G7 Computers

It really ought to be better having essentially smart components which communicate information instead of switching load circuits through every control which should mean less wiring. Until things fail and the only replacement parts that communicate on the bus are those available through Honda.
 
Last edited:
Andrew Shadow said:
If you do go ahead make sure to report back here regarding the results. If they really can fix this problem there are many people who would like to know about it and I suspect that they will get quite a few requests to do so.

Before I would send it to them I would want to know for certain whether they are really fixing the problem or hiding it. When this problem is caused by a defective ECM there is no fault with any part of the knock sensor system. The problem is that the ECM thinks that there is a fault where there is none. I reiterate this because every other ECM repair shop that I have read about here and elsewhere to date have not been able to fix this particular problem with an ST1300 ECM. The best that any of them have been able to offer is to prevent the FI light from illuminating when the problem occurs so that you don't see the FI light. Fixing it means that if there is a true and legitimate fault detected in the knock sensor system in the future the ECM will recognize it and will illuminate the FI light and will display the correct fault code to match the defect as it was designed to do.
Click to expand...
I have asked them directly about my 25/26 issue they said they would fix it and not mask it. I guess the rest is just a gamble . If it works Im buying a bunch of pre 08 ST's and having a Harem.
 
I hate to sound like a broken record but I would want to know exactly what fixing it means to them if I were considering sending them my ECM.

Fixing it to me means that it will function exactly the same as Honda designed it to function without the phantom 25/26 codes.
It means that if there really is a legitimate failure in the knock sensor system the ECM will recognize it, will set the correct failure code for it, will store that correct code in memory for later retrieval, and will illuminate the FI light but only when it is supposed to under legitimate failure conditions?
Perhaps even more important is that the knock sensor system will still function and will still provide the necessary inputs to the ECM to allow it to help prevent detonation and any potential engine damage that can result from it?

The reason for questioning their claim is because no other ECM repair facility that anyone on this forum has approached regarding repairing the ST1300 ECM code 25/26 problem that I have read about has been able to successfully fix it. When pushed hard enough it always turned out that the best that they can do is to disable something within the ECM that prevents it from setting a code 25/26 and prevents it from illuminating the FI light due to a code 25/26. The problem with that is that when there is a legitimate knock sensor failure the system will no longer alert to it.

Apparently none of these ECM repair facilities have access to the specifications of the ST1300 ECM because Honda has never made this data available to them. The same is true of several other Honda products that have this same code 25/26 problem including some of their water-craft. No repair facility that I have read about has ever been able to propeerly repair any of those Honda ECM's because of the lack of technical information provided by Honda. If other ECM repair facilities are able to repair this code 25/26 problem with Honda ECM's there would no cause for concern, but this doesn't seem to be the case. If no other ECM repair facility has been able to crack this nut one has to wonder what they do. Hopefully it means that they have cracked this nut.

In the end, as you wrote, it is a gamble. Sending them a failed ST1300 ECM and testing the results afterwards might be the only way to find out. I hope that they really can repair it. There are at least a few ST1300 owners who would like to have this repair option and who knows, I made need it one day as well.
 
Kurt0721 said:
I have asked them directly about my 25/26 issue they said they would fix it and not mask it. I guess the rest is just a gamble . If it works Im buying a bunch of pre 08 ST's and having a Harem.
Click to expand...

Fixing it would involve replacing the failed Knock Sensor Signal Processing chip on the ECU board but this component is no longer available.....unless they have a stash of them of course, you'd have to ask them.

How can they tell you what they can do to your ECU before they look at it and diagnose the issue?

Have they ever repaired a ST1300 ECU?

They wouldn't be the first shop claiming to be able to fix it....until they actually start looking at one.

And then, if they return it to you with no Code 25, you'll have in fact no way of knowing if they actually revived the knock sensing loop or not.
 
Andrew Shadow said:
I hate to sound like a broken record but I would want to know exactly what fixing it means to them if I were considering sending them my ECM.

Fixing it to me means that it will function exactly the same as Honda designed it to function without the phantom 25/26 codes.
It means that if there really is a legitimate failure in the knock sensor system the ECM will recognize it, will set the correct failure code for it, will store that correct code in memory for later retrieval, and will illuminate the FI light but only when it is supposed to under legitimate failure conditions?
Perhaps even more important is that the knock sensor system will still function and will still provide the necessary inputs to the ECM to allow it to help prevent detonation and any potential engine damage that can result from it?

The reason for questioning their claim is because no other ECM repair facility that anyone on this forum has approached regarding repairing the ST1300 ECM code 25/26 problem that I have read about has been able to successfully fix it. When pushed hard enough it always turned out that the best that they can do is to disable something within the ECM that prevents it from setting a code 25/26 and prevents it from illuminating the FI light due to a code 25/26. The problem with that is that when there is a legitimate knock sensor failure the system will no longer alert to it.

Apparently none of these ECM repair facilities have access to the specifications of the ST1300 ECM because Honda has never made this data available to them. The same is true of several other Honda products that have this same code 25/26 problem including some of their water-craft. No repair facility that I have read about has ever been able to propeerly repair any of those Honda ECM's because of the lack of technical information provided by Honda. If other ECM repair facilities are able to repair this code 25/26 problem with Honda ECM's there would no cause for concern, but this doesn't seem to be the case. If no other ECM repair facility has been able to crack this nut one has to wonder what they do. Hopefully it means that they have cracked this nut.

In the end, as you wrote, it is a gamble. Sending them a failed ST1300 ECM and testing the results afterwards might be the only way to find out. I hope that they really can repair it. There are at least a few ST1300 owners who would like to have this repair option and who knows, I made need it one day as w
Click to expand...

MidLife said:
Fixing it would involve replacing the failed Knock Sensor Signal Processing chip on the ECU board but this component is no longer available.....unless they have a stash of them of course, you'd have to ask them.

How can they tell you what they can do to your ECU before they look at it and diagnose the issue?

Have they ever repaired a ST1300 ECU?

They wouldn't be the first shop claiming to be able to fix it....until they actually start looking at one.

And then, if they return it to you with no Code 25, you'll have in fact no way of knowing if they actually revived the knock sensing loop or not.
Click to expand...
I noticed you have a 03 , do you have the dreaded code ?
 
@Kurt0721 please let us know the results of your venture! G7 could possibly have large customer base if reasonably priced.
 
I liken the situation to my fork in the road with my JK, had I let someone else rebuild the transmission or replace the heater core, what have I really got when I get it back, and where am I now, should I let my kids drive it, last thing I want to hang around with is an unsafe or unreliable vehicle.
It's easy and convenient to send an ECU off for a repair at the quarter of the price of replacement and accept an assurance that it's repaired / restored if all I intend to do with it afterward is sell it.
 
Chris09 said:
It's easy and convenient to send an ECU off for a repair ..........
Click to expand...

......except..... there is nowhere to send the ST1300 ECU to!

No replacements for failed board components are available anymore and none of the specialists on several continents who have tried to flush the code over the last 20 years have managed to do it.
 
I don't see these as remotely the same. Transmission and heater care repairs are ones that are successfully accomplished on a regular basis by many repair shops in many locations. Choosing not to trust these to someone else is a personal choice that does not make them unknown failures requiring highly specialized repair capabilities. This ECM repair has never been successfully accomplished. The root cause of this failure is unknown. No one seems to know how to fix it and there is no DIY option presently. Considering having someone else do this repair is not a matter of preference but of necessity, and because of these unknowns it is a leap of faith.
 
MidLife said:
Dark Sun and others have clearly shown that a fairly common failure mode is the breakdown of the Knock Sensing Processing Chip on the ECM board, due to overheating.
Click to expand...
Assuming that is the case I would regard the overheated chip as the end result not neccessarily the cause. The reason behind the overheat condition would be the root cause. This could be a simple matter of a chip that is not robust enough for the task. Unfortunately it could also be the result of something that causes the knock sensor circuit to overwork and to overheat. In this case a replacement chip would likely suffer the same fate if the source of the overheat is not dealt with. Unfortunately we probably will never know until a replacement chip becomes available to test or Honda releases the ECM data to the repair facilities so it can be analyzed and a repair can be developed.

ECM repairs are routine with modern equipment. It is frustrating that what has become a standardized repair with other ECM's is not available on this one.
 
Andrew Shadow said:
.....something that causes the knock sensor circuit to overwork......
Click to expand...

As shown before on this site, the Knock Sensor Processing Chip is constantly hard at work. Lots of noise constantly coming from the engine and the chip is constantly filtering the noise to check for knocks. It is also constantly talking to the CPU, indicating either "All OK", or "Here is a Knock, take action". If the chip fails and the CPU doesn't hear from it anymore, it will trigger the code and default mode.


Andrew Shadow said:
......... a replacement chip becomes available .......
Click to expand...

Once they are gone, they are gone. Post 07 have a "replacement" chip, different design, different footprint requiring different traces on the newer boards.

Andrew Shadow said:
...... a repair can be developed.
Click to expand...

Repair would be replacing the failed chip....that is no longer available.

Andrew Shadow said:
ECM repairs are routine.......
Click to expand...

Nothing routine about replacing a component on a potted assembly. Costly as well.
 
MidLife said:
Nothing routine about replacing a component on a potted assembly. Costly as well.
Click to expand...
I have to wonder about that based on what is commonly done in the automotive trade.
The ST1300 ECM is particular in that it is such a small market that no ECM repair shop would probably be interested in investing in whatever is required to diagnose, repair, and test them. They probably would never be able to make their money back based on volume so what they would have to charge for each individual repair would be to high to attract customers.

ECM's repairs are done all the time very cheaply however. There are countless ECM repair facilities, and repairing ECM's has been routine for decades. It is not a new endevour. I used to send automotive ECM's out for repair/exchange way back in the eighties. Not just vehicle engine control ECM's get repaired now either. There are ECM's for all kinds of things nowadays to run all kinds of electric gizmos. They all have failures eventually and many of them do get repaired. Many of the places that service the automotive trade have fixed flat-rate repair costs regardless of what the defect is. Many also use an exchange program flat-rate system wherein the failure of the one being sent in is irreleavent, it is the same fixed flat-rate exchange fee no matter what. In the automotive trade this is often offered at well below the $500.00 mark. I can only assume that of the countless number of ECM's that are being repaired there has to be a significant percentage of these repairs that involves removing potting compound to do the repair. I can't see that it is anything but mostly routine for them without very high risk or they would not be using these flat-rate business models. I wouldn't be surprised if the most costly part of the whole process for them is the investment in the testing equipment needed, not the actual repair process itself. But then again they have a big advantage in that they have access to all of the ECM data to be able to diagnose and repair them, something Honda doesn't seem to want to share for the ST1300.

As for removing the potting compound, I have no idea how difficult it is or isn't as I have never done it. Based on what I have observed my suspicion is that the difficulty might vary greatly depending on what type of potting compound was used. Where I used to work I have seen potting compound being removed in the electronics shop and it was not a big a deal at all for them to do it. They were specifically equipped to do it however and again, I'm sure that it depends on the type of potting compound. They used some chemical agent that they would allow to soak on the potting compound for a while and it would mostly turn to a thick sticky liquid that they would then remove using stiff soldering brushes and some cleaning agent. It was toxic I guess as they would wear a respirator and use a vapour booth while doing it but it didn't seem to be that big of a deal to them. Not a very scientific or particularly helpful description I know. It was not the department that I worked in so I never really paid much attention to what they were using or how they did it, which is to bad now looking back.
 
MidLife said:
Then you also saw he found the issue was due to the failure (overheating) of a signal processing chip on the ECU board. This component is no longer available; hence no real repair is possible.

All mods will have to be done via code alterations. Some have identified which code lines could be changed but so far nobody has been able to reprogram the unit (Honda seem to be using an uncommon K-line communication protocol and timing for this ECU).

Given that the ECU is not actually repairable, you may consider if you really want a code modification.

First mod would be for disabling the light, which, as you say, can be done with tape, or just ignored.

Second would be to code out the limp mode....that is not actually a limp mode as many will tell you (including me) the bike is perfectly ridable with the 25/26 code on.

This "limp" mode is actually a slight ignition retarding that provides a bit of safety margin against pinging, given that the ECU is no longer able to detect pinging and take action in case it occurs. At the cost of a slight mpg drop.

Code this out and the bike mapping will be closer to optimum but you'll lose the bit of protection the "limping" mode offers against pinging as the ECU can do no retarding anymore in case pinging occurs.

You may be better off riding it the way it is.

There is a growing pool of owners riding with the code on and no report of V4 pinging themselves to death.

If G7 can actually get you out of limping, you'd be the first to ride around with no pinging margin at all and we will follow you with interest.
Click to expand...
I limped 4000 miles on a trip this summer and kept up with everyone. It probably knocks off 5 MPG . It's more of an annoyance than anything.
 
You must log in or register to reply here.
Back
Top Bottom