2018-2020 Roush F-150 Transmission Tune PSA

[engineermike]

I've been working for months on and off to fix some idiosyncrasies in the Roush F150 [presumably] 2018-2020 OSID KTGV2 trans logic. There some errors in the logic that can be quite challenging to tune around so this is to help others who might run across the same thing.

1. Table 38234 "Trans rpm for normal engine speed limit" is obviously wrong, actually works backwards, and will trigger premature shifts. If you raise this value to 7000 or above, it will cause the trans to short-shift. Default is 6550 and if you try raising shift points too much you will encounter it. Ive had luck lowering it to 5000 rpm and it doesnt seem to interfere.

2. Base downshift table 21484 has some very strange issue where the 100% row is seemingly corrupted or used incorrectly by the logic. The 100% row affects the whole table, as far as I can tell. The truck will not follow the commanded downshift speeds and pedal positions and the 100% row is somehow causing it. After considerable trial and error, I've found that filling the 100% APP row with 0s fixes the problem. I admit this sounds very unlikely but trust me when I say I tried almost everything and this is the only solution that worked for me. Other shift maps such as Tow do not have this problem. I even copied the Tow data, which worked correctly in the Tow table, into the Base table and it did not work the same as it did in the Tow table.

3. There is some sort of limiter that kicks in on the 4-5 and 5-6 WOT shifts (possibly more) that shuts down the power for several seconds or more. This was when attempting to shift at 7050 rpm using the WOT engine rpm table 9690 to trigger the shifts. So far, I've had luck lowering these shifts to 6600 and triggering using the OSS table 21468. I'm not sure at this point if the lower speed shift or the triggering table is what fixed the problem. I tend to think it's the difference in triggering table based on the other logic errors. And yes, I've tried raising all sorts of limiters and nothing else worked. All of the shifts from 1-4 work fine at the same 7050.

Hope someone finds this useful.

[kenny c]

Thanks Mike. How are the Roush 10r’s different from say a “normal” 10r trans? Just curious if this info could apply to all tune files and hardware or Roush specifically. Thanks!

[engineermike]

The Roush 10r is physically the same as the base truck. I had assumed these issues were specific to the OSID KTGV2 but I suppose it’s possible other 18-20 f-150’s could exhibit the same oddities.

[gmguerra8]

My Roush F150 has a harsh 3-5 shift at times and it has a huge delay when shifting from Reverse to drive. Any way that can be fixed?

[engineermike]

Honeslty I just did away with 3-5 skip shift. I prefer it this way. On the slow engagement, I havent tackled that. Mine is fairly slow as welll and may just be a physical characteristic of the 10r.

[skylinedan]

I,m gonna have to try the trans RPM limit, one thing I have run into with my Fusion Sport, power goes up and the shifts get way to short. Been raising the shift tables, but it really needs more. Years ago and guy said to just set them all at 7000, I've run them both ways, stock and 7000, get the same results.

[engineermike]

I feel a little stupid now but I may have solved 3. Parameter ECM 2364 might have been causing the 4-5 shift "shut down" I was experiencing. I had originally figured that the 2364 parameter wasn't causing the issue because the 1-2, 2-3, and 3-4 shifts were all working fine at the same rpm. But I've done some limited testing and it appears to allow 4-5 and 5-6 shifts above 7000 rpm now.

[gmguerra8]

Mike can you post the tune file you have worked on for the Roush F150. I want to have a reference in order to work on mine.

[engineermike]

I've spent probably a year working on this tune. But now the truck runs 11.1 at 125 (dragy) with just tuning and a few minor bolt-ons. This is a crew cab Roush Nitemare. I have it working as a flex tune, plus boost-by-ethanol content and boost-by-drive mode. However, it's in PCMTec format. If you have specific questions, I'm glad to discuss.

[gmguerra8]

I am only trying to smooth out the transmission on the truck during normal driving. It tends to have hard shifts on light throttle only on certain times. I feel that it could be related to the errors in the Roush logic that you mentioned.

[engineermike]

Roush copied all the mustang gt shift characteristics then only activated character 1 for all modes. If you’re looking for less aggressive shifting then consider switching it to character 0.

[gmguerra8]

It is only the 3-5 shift that's hard intermittently. In some occasions, it shifts smooth as silk. It only happens at part throttle. I tried to remove the skip shift also. Without the skip shift, it makes the 3-4 shift really hard.

[engineermike]

Could also be sticky solenoids or the cdf drum issue.

Either way, it’s so easy to try different characters I think I’d start with that.

[gmguerra8]

Can you explain how the downshift map table 21484 is causing the issues. Mine has a hard time downshifting from 10th or 9th at light throttle at highway speeds. Could this be the cause of the issue?

[DodgeBoy_8HP]

Can you explain how the downshift map table 21484 is causing the issues. Mine has a hard time downshifting from 10th or 9th at light throttle at highway speeds. Could this be the cause of the issue?

Share your tune.

For remote tune support: [email protected]

[engineermike]

Can you explain how the downshift map table 21484 is causing the issues. Mine has a hard time downshifting from 10th or 9th at light throttle at highway speeds. Could this be the cause of the issue?

There is clearly some OS level error regarding table 21484. It drove me nuts for a few weeks. I wound up figuring it out by zero'ing out large sections of the table to see how it reacts. The dead giveaway was when I zero'd the entire table except the 100% app row and it still downshifted somewhat normally at part throttle. Another dead giveaway was when I copied all the tow upshift/downshift/coastdown tables into base mode and it acted totally different than tow mode. That lead me to zero the 100% row and populate the rest of the table which actually seems to work. In your shoes, I'd start by using drastically high numbers in the entire 9th gear column and see if it changes or improves. If not, then try setting the 100% app row to 0 and see if that helps, then go from there.