzXD/Screen...t-12-00-49.png


Obviously, I can't say if this is your issue but it's worth looking into. The example and tables above is manual transmission. I don't have a K01I ROM and I don't have that much experience with automatics, so your issue could be something different.

Time constant definitions for transient timer?
 

Does anyone have a definition for the time constant on the lift-throttle retard timer? (I'm running a B01C / OFv3 beta based tune, though substantially modified.)

In my runs, I've seen frequent cases where the lift-throttle retard works as Tor just describes, but for not quite long enough, and then I get a knock correction event just as the advance returns toward normal. I'd really love to add another 20ms or so to that retard timer, particularly if it happened to be RPM dependent.

Throttle blips have also been problematic for me. For example: https://datazap.me/u/baront/log-1541...8-394&mark=348 Here the retard logic seems to be kicking in for the open-throttle stab, may or may not be continuing for the lift-throttle release, and then I get a knock correction event.

The steady-state advance tables that I'm running are pretty good, though I've made them a bit less aggressive than I could (for the steady-state) in order to partially work around the transient problems. But I'd personally prefer to address the transients though the transient mechanism instead.

I essentially made the CL tables match the OL and it didn't fix the bucking, but it did improve drivability in my opinion. Feels like there's more power and the transition to OL is almost undetectable. Weirdly enough, fuel efficiency seems to be the same or better.

I have the tables on the right and it is taking a lot of timing. What are the impacts of increasing the min CFP delta table entries so that it doesn't do it as much?

What's problematic? It retards 0.69 deg - that's next to nothing. If you want to tune your car to have no FBKC and no FLKC you are going to de-tune it immensely.

That these logics activate once in a while is absolutely normal.

I described it in the previous post. If you increase the CFP delta is doesn't take out a lot of timing for small throttle changes where it isn't necessary.

I'm not sure it's your issue though, as I would be surprised if your tune doesn't already have the MY17 CFP delta table... given that it has the MY17 transient retard.

As @Tor said, is it really an issue? Unless the retard is of a large magnitude or causing drivability issues then I'd leave it as is.

When I was seeing events in the -2.5 range, those corresponded nicely to some low-to-mid-throttle bucking. But yea, now that I'm down to this level it all feels fine. I'll stop trying too hard... :D

That said, I'm totally okay with smoothing out my low-to-mid level performance by "filing off" a few better performing lumps, as long as it doesn't make the efficiency really suffer. Of course, WOT is another matter entirely: when giving it the full beans I'm okay with a few lumps. As it stands I've gotten my tune about as smooth as the OFT 4, better sub-3000 RPM cruising performance, and basically back to stock fuel economy. (OFT 4 was always a couple MPG hit.)

Adopting the MY-17 Minimum CFP Delta and more aggressive Transient Retard tables has really helped the mid-range drivability of my USDM '14. Thanks for sharing those hints!