Living with Code 25/26

UPDATE we have indentified knock disable in HONDA software/bin, am now looking at raising RPM for initial knock self test and then altering knock window/knock map. Will trial this on another HONDA ECU, have also had success with ZOMBIE wiring fix on another bike, we only removed the LOGIC GROUND and CODE 25 disappeared from previous hard fail
 
UPDATE we have indentified knock disable in HONDA software/bin, am now looking at raising RPM for initial knock self test and then altering knock window/knock map. Will trial this on another HONDA ECU, have also had success with ZOMBIE wiring fix on another bike, we only removed the LOGIC GROUND and CODE 25 disappeared from previous hard fail

Impressive! Thanks. This is all bench work, right? Would it work on a 25/26 Coded ECU or would that be preventative for functional ECUs only?

Removed LG.....pulled it out at the harness end, right? No regrounding?

I think somebody here tried it, but no success, probably depending on ZOMBIE stage.
 
UPDATE we have indentified knock disable in HONDA software/bin, am now looking at raising RPM for initial knock self test and then altering knock window/knock map. Will trial this on another HONDA ECU, have also had success with ZOMBIE wiring fix on another bike, we only removed the LOGIC GROUND and CODE 25 disappeared from previous hard fail

What colour has the LG?
 
prett sure it is green, bit tricky to remove pin from connector, have a look at youtube for help, and yes leave the LG disconnected
 
GREEN with PINK stripe by the looks
Been following this with great interest. Almost afraid to go out and ride my '04 st13 (245k km's) without adding more grounds everywhere. But no codes so far,, touch wood. Isn't it ironic that this issue seems to share grounding concerns,,, which is an ongoing issue for Honda's over the years. When I did have some intermittent headlight faults, for a brief period several years ago,,, I bought this recall kit from an old stock ebay source,,, thinking it could only help. It plugs into the red diagnostic port located in the same cavity as the pair solenoid. And all it is comprised of is an additional ground,,, as I recall. I never installed it,,, as my intermittent issues cleared up. I have also been following Darksun's posts on the GL owners site. Thought I would post about the recall kit here,, in case it has any relevance to the threads subject. Thanks for all this work,,, good luck,,, CAt'

kit1.jpg
 
If I get time tonight I'll pull the green/pink wire and see what happens. I also have a new knock sensor wiring harness ready if that does not fix it.
 
I finally got some time to replace the wiring harness for the knock sensors and found that removing the green pink ground wire does not get rid of the error. I have what I think is a good ecu that I have installed and now hoping for the best
 
I finally got some time to replace the wiring harness for the knock sensors and found that removing the green pink ground wire does not get rid of the error. I have what I think is a good ecu that I have installed and now hoping for the best
Did you end up removing the throttle bodies to install the new harness?
 
My light hasn't come on since April when I went to GA-STOC. I expect it'll come back on in a couple weeks at the Blue Ridge Gathering.
 
Bit of an update, had limited success on a similar model just removing Signal Ground wire from plug. Code 25 which was a hard fault went away. Tried it on another one no change.
This an ECU from an 03 VFR800, plugs are the same as ST1300 with 26 pins.

vfr 800 ecu.jpg

VFR 800 doesnt use knock sensor and knock sensing isnt used from factory. But this ECU has the knock sensor amplifier fitted, only 1 though not 2 like ST1300.
Knock amplifier also appears to be a different model/supplier. Hard to see but on the board is also sockets for a second knock amp.

vfr 800 ecu2.jpg

I wasnt surpised to see the knock amp installed but thought there would be 2 of them. Next step is to reflash this ECU with flash from another model (inline 4 with knock sensing)
and see if it works OK. My theory is that Honda makes these ECUs hardware the same but fit to different models with model specific flash/programming .
The CBR 954 ECU also uses 26 pin connectors, dont have one yet to open up, but expecting to find a knock amp as well, CBR 954 engine doesnt use knock sensing.
I guess there may be other Honda 4 cylinder models using the "same" ECU as well from that era. Flashing these isnt easy and interface have to make myself then experiment to communicate with ECU in "debug mode" also known as "the backdoor method". Bought the VFR800 for $20 as an unknown mainly for the connectors and had already been opened up.

Are any of you guys able to open up a ST1300 dead ECU and take some photos for me ? as you are aware even 2nd hand ones are nearly impossible to find.
 
Bit of an update, had limited success on a similar model just removing Signal Ground wire from plug. Code 25 which was a hard fault went away. Tried it on another one no change.
This an ECU from an 03 VFR800, plugs are the same as ST1300 with 26 pins.

vfr 800 ecu.jpg

VFR 800 doesnt use knock sensor and knock sensing isnt used from factory. But this ECU has the knock sensor amplifier fitted, only 1 though not 2 like ST1300.
Knock amplifier also appears to be a different model/supplier. Hard to see but on the board is also sockets for a second knock amp.

vfr 800 ecu2.jpg

I wasnt surpised to see the knock amp installed but thought there would be 2 of them. Next step is to reflash this ECU with flash from another model (inline 4 with knock sensing)
and see if it works OK. My theory is that Honda makes these ECUs hardware the same but fit to different models with model specific flash/programming .
The CBR 954 ECU also uses 26 pin connectors, dont have one yet to open up, but expecting to find a knock amp as well, CBR 954 engine doesnt use knock sensing.
I guess there may be other Honda 4 cylinder models using the "same" ECU as well from that era. Flashing these isnt easy and interface have to make myself then experiment to communicate with ECU in "debug mode" also known as "the backdoor method". Bought the VFR800 for $20 as an unknown mainly for the connectors and had already been opened up.

Are any of you guys able to open up a ST1300 dead ECU and take some photos for me ? as you are aware even 2nd hand ones are nearly impossible to find.

Fascinating detective work!

If I recall, @CruSTy may have peeled a board off. From your pictures, it is obviously not a simple task.

It might help if you could highlight the location where you'd expect to find the Knocks Amp on the board? To focus the peel-off task in case somebody has the skills to attempt it?

'
 
Fascinating detective work!

If I recall, @CruSTy may have peeled a board off. From your pictures, it is obviously not a simple task.

It might help if you could highlight the location where you'd expect to find the Knocks Amp on the board? To focus the peel-off task in case somebody has the skills to attempt it?

'
Just getting the outer box off required using a saw. Once the outer box is partially removed continuous soaking in acetone and gentle brushing over a 2 week period. Even at that I was only trying to harvest the header connectors. Mother Honda definitely does not want anyone to repair, explore, or reverse engineer this. Nearly everything here is proprietary. I am making slow progress on the adapter harness. Would like to find a 2007 bike with the existing code 25/26 for testing. Prefer one without a top box as removal may be required for access to the ECM.
 
I have photos of my ECU from 2005. On the element side you can see 2 thick-layer signal amplifiers (green) from the knock sensors, which are probably damaged. Before I disassembled the ECU I did an experiment and put the ECU in the fridge and after freezing I put it in ST1300 and after starting the error 25/26 did not show up. When the ECU warmed up the errors returned





20240909_133850_resized.jpg

20240909_133921_resized.jpg
 
Before I disassembled the ECU I did an experiment and put the ECU in the fridge and after freezing I put it in ST1300 and after starting the error 25/26 did not show up. When the ECU warmed up the errors returned

Great idea. Seems to confirm the heat sensitivity of the Amp, as @oztrax suggested.
 
Also seems to confirm the heat sensitivity for HOT running bikes that are not tuned well, or wrapped in heat foil under the tank/seat/wire harness etc. ;)
 
This is my 2nd damaged ECU. I bought the bike in winter and everything was OK. Summer came and the temperature jumped to 30 degrees C (86F) and errors 25/26 appeared. That's where the idea of freezing the ECU came from. I replaced the ECU with another one. I managed to ride another season without errors and this season errors 25/26 appeared again. I was riding on mountainous routes on bumpy asphalt and the temperature was of course high 30 C.....
Now I have another idea, I want to ,,transplant'' the entire harness together with a newer ECU from 2008 upwards. They are built differently and supposedly don't break down.
 
Back
Top Bottom