I've checked wires for continuity& shorts on both sides all good,I've replaced the sensor that fixed it for a few days, now the light back on. Funny thing, tested sensor for resistance, or continuity there is none. Did my new sensor go bad
Code 26
- Thread starter Bern
- Start date
- Joined
- Feb 11, 2006
- Messages
- 9,661
- Location
- Jacksonville
- Bike
- GL1800 R1200RT NC700
- 2024 Miles
- 010688
- STOC #
- 6651
There have been a few threads on Code 26 necessitating replacement of the ECM to be done with it. I don't think I ever saw in the service manual how to check the knock sensors except for continuity of the harness connections to the sensors.
Knock sensors are basically piezo electric microphones. They generate a small voltage when a vibration is detected. You can check them with a digital voltmeter. Pos probe to the pin on the sensor, ground the other probe to engine block. Set meter to lowest AC volt setting if not an autoranging meter. Tap on the engine block or sensor itself and you should see a voltage readout albeit a small one.
Fault code 26 | ST1300 Related
Last night while riding in hot weather the bike gave fault code 26. So I read about it here and went and cleaned the five way t connector under the tank. It was not completely blocked but one port was very restricted. After reassembly and reseting the computer the engine seems a lot smoother and...
I checked my sensor in that manner, and did register a brief small voltage.but when testing for continuity,it's open circuit.
There is no direct path to ground through the sensor. 
Thanks , Paul the answer I've been looking for. Do you know if there's a way to determine if ecu is bad?There is no direct path to ground through the sensor.
Not to my knowledge, other than swapping it with another unit.
Clear your codes, then go riding, and keep it below 4k RPM, and see if your codes come back.
"If your ECU is going bad, it will throw the code if you hold the throttle above 4k RPM for a few moments.
Did you inspect/clean your 5 way tee?
I would do that before you start throwing money around on replacing stuff you may not need.
What year is your bike, and how many miles on it?
Civilian model or PD?
Auction bike?
So your saying ,my 04 civilian 2nd owner bike can through a 26 code below 4k if ecu is good? Mine never lights untill I exceed 4000rpms for about 10 secs.yesterday I removed the airbox and found rodent damaged wires going to the injectors,mostly just the insulation, I checked the wires for continuity all still good,so I taped them up. Also removed the small 5 way tee.and run a wire through it and blowed it out. It still lights up after 4k.
Yes I removed,and rodded tubes and tee. Now this is the one that attaches to map sensor?
NewtonNole
Site Supporter
We have an 03 and 05 and replaced both ECMs. I would avoid used ones on Ebay, I bought two that did the same thing. You can find them for mid $800s new. I tried everything before the purchase including new sub harness.
- Joined
- May 27, 2021
- Messages
- 1,027
- Age
- 63
- Location
- Coquitlam British Columbia Canada
- Bike
- 2009 ST1300
cant remember much about piezio applications, but just wondering if you could simulate one with a resistor or couple of electronics to I the ECM that all is fine, there is so much ____ reformulated gasoline out there that maybe (in the interest of being concise), god is trying to tell you something, these things are like, 10.6 or 10.8 : 1 (tell you exactly come October), my take is the last hard look these CR / fixed ign Curves / o2 content / real fuel value had was ten years ago, I know with the carbed bikes its getting to where you really have to rejet not only the mains.
Its not resistance it makes voltage . and the way it works is the ecm at a predetermined point will keep advancing the timing till it see a knock. (you will not notice it) the it will retard the timing so it has the most advantageous timing. In auto (work the same way) you can get a code if there is too much uncontrolled knock or none at all. I presume that when the code(light) is activated it goes into a limp mode timing so it doesn't cause engine damage.
I believe so.
When traveling at a steady speed & throttle lock in place, I can notice a very small power loss when the light turns on.
It does feel a lot like the timing is retarded to a "default" setting to ensure that there is no engine knock, so that's what I'm guessing is going on.
So far, I've put over 2,000 miles with this error... I'm debating whether it's even worth trying to fix.
I'm looking at $1,000+ to fix a bike worth maybe $3,000... and already has 70,000 miles on it.
I may be better off just pulling the FI light out of the dash and forgetting about it.
I understand your point of view in regard to not getting a replacement ECU. But you could get a good used one for about $400, and I really didn’t like the fuel economy and the way it restricted my riding style, especially in town and traffic. You can also get a new ECU for about $800, which is what I did, for peace of mind. If you look at the power curve on these engine it really wakes up around 5K and up. And most importantly for me, a bad ECU can quit altogether, the get you home mode fails also. The other problems are to much fuel causing ring and cylinder wear and cat damage.
I went through all of this analysis and denial just like everyone with a bad ECU. At first my bike didn’t throw a code above 4K when stationary, then it started coming on when it was. The situation got worse. The way I looked at the cost of replacement in the end was to compare it to two sets of tires, and the long term view of ownership. Thankfully my wife understanding my disappointment was very nice about it really. The real cost was a much anticipated electronic cruise control. May be next year.
Last edited: