Anyone had this problem after a DCT reset?

[Mike5100 2,061]

The DCT creep on my 2018 (ride by wire) Africa Twin had come back before it went in last week for its 8k service. I should have pointed this out to the dealer but I thought their electronic gizmo would reset it anyway as part of the service. Anyway it came back yesterday but still with the 'creep'. So as the DCT reset had worked when I first got the bike I thought well I will do it again myself.
HOWEVER - this time it hasn't worked. Well it has in that the creep has been corrected and the gearbox seems to be functioning correctly, but the Dand S symbols are both continuously displayed (so I can only tell whether it's D or S by the feel of the gear changes.).
During the DCT reset procedure everything worked as before until the engine restart when the - sign which is supposed to keep flashing at 2 second intervals then go off when initialisation is complete, began flashing at half second intervals as soon as the engine was started. I tried the reset several times, and once left it for 10 or 12 minutes doing the half second flashing but no joy.
Any ideas anyone?
Mike

[Griff 1,556]

One wonders if there has been a software update which necessitates a slightly different reset procedure. My X-Adv is a 2017 but its reset procedure is different to the previous years NCX in that on mine the throttle is not held open whereas instructions for the 2016 suggest it should be held wide open. I saw where someone suggested disconnecting the battery for 30 minutes and retry. There was mention of the sidestand being down could affect the reset procedure. Someone else suggested the ECU reset first and then redo the DCT reset. One chap did a combination of all the suggestions made and presto all was well again. As a result he has no idea what worked. Personally I have not come across this issue.

[Mike5100 2,061]

Thanks Griff but the DCT reset worked a few months ago on this bike.  Only difference is it's now got a powercommander fitted but I don't think that would be confusing things.  How do I do the ECU reset?

Mike

[Andy m 23,473]

The power commander is transmitting false signals/modifying signals on what, the engine RPM, TPS, Lambda sensors.....? 

 

Seems to me it could well confuse an ECU trying to establish a factory zero mode where it is looking for zero or full travel readings on various inputs. If this is what has changed, going back to what Honda designed would surely be a logical step? 

 

Edit to add. Out of interest I just read the Power Commander installation instructions for the CB500 (can't see AT ones). They seem to use scrag connectors in the injector and TPS wires. These are an obvious source of pain and none functioning, if not now then in the future. I hope you don't have them. 

 

Andy

Edited October 9, 2019 by Andy m

[SteveThackery 3,090]

4 hours ago, Griff said:

Someone else suggested the ECU reset first and then redo the DCT reset. 

 

There is no ECU reset.

[Mike5100 2,061]

Sorted.

Left the bike overnight then just did the DCT reset again.  This time it all worked like last time.  

The only difference from yesterday was that I tried to do it on a bike that had just finished 2 hours of riding whereas this morning it was done from cold, but only of course when the L went off.

Weird

Mike

[KingJames 1,262]

Was wondering if it would be oil pressure related as just topped up, sounds like it could have been.

[Mike5100 2,061]

Hmmmmm .... you've got me thinking as it wouldn;t be the first time that the dealer had overfilled the oil.  Will check later

Mike

[Griff 1,556]

18 hours ago, SteveThackery said:

 

There is no ECU reset.

 

Was wondering about that but put it down to something similar to my former 690KTM where there was a process to reset the ECU by starting the motor from cold without touching the throttle and running it at tickover for 12/15 minutes. My former Triumph Explorer had a similar process. Perhaps there is something similar applicable to the NC motors ? 

[MikeBike 5,005]

Mike, I've had something similar when doing the procedure on my 2015 NC DCT at the end of a days ride. Thought I'd broken it. But like yours after some patience and retries it sorted itself in the next day or so. I don't remember if I disconnected the power for a while but think I might have tried it. Glad it's OK again.

[SteveThackery 3,090]

15 hours ago, Griff said:

 

Was wondering about that but put it down to something similar to my former 690KTM where there was a process to reset the ECU by starting the motor from cold without touching the throttle and running it at tickover for 12/15 minutes. My former Triumph Explorer had a similar process. Perhaps there is something similar applicable to the NC motors ? 

 

I've researched this extensively, mostly at Honda's R&D site where there is a lot of technical information on the motor, plus having read the official workshop manual from front to back.  There is no mention anywhere of such a process.  Now, you might argue that one cannot prove a negative, and such a reset may therefore still exist even though it's not written up anywhere.  But I would argue that if there was a reset process it would be essential that the Honda technicians would know about it.  After all, there are instructions in the workshop manual on changing the ECU, but no mention at all of any "training" or "reset" process afterwards.  This is in stark contrast to the extensive information given about the calibration process you must do after changing the DCT ECU.

 

Incidentally, those other bikes you mention: was that "reset" or "calibration" process actually written down in the official workshop manual?  Did you see it with your own eyes?  I ask simply because there is all sorts of information on the Internet that began as someone's assumption but gets repeated everywhere until it becomes a "well known fact", even though it has no basis in reality.  

 

What complicates it is that you will find plenty of people who swear their bike runs loads better after doing some reset or recalibration process.  Unfortunately it's impossible to eliminate "magical thinking" or the psychological factor - you will perceive it runs better if that's what you expect, or want.  It is basically impossible to rely on subjective data in this kind of scenario.  After all, my bikes run at their very best on the way back from the MOT station with a pass certificate in my pocket!

[Griff 1,556]

1 hour ago, SteveThackery said:

 

I've researched this extensively, mostly at Honda's R&D site where there is a lot of technical information on the motor, plus having read the official workshop manual from front to back.  There is no mention anywhere of such a process.  Now, you might argue that one cannot prove a negative, and such a reset may therefore still exist even though it's not written up anywhere.  But I would argue that if there was a reset process it would be essential that the Honda technicians would know about it.  After all, there are instructions in the workshop manual on changing the ECU, but no mention at all of any "training" or "reset" process afterwards.  This is in stark contrast to the extensive information given about the calibration process you must do after changing the DCT ECU.

 

Incidentally, those other bikes you mention: was that "reset" or "calibration" process actually written down in the official workshop manual?  Did you see it with your own eyes?  I ask simply because there is all sorts of information on the Internet that began as someone's assumption but gets repeated everywhere until it becomes a "well known fact", even though it has no basis in reality.  

 

What complicates it is that you will find plenty of people who swear their bike runs loads better after doing some reset or recalibration process.  Unfortunately it's impossible to eliminate "magical thinking" or the psychological factor - you will perceive it runs better if that's what you expect, or want.  It is basically impossible to rely on subjective data in this kind of scenario.  After all, my bikes run at their very best on the way back from the MOT station with a pass certificate in my pocket!

 

Never saw a manual description of the KTM process with my own eyes but it was passed on to me by a well respected KTM mechanic and was otherwise well documented. The difference it made to the 690 was that it more or less "cured" its tendency to stall at low revs when coming to a halt and I certainly experienced this. It also helped with starting issues sometimes. However it wasn't always possible to complete the timeframe for the process as the motor would sometimes stop before completion. When that happened one had a problem of a different nature and a prelude to the various KTM 690 fuelling issues that were an absolute pain to find. 

[Andy m 23,473]

On the brake systems, the often repeated "reset" is just clearing the error memory.  You may as well photograph the warning light, delete the picture and save the time/money involved in grounding the K-line of whatever sequence they try.

 

Andy

[embee 7,288]

Good points made by Steve at al.

 

One aspect of the DCT clutch initialisation routine is, If I recall correctly, that the fan should not operate while it is doing it's stuff. I can only guess that it can potentially result in voltage kicks and/or slight speed variations etc as the electrical load is switched on/off. I may be wrong on this, it's just something in the back of my mind, I can't be bothered to trawl the manual to find it. Some things may simply be treated as failures in the process as a matter of course (if fan switches on during the process it is treated as a fail for example, whether or not it actually affects anything in reality, the NC fan is controlled by the ECU not a temperature switch).

[Mike5100 2,061]

On 10/11/2019 at 12:10, embee said:

Good points made by Steve at al.

 

One aspect of the DCT clutch initialisation routine is, If I recall correctly, that the fan should not operate while it is doing it's stuff. I can only guess that it can potentially result in voltage kicks and/or slight speed variations etc as the electrical load is switched on/off. I may be wrong on this, it's just something in the back of my mind, I can't be bothered to trawl the manual to find it. Some things may simply be treated as failures in the process as a matter of course (if fan switches on during the process it is treated as a fail for example, whether or not it actually affects anything in reality, the NC fan is controlled by the ECU not a temperature switch).

Hmmmm ... That's interesting Murray.  When I tried to do the dct reset after a 2 hour ride I noticed the fan cutting in and out quite often.  The reset didn't work.  The next day I let the bike run for a while - I was going to wait until the fan came on but thought I would try it before that.  When I first started doing it the L came on (for temp too low) then next time i did it about 30 seconds later, no L came up and the process worked, and I don't remember the fan kicking in at all, so maybe you are right.  (maybe we need the engine warm but not too warm)

Mike

[Stu Brown 50]

Is the grade of oil correct? I think the oil pressure is affected by oil temperature and viscosity and the sensors use the pressure as an input into the gear selection. When I replaced the oil on the Integra with the recommended 10-30 grade it ran worse than with the previous oil ( which was in  the bike when I bought it. and I think was 20- 40 from the service records) A reset improved the overall running straight away.

[SteveThackery 3,090]

10 hours ago, Stu Brown said:

Is the grade of oil correct? I think the oil pressure is affected by oil temperature and viscosity and the sensors use the pressure as an input into the gear selection. When I replaced the oil on the Integra with the recommended 10-30 grade it ran worse than with the previous oil ( which was in  the bike when I bought it. and I think was 20- 40 from the service records) A reset improved the overall running straight away.

 

FWIW, mine wouldn't change smoothly when cold on 10W-40.  Changing it to 10W-30 cured it.