Code 25-26 knock sensor - ECM repair

[TechKnowViking]

Okay… lots to pick apart here.

First off, Hi I’m the guy referenced in the original post who made that video. My channel is now under the name TechKnowViking.

Second, a lot of things have been said about this knock module issue, most of which are not correct.

So let me clear some things up.

The actual issue, leading to an unnecessary code 25 or 26, was found by a few very talented electronics technicians. What they found was that the KC2-B0 modules, at least the ones manufactured at the time of their use in the Keihin ECMs, were faulty, having some contamination trapped between an Op-Amp and the PCB surface, causing a 12V trace to bleed over to a 5V reference trace, thereby (eventually) destroying the IC on the board, resulting in the discreet output from the module sending a “Knock Detected” signal to the ECM Processor.

Also, replacement ECM’s for most Honda (and some other manufacturers) products come with those original modules on the board. Some fail quickly, others don’t fail at all.

The video I posted was for a 2003 GL1800 that I repaired myself for my own use. At the time, I was not offering to repair them, now I am. The cost for ECM’s which contain two modules is currently $545 and for ECM’s containing one, $495. This is for US based customers. All the Aquatrax skis have one module, the ST1300’s, GL1800’s, and other two bank engine motor sports vehicles, have two. This is the reason for the cost difference.

My repair is a KC2-B0 module replacement, not a circuit disable like is offered by Gen-tune. I am a former Aviation Electronic Warfare Technician of 20 years, and I am IPC-600 certified and perform my work to level 3 IPC standards. This is why my work includes an 2 year limited warranty.

I also want to point out that the part number for the module has always been in the description of those videos. If you want to attempt the repair yourself, feel free to watch my videos for reference. You’re even welcome to send me a message on my Facebook business page if you want to ask questions.

Here’s the page:

https://www.facebook.com/share/1BKfqhGgSK/?mibextid=wwXIfr

Here’s my channel:

https://youtube.com/@therealtechknowviking?si=twqXHWTfN3IBilCe

If you’ve read this far, I have a full, start to finish, video of a GL1800 ECM repair publishing on the 18th next month. If you’ve read want to fix yours for yourself, I suggest waiting for that video before trying.

I’ve added a few photos from some of my most recent customer’s repairs.

 

[Larry Fine]

 

[ST Gui]

Everything you thought you knew is wrong.

So let me clear some things up.

Ok, I know the first bit wasn't the sentiment you expressed but I reached a bit because it made me laugh. Appreciate knowing your bona fides.

I'm not in need of an ECM at the moment and don't expect I will be anytime soon. But then nobody expects the Spanish Inquisition.

Is there a core charge in that someone's bad ECM need be turned in to get a repaired one?

Can an owner send in their 'dead' ECM to see if it can be repaired?

Is their data in the ECM unique to that owner/module?

If so can it be preserved or is it (as I suspect) lost?

I don't have an FB account so I'm asking here.

 

[Larry Fine]

Yes, is any programming required, or do all F.I. bikes work with the same parameters?

Also, the supplier page shows two modules. Are there any differences between them?

 

[Chris09]

When can we expect to meet patient zero?
Will some reasonably established member of this forum please step forward and say emphatically and with conviction that...
I had the sickness [well, the bike at least,] and now I am cured !
Hope springs eternal [cha ching $$$] and there's a sucker born every minute
Isn't going to happen, ever
call it a hunch
Where's Marty Feldman when you need him

 

[TechKnowViking]

Ok, I know the first bit wasn't the sentiment you expressed but I reached a bit because it made me laugh. Appreciate knowing your bona fides.

I'm not in need of an ECM at the moment and don't expect I will be anytime soon. But then nobody expects the Spanish Inquisition.

Is there a core charge in that someone's bad ECM need be turned in to get a repaired one?

Can an owner send in their 'dead' ECM to see if it can be repaired?

Is their data in the ECM unique to that owner/module?

If so can it be preserved or is it (as I suspect) lost?

I don't have an FB account so I'm asking here.

I don’t charge a core, as there are so many versions and variations of Keihin ECM’s that I cannot keep a running stock for most models. So, I simply repair what my customers send me and return their unit, unless they specifically request some sort of other part number, and I can only do that if I have one available.

Owners sending me their “dead” ECM’s is basically how I’m currently operating. My turn around time is usually less than a week, but that could quickly change with how many people have reached out now that they know about my service. I haven’t yet but I will offer “front of the line” service if someone really needs it done ASAP.

Data on these ECMs is only unique to the owner’s bike in one real way. If the bike has HISS (Honda Immobilization Security System) then it is coded to their bike and would need to be returned specifically to them for use with their RFID equipped key. European models typically come with this system.

Other than that, most of these things predate “mating” software that locks them down, so we are free to replace them at our leisure.

 

[TechKnowViking]

Yes, is any programming required, or do all F.I. bikes work with the same parameters?

Also, the supplier page shows two modules. Are there any differences between them?

No programming is required unless there is an HISS activated on the ECM for a coded ignition key. US based bikes do not have this feature to my knowledge, but the part number on the ECM will tell you, as well as the country code.

The modules in question are KC2-B0, and the component sense website is simply archaic with the double listing.

 

[Chris09]

It's alway a gas with these claims of ECU restorations; never anything verifiable or quantifiable, you can never point to anyone at arm's length or even further and say yeah, he got his fixed. Never anything on a dyno, nothing tangible, nothing empirical, in essence, a lot of well, you can't even say with any confidence; and in this day of the internet, " I know a guy who know's a guy who got his fixed... " or " know a guy who bought an aftermarket and it improved the performance of the bike on the basis of any criteria;
A lot of, send me your [essentially] core that you know doesn't work along with money and I'll...

... step right up...

I'm Canadian but if I was an American, I'd be from Arkansas, which is not the tell me your trade secret state; it's the if you got it, I'll have what you're having and won't ask how you got it type of reasoning,
If you got it, you worked for it and it's worth something for me, no questions asked
Right now Honda wants $2,000 Canadian for a new 2003 Blackbird ECU and going forward I don't know if that's going to be better or worse than the one I already have, which makes the bike perform perfectly;
the one I still have that is
It'll be a cold day in Fort MacMurray or Grande Prarie before I ever send that ECU out
The CBR1100 ECU timing and circuitry is little changed by failure of the knock sensor input, the timing routine; reported dyno tests show these bikes deliver about 137 to the rear wheels - either way.
My opinion remains that the knock circuitry failure is a layer failure on ignition timing that doesn't make any difference on acceleration or load
plus I like the little red light

rebuttal?....
[edit] well, it's a tough crowd...

 

[Larry Fine]

It's alway a gas with these claims of ECU restorations; never anything verifiable or quantifiable, you can never point to anyone at arm's length or even further and say yeah, he got his fixed. Never anything on a dyno, nothing tangible, nothing empirical, in essence, a lot of well, you can't even say with any confidence; and in this day of the internet, " I know a guy who know's a guy who got his fixed... " or " know a guy who bought an aftermarket and it improved the performance of the bike on the basis of any criteria;
A lot of, send me your [essentially] core that you know doesn't work along with money and I'll...

What about eliminating the persistent 25/26 codes and running better?

You could request that the case pieces be let off to see the new modules.

 

[Larry Fine]

Right now Honda wants $2,000 Canadian for a new 2003 Blackbird ECU and going forward I don't know if that's going to be better or worse than the one I already have, which makes the bike perform perfectly;
the one I still have that is
It'll be a cold day in Fort MacMurray or Grande Prarie before I ever send that ECU out

Why would you, if you're happy?

 

[Chris09]

... makes me feel like a rebel...

rebuttal?....
[edit] well, it's a tough crowd...

 

[TechKnowViking]

I think you should quit drinking, Chris09.

 

[W0QNX]

TechKnowViking I did read your entire post and I did watch a few of your video's. Please keep us informed and yes you will have posters here who do not add positively to the thread. Please try to ignore those and stay positive for those of us that look forward to your effort. There will be naysayers until you "prove" this fix works.

We had a major forum crash last winter and lost most of the old posts discussing the Code 25 code 26 error on older ST1300's (2007? and back?). There were MANY posts on the Code 26 issue and there are hundreds of ST1300's running out there with the Code 26 error.

I will say your repair is the first I've seen that opens the ECU and replaces a module by removing and replacing soldered in parts. Based on that It looks to me like a fix does indeed exist for our bike's problem and you are trying to provide that fix. Everything I saw in your GW ecu KC2-BO sub board replacement video looks 100% valid.

I hope this thread can stay useful and on point. I look forward to seeing more useful posts from you.

 

[MattSTlouis]

I have 3 bad ECUs and am tempted to order the parts and try it on the worst one. Yes, all 3 behave differently as to how long it takes for the light comes on and how it performs with the error present. 2 of them come on more frequently and I can tell when the engine looses a bit of power but 1 I can not tell and takes longer for the code to trip. I wonder what would happen if you just unsolder the 2 parts and just leave it like that, I might have to make a test ECU, if no light then its fixed, just use the correct octane, like a normal person.

 

[Andrew Shadow]

I agree completely with W0QNX. There is no voodoo being proposed by this repair. This is a failure of an electronic component like any other electronic component failure. The problem has been that because Honda has never disclosed any information about their ECM no one was able to run a diagnostic to compare the results with known correct values to be able to locate the source of the failure. It took someone willing to take drastic steps and cut open an ECM and find the failure the hard way. Once the failure is identified the repair is not magical. It is an electronic component replacement procedure no different than the replacement procedure of any other electronic component.

The only thing that I am unclear about is if this repair also eliminates the contamination trapped between an Op-Amp and the PCB surface that is stated as being the root cause of the failure. Does the process of replacing these KC2-B0 modules also eliminate this problem? If not, does this mean that these new modules will eventually fail again for the same reason?

I applaud TechKnowViking. There are many ST1300 owners suffering from this failure who can not find a working replacement ECM. There was no possibility of repair available to them either, and therefore no hope. Now there is hope.