Need help diagnosing FI code 26 on 2003 ST1300A

StMark

Firstly, I know that according to Honda code 2-6 indicates a faulty right side sensor but the bike shop has replaced that and still the code persists. Let me give the recent history:

Four months ago, idle wasn't holding until the bike warmed up and this was resolved with a new Map Sensor. Shortly after that, the FI light started coming on once I got above 4000rpm for more than a few seconds. The other symptoms included a reduction in power and increased fuel consumption as soon as the light came on. The light would stay on until the bike was stopped and key turned to off and would only come on again once the bike was held over 4000rpm.

The bike shop noticed that the wires leading to the right side knock sensor were decaying from heat and were repaired. But the code 2-6 persisted. Next the right side sensor was replaced but the code persisted.

Others on this site, had found that a code 2-6 could really indicate a LEFT side sensor, so I asked the mechanics to change that also, but the code still didn't go away. They even swapped over the left and right side sensors. Next they checked the connectivity of all the wires and connections from the ECU to the knock sensors and all seemed fine. Finally they replaced the ECU but still the code persists.

The mechanics are now scratching theirs heads and wonder if the wire repair to the right side knock sensor could be the problem. They say the ECU reads resistance through the wire and that maybe the solder has changed that resistance. An electrician I spoke to thinks this would not make enough difference.

Any ideas or suggestions would be appreciated!!!

Thanks
Mark
 
Shouldn't change the resistance unless it was maybe a cold solder joint.
Is there enough extra wire to cut it out and redo it? I'd try to reflow it 1st but it may have nothing to do with your issue.

By resistance thru the wire do you mean the ecu is seeing a normal knock sensor resistance or the wire has a resistance that is different than the other side as measured at the ecu connector?
 
StMark said:
Others on this site, had found that a code 2-6 could really indicate a LEFT side sensor, so I asked the mechanics to change that also, but the code still didn't go away. They even swapped over the left and right side sensors. Next they checked the connectivity of all the wires and connections from the ECU to the knock sensors and all seemed fine. Finally they replaced the ECU but still the code persists.
Click to expand...

That would leave the ECM and wiring. You could eliminate the wiring by disconnecting the sensors and rigging up a bypass cable that goes all the way back the connector at the ECM.

You can also check whether or not the 25/26 codes are reversed by disconnecting one sensor at a time, repeating your experiment and seeing of you start getting an additional code 25.

They say the ECU reads resistance through the wire and that maybe the solder has changed that resistance. An electrician I spoke to thinks this would not make enough difference.
Click to expand...

I'd agree. There should be only a few solder joints involved, all inside the ECM. Everything outside is crimped.

--Mark
 
Did you or the dealer try manually clearing the code or just wait to see if it goes away by itself? If you didn't manually clear it, the code will persist through some number of start cycles before it will self clear.
 
You could also try disconnecting both ends of the circuit and connecting a test light to operate through that wire as sort of a load test. I've found this method can uncover a suspect bad connection
that may not be that obvious with a dvom continuity test. I'll bet there is a connector in the line that's corroded.

Good luck, and keep us posted,

Geoff
 
gsforster said:
You could also try disconnecting both ends of the circuit and connecting a test light to operate through that wire as sort of a load test.
Click to expand...

That's not going to help much here. The knock sensors are piezoelectric microphones and not resistive sensors, so they're not powered.

--Mark
 
Is there enough wire to reverse the connections so the left sensor connects to the right side and so forth ?
That will narrow down where the fault is .
clear the codes and see if it pops up still as a code 26 or switches to code 25.
 
The wiring diagram shows that the knock sensor wires are shielded. Any chance the shield was shorted to the center conductor of the wire when it was repaired?
 
I had that on my 04 the codes are right but the color code in the service manual is :pie1::pie1::pie1: backwards. I had to replace the knock sensor wireing harness to fix the problem. I guess they cleared the code?
ken
 
Blrfl said:
Originally Posted by gsforster
You could also try disconnecting both ends of the circuit and connecting a test light to operate through that wire as sort of a load test.
Click to expand...

Blrfl said:
That's not going to help much here. The knock sensors are piezoelectric microphones and not resistive sensors, so they're not powered.

--Mark
Click to expand...

Sorry - I was a bit unclear. I know that the sensors are non powered, I didn't mean check for power there but disconnect ecu plug and sensor plug and use the signal wire as part of a circuit for a test light.
Signals from a knock sensor are typically 750mV or less, so a poor connection would be a major problem.

If the sensor plugs could be swapped side to side and the code persists, then it could only be a wiring problem or internal ecu problem, IMHO
 
First thing I'd check is the connector plug itself.

Make sure the code is cleared from the ECM before trying it again.

Mark - can you post a link to my trials and tribulations? It sounds exactly the same as mine, except the new ECM fixed mine. It has to be either the sensor, the wiring, or the ECM. Or ----

Maybe it really is knocking? If not, and the wiring and sensors are good, they put another defective ECM in your bike.
 
unchargedken said:
I had that on my 04 the codes are right but the color code in the service manual is :pie1::pie1::pie1: backwards. I had to replace the knock sensor wireing harness to fix the problem. I guess they cleared the code?
ke
Click to expand...

Many Thanks for all the excellent advice and interest. The mechanics are confident they found the source -- the Sub Wire Harness -- so "unchargedken" was right on the money!! But what an ordeal it has been to get there - what would we do without the collective knowledge of this forum. Many Thanks again. Mark
 
Keep us posted on how it turns out.

BTW - Shouldn't the wiring have been the FIRST thing they tested? It is the easiest AND cheapest thing to test/replace.
 
Hi guys,
I have Fi code 25 on my 2003 ST. That is the left sensor. I unplugged the sensor on the RH just to make sure I am looking at the correct side and ran the bike, code 26 showed up. So it is the left side.
I unplugged both left and right sensors and the ECM and tested for continuity between two ends of the wire both on left and right sides and meter showed 0.00 which means the wires are not broken.
I replaced the left sensor with a brand new one.
Code 25 still showed up.
Now I am wondering if it is the ECM.
If the ECM is bad would it change the code to 26 when I unplugged the right hand side? seems like the ECM is giving out the correct codes for the respective sides.
Anyone has any idea what else I have to check to determine if the ECM is bad or another way to test the wires?
 
vinnyst1300 said:
Hi guys,
I have Fi code 25 on my 2003 ST. That is the left sensor. I unplugged the sensor on the RH just to make sure I am looking at the correct side and ran the bike, code 26 showed up. So it is the left side.
I unplugged both left and right sensors and the ECM and tested for continuity between two ends of the wire both on left and right sides and meter showed 0.00 which means the wires are not broken.
I replaced the left sensor with a brand new one.
Code 25 still showed up.
Now I am wondering if it is the ECM.
If the ECM is bad would it change the code to 26 when I unplugged the right hand side? seems like the ECM is giving out the correct codes for the respective sides.
Anyone has any idea what else I have to check to determine if the ECM is bad or another way to test the wires?
Click to expand...

Did you follow the service manual to reset the stored codes? You may try that.
 
You must log in or register to reply here.
Back
Top Bottom