Jump to content

HOWTO: Disable or reset ZF 6 speed long term adaptive learning (KAM)


Roland@pcmtec

Recommended Posts

If you have a freshly rebuilt box with extra clutch packs you will find that the car bangs into gears and is generally horrible to drive. This is as the adaptive learning has to unlearn the clutch wear. It takes approximately ~1500km of light cruising to achieve this. We believe the learning algorithm learns forward (eg learns worn clutches) much faster than it learns in reverse, this is as logically a clutch can't become "unworn" so the algorithm probably was not programmed with this in mind.

To disable the long term learning algorithm set ZF03182 to 0

To reset the long term learning set this to 0, start the car and put it in gear. Turn the vehicle off then set the variable back to 1.

It is reported that it takes less than 100km for the vehicle to learn the clutch wear on a standard box with 150,000kms if this is done. We have another workshop trialling this out on some new boxes shortly to give us some more feedback on how accurate the learning period is and how well this works on a high hp (800hp) build with lots of extra clutch packs.

Update:

This is not 100% confirmed as working on modified boxes. We have had a report that it does indeed work on a standard box however. Most likely this is as there is only a certain range it can learn as one workshop trialled this on a heavily modified box and found they still had to pull a lot of pressure out of the tune for it to stop banging into gears.

  • Like 4
  • Thanks 1
Link to comment
Share on other sites

  • 2 months later...
On 2/27/2019 at 10:50 AM, Roland@pcmtec said:

To disable the long term learning algorithm set ZF03182 to 0

To reset the long term learning set this to 0, start the car and put it in gear. Turn the vehicle off then set the variable back to 1.

It is reported that it takes less than 100km for the vehicle to learn the clutch wear on a standard box with 150,000kms if this is done.

Just a question,

What does the "reset adaptive" do then when ticked if flashing a tune?

Link to comment
Share on other sites

Any chance it could be added to the professional version? Resetting the tcm adaptations is important when moving the zf mechatronics unit (which includes the TCM) from one box to another as the clutch wear will not be the same between the boxes. Interestingly the BMW guys reguarly reset the  zf adaptions however there seems to be no affordable way to do it with the falcons. 

Link to comment
Share on other sites

I understand, but I only want the reset function as a service function, i have no interest in any of the zf tuning features of the workshop version. After calling a few transmission places you could actually release a tailored "zf adaptation reset" tool and sell this to the service market rather than just tuners. However this would need to under cut ford's IDS software in cost per use.  

Link to comment
Share on other sites

1 hour ago, Blackxr6t said:

I understand, but I only want the reset function as a service function, i have no interest in any of the zf tuning features of the workshop version. After calling a few transmission places you could actually release a tailored "zf adaptation reset" tool and sell this to the service market rather than just tuners. However this would need to under cut ford's IDS software in cost per use.  

There would be no return on investment for us to build such a tool unfortunately. Forscan, IDS, FDRS etc all do these tasks much better than we ever will and support thousands of cars for a tiny price. The Falcon ZF is the only TCM that Ford never released a tool for.

To build, test, QA and release a product like that (which would only support a now extinct model of car, the Falcon) would end up costing us close to $100k in development, not only that but we would lose 3 months of development time on our flagship products (the multi tune etc). We would sell maybe 100 copies of the software and I doubt anyone would pay more than $300 for it.

There are a lot of things we would like to do, but we have to say no to about 90% of them as we would end up building 50 tools poorly instead of 1 extremely well.

You can see why the high end scantools cost thousands of dollars a year in subscription fees, it is a huge amount of work to build and maintain even simple products. So if not many people use them, you have to charge a lot of money to make it anything other than a money losing exercise. Where as with something like a tuning product that people use daily and are willing to spend $150-$500 per car in licensing fees, you can suddenly dedicate several full time staff to polishing and constantly improving the tools for multiple years.

Link to comment
Share on other sites

Yeah there seems to some confusion out there about resetting the PCM adaptations also resetting the TCM at the same time. I have just cconceded that the transmission will need replacing and have found one with matching hardware and software tcm codes. 

Link to comment
Share on other sites

  • Roland@pcmtec changed the title to HOWTO: Disable or reset ZF 6 speed long term adaptive learning (KAM)
  • 9 months later...
  • 5 months later...
On 2/27/2019 at 11:50 AM, Roland@pcmtec said:

If you have a freshly rebuilt box with extra clutch packs you will find that the car bangs into gears and is generally horrible to drive. This is as the adaptive learning has to unlearn the clutch wear. It takes approximately ~1500km of light cruising to achieve this. We believe the learning algorithm learns forward (eg learns worn clutches) much faster than it learns in reverse, this is as logically a clutch can't become "unworn" so the algorithm probably was not programmed with this in mind.

To disable the long term learning algorithm set ZF03182 to 0

To reset the long term learning set this to 0, start the car and put it in gear. Turn the vehicle off then set the variable back to 1.

It is reported that it takes less than 100km for the vehicle to learn the clutch wear on a standard box with 150,000kms if this is done. We have another workshop trialling this out on some new boxes shortly to give us some more feedback on how accurate the learning period is and how well this works on a high hp (800hp) build with lots of extra clutch packs.

Update:

This is not 100% confirmed as working on modified boxes. We have had a report that it does indeed work on a standard box however. Most likely this is as there is only a certain range it can learn as one workshop trialled this on a heavily modified box and found they still had to pull a lot of pressure out of the tune for it to stop banging into gears.

If using this method (ZF03182), should I expect the learned adaptive numbers to zero out?

Link to comment
Share on other sites

If I was doing it that is what I would expect to see. There is a drive procedure to get the adapts to relearn faster than just normal driving. Its for BMW but should apply to ours close enough. The attached doc has the drive cycle in it.  Not sure which software version to follow. You'll have to suss that out. Please report back your findings, this would be good knowledge to share. 

ZF_Workbook-transmission-6-8HP.pdf

Link to comment
Share on other sites

1 hour ago, hjtrbo said:

@Mick No shit. The drive immediately after the adapt reset did you notice any flares or shitty shifts. Or was it the same as before the reset. 

Appears to have had no change at all to the way the car drives

Link to comment
Share on other sites

15 hours ago, Mick said:

If using this method (ZF03182), should I expect the learned adaptive numbers to zero out?

Honestly I have no idea. We have never had a ZF to test with and relied on customer feedback which has been very polarising and not consistent. Personally I don't think it works but some people swear it does. 

  • Like 1
Link to comment
Share on other sites

2 hours ago, Roland@pcmtec said:

Honestly I have no idea. We have never had a ZF to test with and relied on customer feedback which has been very polarising and not consistent. Personally I don't think it works but some people swear it does. 

 

So can you reprogram the values to be 0? I'm not talking about resetting memory, but rather just changing the values.

Link to comment
Share on other sites

15 minutes ago, Puffwagon said:

What's the difference between changing a spark value to zero and an adaptation value to zero?

One is a rom value, one is a dmr. Even if you could write to it it would be overwritten. I'm sure there is a way to make them reset but we have no way to test it without a car and a week of spare time. 

Flashing an ar29 to an 8r29 osid then back again to ar29 (or vice versa) should reset them as it changes the os checksums. Same as how changing the pcm osid resets the kam. 

Link to comment
Share on other sites

  • 2 weeks later...
  • 3 months later...
Quote

This is not 100% confirmed as working on modified boxes. We have had a report that it does indeed work on a standard box however. Most likely this is as there is only a certain range it can learn as one workshop trialled this on a heavily modified box and found they still had to pull a lot of pressure out of the tune for it to stop banging into gears.

So are people with built boxes going below factory pressure values to get a nicer shift feel in lower throttle cruise conditions?

Link to comment
Share on other sites

You can't make a broad statement like that.

The builders tuner won't be posting their tips here. You need to use first principals and a fundamental understanding of a clutch to clutch transmission along with good high sample frequency data logs to work out which direction to go with your settings.  

At this stage with regards to the small amount of detailed ZF information publicly available, there is a strong financial / reliability argument to stick with a build shop / tuner combo that has a good track record.  

  • Like 1
Link to comment
Share on other sites

With that said, it still doesn't prohibit you from going your own way and then come back here with data and questions about what you're seeing. In that case, people can relate their own experiences and perhaps come to some direction that will be beneficial to your build. 

Link to comment
Share on other sites

... and slamming the shift time down to less than 150ms without sufficient pressure to get it done. 

The torque ramp back from a shift is still a bit of a mystery that I'm interested in.  There's a few tables that you notice in the OE cals that give some clues.  

Link to comment
Share on other sites

8 hours ago, hjtrbo said:

Agree. 

What's the current Oceania benchmark for a stock ZF that still does daily duties? 9s or 10s something???

If you set the tune up to run 9s (eg just enough to keep it alive) but then daily it with way more torque reduction, eg basically no WOT downshifts etc you could make it live forever. 

Making it run 0 torque reduction on the higher gear shifts will be fast but it will never live long. 

You could fudge this in the multitune by tweaking speed density (to reduce torque reduction) then fixing fueling via tuning correction to get a multitune with different shift pressures. Be a bit hacky but you could make it work. 

  • Thanks 1
Link to comment
Share on other sites

23 minutes ago, Puffwagon said:

All those wot shift hits on the input shaft, along with the power required to run a 9, will eventually snap it. They twist a considerable amount before breaking and will break from outright power, even with no wot shifts.

That sux, ain't nothing that can be done in the tune when she's gonna shear off just from transmitting bulk torque in steady state. 

If I can get my 220,000km standard box to hold 520rwkw+ for another 20,000km hard daily driving I'm calling that a success. 

Link to comment
Share on other sites

RVO do a modified E clutch set up that uses Powerglide clutched which give more clamping area, he confident that it's strong enough to have WOT shifts with no Torque management interaction, although in mine I still have it enabled

  • Like 2
Link to comment
Share on other sites

  • 6 months later...

ok guys ive  got a build  RVO  zfr80-B from Dave   with a 3200 rpm stall      and in  new to this  pcm tech gear  so  where would i start to look at modifying  for the  stall   thanks 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...