Very odd code 26 problem, might be fixed or something other than ECU

Joined
Sep 26, 2006
Messages
121
Location
Arnold, mo
Bike
2005 ST-1300
I bought a 2005 9 days ago. Test rode it and no problems and everything was fine for the first 90 miles. Then the code 26 started and for the next 6 days it would come on as soon as I got over 4400 RPM for a few seconds. So I got on here and read that it was probably the ECU, so started looking for one. Yesterday morning I got up and pulled the plastic off and checked the continuity on both wires from the sensors to the ECU and it was good. I swapped the sensors right to left and put it all back together. The first 15 miles no light(even though it should have come on) but it finally came back on, I stopped and checked it to verify that it was a 26(it was a 26) and not a 25. For the next 100 miles it would come on but started to be harder to get it to trigger. Now the last 50 mile I have ridden it, which is the last 25 miles of my ride home last night and this morning riding to work, I can not make the light come on and I have tried at least 20 times.

Has anyone got an idea of what is going on, does the bike realize that I will love and take care of it and decided to behave?
 
It has been my understanding that this problem is random and relevant mostly to the '07's and earlier. That does not make it any easier to find an ecu, but if yours is going south, cycle junk yards might be more likely to have one.

Alas, as much as we love these bikes, I don't think they are capable of loving us back. For a start, you might read this...
 
More than likely it is the ECU. I went through the process of changing the wiring harness and sensors to no avail. Ended up replacing ECU with a used one from eBay. No problems since.
 
Had the bike been sitting for a while? I ask as code 25 and 26 are knock sensors. Maybe the engine had some carbon build up (or something else that would cause pre-ignition) and riding it remedied whatever was causing the knock.

Neal
 
More than likely it is the ECU. I went through the process of changing the wiring harness and sensors to no avail. Ended up replacing ECU with a used one from eBay. No problems since.
I think so but just weird that all of a sudden it stopped.
 
Had the bike been sitting for a while? I ask as code 25 and 26 are knock sensors. Maybe the engine had some carbon build up (or something else that would cause pre-ignition) and riding it remedied whatever was causing the knock.

Neal
Probably, I got it from a guy that did not use it much the past year and he got it from his retired neighbor that had not been using it much.

What is the best fuel treatment to run through it?
 
Had the bike been sitting for a while? I ask as code 25 and 26 are knock sensors. Maybe the engine had some carbon build up (or something else that would cause pre-ignition) and riding it remedied whatever was causing the knock.

Neal
The Knock sensor is there to prevent knock and acheive best performance by allowing the timing to be as advanced as possible. When the light comes on it retards the timing to a safe and predetermined setting.
 
The working consensus is the vibration from riding is rattling the ECU breaking the internal solder points. Honda manufactured the ECU as a sealed unit so there’s no way to slice it open and fix it.
 
The working consensus is the vibration from riding is rattling the ECU breaking the internal solder points. Honda manufactured the ECU as a sealed unit so there’s no way to slice it open and fix it.
it has nothing to do with carbon
 
Well since I can not get an ECU, I am willing to try anything. If I had a spare I would try to drimel it open and see what is inside. I did fix a sealed GM electrical component that they claimed was not serviceable with a drimel and some electrical tape. I also resoldered a BMW computer board one time and got it working, so put those together and maybe I can fix this.

Thank you all for the help.
 
Maybe the engine had some carbon build up that would cause pre-ignition ?

Neal


No doubt carbon deposits can cause knocking.

I ask as code 25 and 26 are knock sensors.

Neal


But there are misconceptions about the FI light coming on with Code 25/26, like here:

When the light comes on it retards the timing to a safe and predetermined setting.


The light is not indicating that the ECU is retarding ignition.

The light is indicating that the knock sensor is defective (or at least that the ECU thinks it is defective), which implies that you could be riding with undetected damaging knocks,
 
The working consensus is the vibration from riding is rattling the ECU breaking the internal solder points. Honda manufactured the ECU as a sealed unit so there’s no way to slice it open and fix it.


Sealing, potting, circuit boards is common practice, not just a Honda whim, and repair shops are familiar with the diverse methods used to remove/strip/dilute the diverse potting compounds.

Problem here is that all the specialized shops all over the world that have looked at the cause of the failure over the years have stated they could not fix it.
 
The light is indicating that the knock sensor is defective (or at least that the ECU thinks it is defective), which implies that you could be riding with undetected damaging knocks,
So what does the ECM do when it sees a malfunctioning knock sensor beside turn on the light?
 
The light is indicating that the knock sensor is defective (or at least that the ECU thinks it is defective), which implies that you could be riding with undetected damaging knocks,
if the knock sensor is faulty, the ECU will know this and go into a sort of safe ignition advance mapping... google it, and Matstlouis never mentioned any carbon/preignition knocking.
 
Last edited:
if the knock sensor is faulty, the ECU will know this and go into a sort of safe ignition advance mapping... google it, and Matstlouis never mentioned any carbon/preignition knocking.


Ah, OK! I didn't know that what the ECU was doing when the sensor is faulty and I was not able to find a Google source on that either.

My understanding was that what was being talked about here was the more common case where the Knock Sensor is NOT faulty but the ECU is falsely coding that it is.
 
Top Bottom