The correct part number for your 04 ECM is 38770-MCS-L01... the same part number was used in the 03-07 year models. I went through the same FI26 error and my problem was the ECM.
FI code 26
- Thread starter skipcurt
- Start date
OP
OP
skipcurt
Skip
I did take a pic of my ECM last night so I'm on the hunt for one if it's going to be needed. I should know for sure on Sunday.
Good luck with it... if you are meeting someone with a similar year model with a good ECM, for sure you'll find out if that is the problem. I hated when the error came up as my gas mileage would suffer and later found out that when it occurs the ECM retards the timing. The bike still ran pretty good, but I'm not sure what harm would be done running it for a prolonged time with the timing retarded. I was very lucky to find a used ECM off eBay from a wrecked 06 with 12,000 miles on it for $75. The cheapest new one I could find was from Partzilla for $681.14.
OP
OP
skipcurt
Skip
Blrfl
Natural Rider Enhancement
If it wasn't packed from the factory, just hit it with some plastic-safe contact cleaner. The tail's pretty well shielded from the elements.
--Mark
--Mark
OP
OP
skipcurt
Skip
Well I have good news!!! Today I replaced the right knock sensor connector that is pictured in my 4th post in this thread. Here are my steps:
- Disconnected the battery
- Spliced it together with a heat shrink butt connector and covered with some more heat shrink tubing
- Put some dielectric grease on both of the knock sensor connectors and plugged 'em in
- Connected the battery
- Started her up in neutral with kickstand down until it reached 3 bars. Probably took about 3-4 minutes.
- Straddled the bike and raised the kickstand while in neutral and proceeded to run the RPM's up to 4k and held it there for 45 seconds. No FI light!!
- Then I revved it up and down while still in neutral between 4k & 6k RPM's for 45 seconds. No FI light!!
- I then put it in 1st gear, kickstand up and again revved it up and down between 4 & 6k for another 45 seconds. Still no FI light!!
As for now, I'm going to say this is solved unless anyone can think of any other tests to perform. If not, I will put all the tupperware back on tonight and hopefully I'll give her a spirited ride tomorrow and see what happens.
I want to thank everyone on this thread that gave me advice & guidance. I've learned some new skills albeit I'm not ready to take any big complicated wiring job but I can now perform some basic electrical testing. I have done more internet reading in the last 5 days than I've probably done in the the last 3 months.
I will follow up tomorrow and further down the road to keep the thread factual. Again, thank you all very much for your patience and wisdom.
Skip
- Disconnected the battery
- Spliced it together with a heat shrink butt connector and covered with some more heat shrink tubing
- Put some dielectric grease on both of the knock sensor connectors and plugged 'em in
- Connected the battery
- Started her up in neutral with kickstand down until it reached 3 bars. Probably took about 3-4 minutes.
- Straddled the bike and raised the kickstand while in neutral and proceeded to run the RPM's up to 4k and held it there for 45 seconds. No FI light!!
- Then I revved it up and down while still in neutral between 4k & 6k RPM's for 45 seconds. No FI light!!
- I then put it in 1st gear, kickstand up and again revved it up and down between 4 & 6k for another 45 seconds. Still no FI light!!
As for now, I'm going to say this is solved unless anyone can think of any other tests to perform. If not, I will put all the tupperware back on tonight and hopefully I'll give her a spirited ride tomorrow and see what happens.
I want to thank everyone on this thread that gave me advice & guidance. I've learned some new skills albeit I'm not ready to take any big complicated wiring job but I can now perform some basic electrical testing. I have done more internet reading in the last 5 days than I've probably done in the the last 3 months.
I will follow up tomorrow and further down the road to keep the thread factual. Again, thank you all very much for your patience and wisdom.
Skip
Last edited:
DavidR8
Site Supporter
- Joined
- May 31, 2017
- Messages
- 1,220
- Location
- Vancouver Island, British Columbia
- Bike
- 2013 BMW R1200RT
- STOC #
- 8968
Hurray!
That's great news. Let's hope a road test will show the same results.
Sent from my iPhone using Tapatalk
That's great news. Let's hope a road test will show the same results.
Sent from my iPhone using Tapatalk
John Anthony
Moderator
Great job, Skip! Way to hang in there. Kudos to all the members who lent a hand while you were sorting it all out.
John
John
Always a good feeling when you fix lectrical issues. Congrats
Blrfl
Natural Rider Enhancement
I think you're good. If the ECM's happy, you should be, too. Learning something new is the best part about these outings.
--Mark
OP
OP
skipcurt
Skip
Bad news!
Took the bike out for a test ride this morning and it was looking promising. Riding around at some times above 4k RPM as much as 5500 RPM and the FI light didn’t come on. Stopped and filled up with 93 octane and left to hit the interstate. Not even a mile after getting fuel but at 9.4 miles in while getting on it to merge in the light came on. Major bummer! What a disappointment. Same code 26.
So my next thought was maybe, just maybe the disconnection of the battery didn’t reset the ECM. So I’m going to try the published method from the ST Wiki pages. My question is this. It says short the brown and green/orange wires on the diagnostic connector. Which one is that? The grey or the black one or does it matter which one? Both of those color wires are in both connectors. Figured I’d try that and then re-test ride before my swap/check/test with another ECM on Sunday.
Took the bike out for a test ride this morning and it was looking promising. Riding around at some times above 4k RPM as much as 5500 RPM and the FI light didn’t come on. Stopped and filled up with 93 octane and left to hit the interstate. Not even a mile after getting fuel but at 9.4 miles in while getting on it to merge in the light came on. Major bummer! What a disappointment. Same code 26.
So my next thought was maybe, just maybe the disconnection of the battery didn’t reset the ECM. So I’m going to try the published method from the ST Wiki pages. My question is this. It says short the brown and green/orange wires on the diagnostic connector. Which one is that? The grey or the black one or does it matter which one? Both of those color wires are in both connectors. Figured I’d try that and then re-test ride before my swap/check/test with another ECM on Sunday.
Last edited:
wjbertrand
Ventura Highway
Not sure what you mean by "both connectors" as there's only one diagnostic connector. It's red and located to the right side under the passenger seat.
Shrink tubing helps relieve the stress at this point.
OP
OP
skipcurt
Skip
Gotcha! Thank you for pointing me in the right direction. I saw what you referred to and confirmed in the service manual.
So I did that now too and took it out for a spin and unfortunately the dang light came back on again.
wjbertrand
Ventura Highway
I have heard that sometimes this knock sensor warning cannot be cleared from the ECM and a new ECM has to be fitted to solve it.
The ECM does not store a knock sensor error... when the key is turned off, it resets.
Gandolkf
Site Supporter
I have same problem.F1 code 26 blinks.
have not had time to deal with it.
not sure I want to now.does not run bad or kill mileage.
Does not come on at any specific time, sometimes after a few minutes----or a few hours.
had a few days no F1 light.
Maybe I'll wait to see what happens.
have not had time to deal with it.
not sure I want to now.does not run bad or kill mileage.
Does not come on at any specific time, sometimes after a few minutes----or a few hours.
had a few days no F1 light.
Maybe I'll wait to see what happens.
wjbertrand
Ventura Highway
When I unplugged my knock sensors as a test for my misfiring problem the computer set an error code for both the left and right ones as soon as the engine reached around 4000 RPM. When I plugged the sensors back in the light went out, but the codes were still in the ECM before I cleared it.
We need a Honda engineer to answer this... the ECM on the 03-07 bikes obviously have some defect with the reported number of failures and Mother Honda seems unconcerned about it...
Blrfl
Natural Rider Enhancement
Those bikes are long out of warranty (as is every one put into service before 2014), there's a known way to repair it, the failure isn't a safety hazard or even what I'd call widespread and the part was superseded on later bikes. What should they be concerned about?
My long-held suspicion is that a component in the bit of analog circuitry that interfaces the sensors with the computer is failing. It could be a design defect or that Keihin got a bad batch of parts that ended up in the assembly process when the boards for this ECM were being made. The failure rate may not even be high enough to count as significant, but only Keihin and Honda know that and I doubt either would tell us.
What Jeff is describing is the expected behavior if the ignition was shut off before he reconnected the sensors.
--Mark