ZF8 shift pressures and timing - for Steven@HP

[ic3man5]

I did a bit of reverse engineering on which clutches are active in each gear as I couldn't find any useful information online for the ZF8HP75. This is for a 2019 RAM 5.7L which I don't believe should matter as I doubt there are variants of the ZF8HP75.

	1	2	3	4	5	6	7	8
A	Yes	Yes					Yes	Yes
B	Yes	Yes	Yes	Yes	Yes
C	Yes		Yes		Yes	Yes	Yes
D				Yes	Yes	Yes	Yes	Yes
E		Yes	Yes	Yes		Yes		Yes
	C	E	C	D	C	E	A	E

[ic3man5]

Edited my response to include downshift tuning also.

[ezhik89rus]

Edited my response to include downshift tuning also.

You wrote earlier
"Shift Pressures -> Downshift -> Offcoming Clutch
Like above, this is named backwards and might actually consider this a bug in HP Tuners or the ZF naming scheme is just strange. I've gone and increased all these pressures by 20% without issues. The stock pressures are too weak if you downshift with the torque converter unlocked IMO. I've reduced the pressure around idle shifts by 50% to smooth out the downshifts. There is little to no "bump" felt now."

Why and why did you decide to raise the pressure by 20%? I think it interferes

[ic3man5]

You wrote earlier
"Shift Pressures -> Downshift -> Offcoming Clutch
Like above, this is named backwards and might actually consider this a bug in HP Tuners or the ZF naming scheme is just strange. I've gone and increased all these pressures by 20% without issues. The stock pressures are too weak if you downshift with the torque converter unlocked IMO. I've reduced the pressure around idle shifts by 50% to smooth out the downshifts. There is little to no "bump" felt now."

Why and why did you decide to raise the pressure by 20%? I think it interferes

I need to update all the downshift stuff, it doesn't work the way I originally thought it did and I still don't fully understand it. I need to spend a couple more hours with it to completely figure it out (5 week old newborn is preventing a lot right now). From what I can tell we don't have full control over the downshift. Oncoming clutch is used just enough so we don't drop RPM (opposite of a flare) and the offgoing clutch is used to smooth out the apply of the oncoming clutch calculated by the TCM. I "believe" the TCM is calculating it purely from shift timing. I have zeroed out both oncoming and offgoing tables and the truck will still shift (really hard though). I wish I would have written it down but I believe a "slow" shift time paired with a high offgoing clutch pressure will throw a code saying it can't complete the shift. Most of the time if you zero out the oncoming clutch pressures it will still work almost 100% of the time as I believe the ECU will attempt to hold the RPMs of the engine. To summarize, increasing offgoing clutch pressure to smooth out hard hits when lowering shift timing seems to be the "basic" way to do things here.

I also know the transmission goes into a "fail safe" mode when the previous shift doesn't happen in time for the next shift (ie. OSS for 2-1 is 200 but we are still doing 3-2 until 180 OSS). It will increase pressure to lock the clutches and then immediately shift really fast for the next shift. This threw me for a loop for a while and I believe its the reason you will feel the 2-1 shift "bump" harder on a quicker brake. We really need a deacceleration factor in these tables to make it shift perfect and we don't have that.

[james1595]

not to throw off the conversation here to much, but im having an issue where my car doesnt want to shift from 1st to 2nd at wot. ive tried just about everything i can think of to fix it. i have it set to shift at 5600rpm, and the last time i was at the track(which its been a few months) it shifted at 6600rpm. is there anything else im missing? i can post the tune i had, and what i have now. ive done a cam swap sense my last cam swap, so still working out a few bugs, and making adjustments on normal shift firmness

[jhcasey]

ic3man5 Congratulations on new born...now we know why you had to take a quick trip to hospital . Have you got any idea what the last row in the clutch table represents? FYI- this table is good for the ZF8HP90 also. Thanks

[jhcasey]

Torque Management -> Upshift -> Forced Torque Intervention
This is a turbine speed delta by engine trq map. After the shift has gone through and tries to apply line pressure at the end. If its botched and still cant make the hand off this is the map that kicks in and pulls trq to get the gear back in line. Its the final trq reduction and a last ditch hope. You see this usually as timing that doesn't come back right after a slip shift. I found out during the early days disabling some or all trq management cant remember but 4-5 slipper was the outcome with that table able to change how much timing at the end was held till the gear came in line.

how is the turbine speed delta determined...i.e. what are the 2-turbine speeds subtracted from one another to calculate the delta? Thanks

[ic3man5]

ic3man5 Congratulations on new born...now we know why you had to take a quick trip to hospital . Have you got any idea what the last row in the clutch table represents? FYI- this table is good for the ZF8HP90 also. Thanks

thanks!

Which table are you talking about? I've learned a little bit more about how the transmission works. The slip times seem to be purely a fail safe, if it slips over the specified time it goes into a fail safe shift mode that basically uses line pressure for the oncoming clutch and uses the offgoing clutch table to offset how aggressive it is. You can see how poor your shift pressures are setup by maxing out the slip times and lower the offgoing clutch pressures a lot (you can safely zero it out, it will just drop RPM until this happens). It also goes into this fail safe shift mode when the gears aren't shifting fast enough (Target OSS shift is about to be missed and its still executing another downshift).

[jhcasey]

Merry Christmas ic3man5...It was the table that you had included in a previous post #46. I had never seen the engaged clutches per gear table with that row but I figured it out ...it's the oncoming clutch.

[durangoRTRob]

I actually took an approach basically identical to what you just described on a test mule last weekend. I had been thinking about it and remembered we had the ability to control the estimated tq values. I figured if I could get the PCM to think the tq value was lower the TCM wouldn't step in and reduce tq. Once I figured that side out on the dyno, then I worked on the tcm shifting pressures as well as how much tq reduction was used for shifts. When finished it drove like stock under normal conditions, and then when hard in the throttle shifting became very firm and quick. I had some others drive it as well and they loved it so I'm going to implement it into one I'm working on now to see how if I can replicate the results on a different setup. Thanks for the advice, good to hear someone else came up with the same approach and has had success with it as well.

This is exactly what I?d like to do, but I?m no tuner? I?ve had my Durango dyno tuned and I hate how the tranny is now? always thought it was just engine tune, but the more I drive it the more I think the tranny needs adjusted for the 274 cam in the 5.7

[DodgeBoy_8HP]

This is exactly what I?d like to do, but I?m no tuner? I?ve had my Durango dyno tuned and I hate how the tranny is now? always thought it was just engine tune, but the more I drive it the more I think the tranny needs adjusted for the 274 cam in the 5.7

Hello durangoRTRob,

I can support your tune! I just sent you a private message.
DodgeBoy ([email protected])

[DodgeBoy_8HP]

How do you fake the trans out dodgeboy_8hp?? Do you lower the max flywheel torque table values and continue to use that table or??? Seems the trans would shift slower and softer if it thought my hemi was making half the torque?????

This aint pro stock or factory experimental.

Or is this is your livelihood and you are on this forum only to advertise services?

Do share. Please do : )

You have a couple of options:
- You could make engine to broadcast a smaller value (in case your engine is delivering more torque than tranny can handle and tranny is "cutting" torque)
- you can increase max allowed torque on tranny side
- you can play around with filling / pressure / timing / gradient / torque management / timing

If not done correctly, tranny will re-learn and your shift will become softer (most likely) or harder.

People need to understand that indivertibly increase pressure will: damage the tranny hardware (short term) and tranny will soft the shift over time due to GLS Adaptation (mid to long term). Than, you will probably have undesired slip & flare.

[DodgeBoy_8HP]

Thanks for sharing. I hear ya on the learning part. Makes it tricky to tune huh? Would suck to figure out something that works then the ?learning? softens it up. You see the log I posted for spoolboy? What do you see? Thx

I see a couple of issues:
- Torque Converter Closing
- Shift Pressure & timing (this is causing the shift to not fully complete before the next one - which causes "engagement shocks")
- Shift Schedule (VERY uneven - specially at 80% throttle and above) -> attached the chart showing your current shift pointsScreenshot 2023-03-02 063932.png

yeah, I would change quit a bit in this tune

[LilSick]

What are you trying to do?

^^^

[DodgeBoy_8HP]

That is what ya saw that you say is messed up in my TUNE. Whatcha see wrong in the LOG ?

What is tc closing? i run the stock maps on the street and leave that solenoid closed at the strip.

Engagement shocks? where do i see those on a log?

Shift Schedule - Yeah, I made it uneven on purpose. Drag racing you want it like that. The car is quicker like that because of how the ratio's are spread out. I am gonna try using the paddle for the top of 3rd cuz it wont shift soon enough no matter what values are in the tables. Optimum rpm for my car is 66-6700 low, 6400 2nd, and 6000 3rd... Top of 4th (what rpm it shifts into 5th) does not seem to effect my E.T.

All the potential modifications I mentioned were based on the log you posted as well.

I personally do not like to keep Torque Converter (TC) open till 4th or 5th gear... you are loosing power and unnecessarily overheating the whole system.

Yes, engagement shocks because transmission does not fully complete the upshift and than starts another one (upshift are made up with roughly 4 phases - based on your log, around the third phase on 1st to 2nd, the first phase from 2nd to 3rd starts -> this creates a system shock)

On shift schedule I have a different opinion than yours: I understand making it to always upshit in the best torque vs. power point. However, yours is uneven... look at 3rd to 4th: it is set to upshit at 5475 rpms.

Anyways, if it works for you, great! I just would do it in a different way

[LilSick]

After you have tuned them how many zf8 cars with 500 or more hp and a 370 gear get thru all 4 stages of 1-2 shift before starting stage one of 2-3 ?

Mine has 91" of rollout

[16AudiS8+]

bumping this just because its a thread with a bunch of information. Appreciate it all.

[ezhik89rus]

I've had a chance to experiment with stuff more. Take this with a grain of salt as this could be completely incorrect and only based on my experience and tuning so far.

Shift Pressures -> Upshift -> Oncoming Ramp
Shift Timing -> Upshift -> Ramp Time
These two seem to correlate to each other. I'm going to call this the Oncoming Clutch apply rate. I've zeroed out these two tables and its actually the "proper" way I'd go about it for tuning the correct pressure since this seems to mask any issues with correct pressures and timing. If everything is zeroed out you can think of the oncoming clutch pressure as "on" and 0 torque (clutch plates just touching) as off, it is instant and as fast as possible. The ramp smooths out how fast this happens. 50ms and a ramp of 0.5 is a good starting point for initial tuning.

I don't completely understand the math behind these two parameters but it seems to be inverted from what you'd normally think. The lower the ramp rate (bar/s) the less it dampens.

EDIT: 1.0 ramp near the upper torque rows is probably better to start on.

Shift Pressures -> Upshift -> Oncoming Clutch
This is the actual pressure that gets applied to the clutches and is what you feel as the final portion of the shift or the RPM differences from what I can tell. If the ramp is too aggressive it makes this stage really hard to tune. 6 bars is about the max you want to use here. I've gone to 9 bars and at the point where it feels like your going to break internals or driveshafts. 0.6-1.0 bar seems like a decent place to be on lower torque areas and 5-6 is a good spot for firm shifts in the higher torque areas. 2.5-4 seems to be the sweet spot for "smooth" shifts.

Shift Pressures -> Upshift -> Offgoing Clutch
This is just how much pressure is used to hold the current clutch while phasing in the next clutch. When going for aggressive tuning, I always zero this out. My theory is if the shift is fast and firm enough you don't need to hold this and its too hard to tell if you are doing premature wear until its too late.

Shift Pressures -> Upshift -> Fill Offset
I don't have as much experience with these tables but this seems to pre 0 torque condition of both plates slipping. Not enough pressure here and you'll see a flare and too much and you'll have an overlap condition. The ZF8HP75 on the 2019 RAM for example I couldn't figure out for the life of me why the 3-4 shift was "binding" until I lowered these. My advice is if you increase oncoming clutch pressure by say 20% you should lower the fill offset by at least 10%.

EDIT: After more tuning, increasing the oncoming pressure, less ramp, and possibly holding the offgoing clutch longer is a lot easier. The values in the fill offset are extremely sensitive. Changing by 0.01 bar was enough to make it flare 100-200RPM during the shift. 3-4 on my vehicle (2019 RAM) the fill offset on the low end were too low by ~10% and the upper I couldn't change from stock without flare so I increased the ramp (pressure not time) to avoid binding.

EDIT2: After more tuning I realized I was getting owned by adapts. It seems the current settings here are only good until a key cycle. The second ignition cycle the adapts adjust everything and will mess up tuning in this section. The stock tune for example on the 3-4 has a lot of values that are way too low on the low end torque and low RPM. (0.10 bar @ 74 lb-ft / 1702RPM). After driving on the first engine start and having flares, the second start would see the majority of the flare go away, third even more. The farther away from "correct" the more key cycles it takes. The same value I mentioned before took almost x3 more pressure to not flare after an adapt reset. I suspect without an adapt reset any added pressure is too much as the adapts have already adjusted for it and its too dumb to know we modified the tables.

Edit3: This seems more like a preliminary stage of the oncoming clutch pressure. If the fill offset is 1.0 bars and the oncoming shift pressure is set to 2.0 bars and there is a ramp the actual pressure applied seems to be 1.0 to 2.0 bars. the fill offset is the "floor" or base of when the pressure gets applied. I don't believe this should be above the oncoming shift pressure. Consider it fine tuning for the ramp and a way to move the shifts closer together.

Shift Pressures -> Upshift -> Flare Adder
I haven't actually played with this one yet but it seems pretty self explanatory. If the Fill offset isn't high enough these tables say how much pressure needs to be added to make up for the slip.

Shift Timing -> Upshift -> Nominal Slip Time
As far as I can tell this doesn't have any say on how fast the shift actually takes place. This held true for my tuning of the 6L80 (GM) also and I'm going to assume it applies here also. This is more or less the time it takes for the Torque Reduction stage to happen, the shorter this time the faster and less time the engine pulls back torque (timing). If you have a flat spot or hang in the middle of the shift, more than likely this is too long.

Torque Management -> Upshift -> Forced Torque Intervention
I didn't play in this area too much but this seems to be the pre-cursor torque management stage. I played around with this by making it so the all the values would equal 0 engine torque (5500 turbine speed @ 350 ft-lb = -350) would actually cause the transmission to never shift even at WOT because it guaranteed the output shaft speed never hit the right speed to shift. It went into a really fun bounce back and forth of the RPMs but never shifted. If I added power to the engine I would account for it here to protect the transmission.

Torque Management -> Upshift -> Main Torque Intervention
This is basically the same as forced torque intervention but is applied DURING the shift instead. Again if I added power to the engine I would account for it here to protect the transmission. From what I can tell on the RAM 200 ft-lb is roughly what OEM spec'd at as reliable; This is a huge increase from the 6L80 as it usually never seen over 40-70 ft-lb on a shift.

EDIT: WOT drops the torque a little bit lower. Roughly around 130 ft-lb which is what I'd expect.



NEW 5/19/22:

Shift Timing -> Oncoming Inertia Mode Delay
I haven't quite figured this one out but reducing this and keeping this LONGER than the offgoing inertia mode delay seems to help the shift timing. I believe this is when the clutch start to engage.

Shift Timing -> Offgoing Inertia Mode Delay
I haven't quite figured this one out but reducing this and keeping this SHORTER than the ongoing inertia mode delay seems to help the shift timing. I believe this is when the clutch start to disengage.

Shift Timing -> Fill Time
I've had really strange delays in WOT shifts and randomly in other spots (very obvious in WOT shifts). Reducing this when increasing the shift pressure seems to have helped here. I suspect too long of a fill time causes the transmission to attempt to reduce or "bleed" the pressure before the shift actually happens causing a really long delay that can be felt. 100ms seems okay here across all temperatures so far. I haven't tested in cold weather yet, this will have to come in the winter months.

Shift Pressures -> Downshift -> Oncoming Clutch
Removed: Update later

Shift Pressures -> Downshift -> Offcoming Clutch
Removed: Update later

NEW 8/6/22:

Downshifts: REMOVED, will update later with better information.

I have found that almost all the pressures in the stock tune are way too low. This is the opposite of the 6L80 tuning GM does where the pressures are set high and the adapts bring it down. The ZF8HP75 has the stock pressures low and the adapts increase pressures.

When to wait for a new article that will help many people navigate the program for tuning. I decided to start all over again in order to achieve the perfect job. But I'm worried that maybe I'm missing something or there is new information ..

[LilSick]

Get the converter locked-up earlier and it can shift quicker. The converter tables from the TRACK transmission settings work really well with a stock converter.

the input shaft accelerates at a rate of X

X varies depending on where you are on the pass

the torque convertor is NOT between the input and output shaft

[LilSick]

can you post of log of one of those passes and the tune that was in it?