I understand the point of the thread, more general "musings around the camp fire" than "does Honda have a problem that I should think about suing them for?" Someone pointed me at this thread, and, by-and-large, I agree with most of the replies... but it's probably worth a few "back of the envelope" calculations just to join in while we drink beer around the campfire.
That said, it's a warm Sunday here, and I've just finished building decking in the back garden, now laid in a hammock having the lovely MsHom hand me a beer once in a while... so, excuse the ramblings that will no-doubt be below...
ECU's / ECM's are mostly interesting to me, because they're the perfect blend of hardware, mixed with near-realtime software. Considering that the engine (in all vehicles) has certain manufacturing tolerances built in (pistons to a certain number of thousandths etc), the software has to be able to learn how it should operate
that particular set of nuts and bolts. Any of you that have worked in a dealership probably knows that the first time you fire an engine after building a bike from a crate, it don't sound right at all, but then the ECU learns & adjusts, and all of this happens pretty quickly. If you think a brand new build sounds perfect on first fire, please go get your ears checked (or mine are just more mechanically sympathetic I guess).
What's my point with the waffle? Well, manufacturing tolerances, but also, software bugs. It strikes me that the most commonly reported issue (here on the forum) is code 25/26, and once triggered, no matter what you do, the ECU seems to remember it. That sounds like a software bug to me, which
should be possible to find a workaround for.
What I found was a bath of epoxy-like resin....Good luck repairing that !
You'd be surprised the number of electronics that are covered in epoxy. Mostly it's weather proofing like the case you mention in the 1100, sometimes it's tamper protection. Nintendo include a JTAG on some of their circuits, but like to cover them with epoxy. A guy I know asked me some questions about how to get at the JTAG so he could let his kids play hacked games... after the conversation of "don't developers deserve to get paid"... I told him how to safely heat it up & work with it. He apparently got bored, and used a paint stripping heat gun. Next day, he comes to me with "I followed your advice, and all the components dropped out of the board!"... "Err, that wasn't my advice, ain't my problem you're impatient!"
Epoxy resin isn't a show stopper, it just makes the job 1000x slower.
Is the percentage of ECU failures still relatively low? Like 0.1%? Or more like 5%?
So, back of the envelope calculation time...My STOC # is 9004, not all members of this forum will have a STOC #, and not every owner will be a member of this forum, or have a STOC # from some other source. So lets say there are 10,000 of us that care enough about these bikes to register a STOC. Those that care enough, probably on average have more than one ST (some like Bob buck the trend and I'm surprised his wife puts up with it), some will only have 1 bike that they look after better than their kids. So lets say on average each STOC registered person, has 2 bikes.
I make that a ballpark 20,000 bikes. I've been on this forum around 2 years, and I can recall on one hand the number of threads that resulted in an ECU swap fixing the problem. So lets say that there's 5 ECU swaps every 2 years, my ST1300A2 is now 19 years old, so following the average, lets be charitable and say that's 50 ECU swaps that STOC members are aware of, in 20 years. Note here that my number of bikes is lower than actual sales, and that my number of "known ECU swaps" is probably higher than actual, which means the final number is a gross overestimate of frequency.
I make that 50 swaps from 20k bikes, or 0.25%, or 1 in 400 bikes will at some point in 20 years, need it's ECU swapping. Again, that's a gross over-estimate. I'll suggest half the STOC numbers are 1100 riders, half are 1300, so for either bike, it's closer to 0.125% chance in 20 years. Of course, I bet STOC numbers are heavily weighted towards the 1100, but this is an envelope calculation.
"But wait, ad.hom, 1 in 400 bikes sounds like I should sue Mother Honda!" I hear someone at the back cry... well... some perspective is in order...
Let's say you're 65 years old, a smoker, and live in the US. You have a 1 in 400 chance of dying in the next 10 years from any of: heart disease, stroke, cancer, pneumonia, aids, COPD or any accident like crossing the street. If you
never smoked, your chances are 1 in 800. If you're not interested in suing a big tobacco corporation, I probably wouldn't recommend you try suing Honda either. Again, I know it wasn't the point of the thread, but it's worth exploring these things anyway.
1 in 100 die of obesity. 1 in 500 of a food borne illness. 1 in 33 develop skin cancer. 1 in 200 die in a homicide (US data, 2009).
Again, my number was grossly over-estimated.
So, theory and speculation out of the way...
I came across this site claiming to be able to repair the ecu
Be very careful of those sites. Assuming they know what they're doing with any ECU, there's a strong chance they'll brick an ECU if they're not familiar with the ST ECU. Most sites like this tend to have done it once before, and they got lucky. With the number of people wanting different fuel maps for supposed throttle issues on STs, companies that will do fuel maps are still pretty rare - and that has nothing to do with the age of the bikes. I honestly wouldn't waste my money unless they were willing to put you in direct contact, on the phone, with someone that has used their services and still rides the ST they modified.
Fun fact too, I was trying to figure out how to tweak the ECU on my Jaguar many moons ago, so I put an industry standard tool into diagnostic mode, seeing if it could just do a raw dump of the CAN traffic for me to analyse later. This was a 2005 car. About 10minutes into my journey, the LCD on my dash lit up "TAMPER DETECTED. SHUTDOWN IN PROGRESS." - and the whole electrical system went dead. At 70mph, I lost power steering, ABS, ignition, radio, everything. If my 2005 car had that, what's the chances of Honda putting tamper detection into ECUs?
Can the ECU be tampered with? Probably. In my line of work I like to think that nothing is impossible. But, everything comes with a cost in some shape. I'm actually interested some day to compare a police ST ECU with a civilian one, mostly for the sake of curiosity - but assuming there's some tamper detection baked in, I'd probably burn through 6 of each type before I knew the differences. I'm not sure as a group, we have 12 spare ECUs for me to play with, just for a curiosity.
Then there comes the supply & demand aspect to it... if we're talking about learning how to fix FI fault codes, and lets imagine I did figure that out. I'm looking at less than 50 sales over 20 years. That makes my unit cost so high that it's just cheaper to pay the price for a new/used functional ECU. It'd also have to take it's own priority in my "pet projects" queue, unless someone is going to pay me enough to figure this out that I can quit the day job.
Even if ECU's turn into unobtainium like the wind deflector kits, it's sad to say, but you're insane if you want to pay me to dig into ECUs rather than just buy another bike.
