If you unscrew it on a cold engine and manage to keep it out of adjustment and it is still in the 0 position then in the software you will be able to see if the throttle position changes as the temperature rises, it shouldn't because the sensor isn't screwed on I just don't know if you can keep it out of adjustment ideally you would have the hod ecu hacker on while unscrewing to see if it is 0.
When you fire up the bike and it starts to warm up, if the TPS starts to change values by some miracle then you know it's something with the electrics, and if the TPS doesn't show then screw it on and see if it starts to show 1.5 like it does now that would mean something is going on with the throttle as it warms up.

check loose throttle cable.

Sigh. it's not a loose throttle cable.

Before you do that you should try to adjust the throttle on a warm engine with this screw then wait for it to cool down and see if it is ok.

I've already replaced the TPS with a new one, and it is still doing this.

I don't understand your logic on how you get to "problem solved".

Charles.

The hod ecu hacker program shows that you change the position of the throttle as the temperature increases, warm up the motorcycle and loosen the screw slightly, mark it as originally set in case of emergency.

Loosen what screw?

Charles.

this screw adjust throotle position do this with hud ecu hacker on to see position

In post #88 the idle screw that Lukas points out there is a spring behind it and that's why I was wondering if your throttle snaps back. That could be the cause of a fluctuating idle?

One other thing here, if you like to jump on the throttle it can bend the tang on the return spring I mentioned. Meaning that bend could cause a loose return to idle? Thinking out loud.

No, there is nothing mechanically wrong with the bike. The throttle snaps back fine. When the problem is happening, attempting to close the throttle more has no effect. Nothing is loose or bent or hanging up.

Charles.

If you will when you have this problem push with your finger on that arm in post #88 back to the idle stop. If that doesn't do it then it is your ecu

That's why I wanted him to unscrew the TPS and warm up the motorcycle, only not to unbolt it, if the motorcycle after warming up starts to show in the hod ecu hacker that the throttle is open as it shows on both log file about 1.5% and it is unscrewed from it, then it is clear that there is some short either on the wiring harness, or in the ecu some resistor, or some other damn thing I mean to check the throttle.

When you unscrewed the bolt, in the software what was the throttle position value 0?

Warm up the motorcycle so that the problem occurs and unscrev the TPS, set it manually to zero so that the software shows it as 0% and see if the motorcycle continues to run bad with unscreved tps with position 0% OR if you have a second TPS just replug the plug to second TPS you don't have to unscrew the one in the throttle.

So I could go through all of that, but I called CSC and they priority-mailed me a loaner ECU. It's an early ECU that mistakenly got shipped with US bikes, as it throws a code for missing side stand switch (which the US models don't have).

At any rate, I removed my old ECU, and... the area under the ECU was covered in moly grease. The bolts were also covered in it. Thick, heavy, dark grease. I cleaned it all up with brake clean and then electronics cleaner. I'm not sure if the ECU body is supposed to be grounded well in order for it to operate, but I'll bet that grease wasn't helping.

After everything was nice and clean I plugged the loaner ECU in, started the bike, and took it on a decent ride. So far, no problems whatsoever. No high idle, even after sustained highway riding. Idle is 100% stable at all times. I also didn't throw a single O2 sensor code for the entire ride. And this was easily 5x as long as the other rides that would start the bike running badly.

It's hard to say "it's fixed" from one ride... but it appears to be. I'm going to rip around for a while with this loaner and then get a new one from CSC if it continues to go well.

I wonder if a component in the ECU fried, or if the software glitched, or what. Very odd that the ECU would otherwise work normally...

Charles.

Some resistor has burned and as it gets temperature, it expands and gives a signal it can be located you need to go from TPS along the cable to the ecu plug and check to which pins it is connected then check where the pins lead on the board and you will find the damaged component, which should be examined with a meter it can be repaired if I were you, I would check the batery charging if there are no problems with it, because the ecu rather not damage itself if you do as I said, that is warm up the motorcycle and unscrew TPS then you would find that it is not the throttle but the ecu, but never mind check the charging.

Glad to hear that you finally appear to have it running right again. See, nothing is unfixable.

If it is a bad ECU, I would follow Lukas's suggestion and check your stator and rectifier output, since ECU's are designed to be pretty robust electrically. It is not normal for them to fail. From the wiring schematic, the ECU appears to be grounded via pin #9 on the gray connector and pin #2 on the black to the chassis ground. Pin #4 (O2BHTR) and #15 (O2BHI) on the gray connector and pin #7 (O2AHTR) and #17 (O2AHI) on the black ECU connector deal with the O2 sensor; it would not hurt to check them for continuity, while making sure that the terminals haven't backed out of the connector, possibly causing a poor or intermittent contact. I would put the old ECU back in and see if the problem returns; if it does and the wiring checks out, then it confirms that the ECU itself is bad.

As to what caused the ECU to fail, if the electrical system is outputting properly and not hitting it with power surges or badly under-volting it (digital electronics do very strange things when they do not receive enough voltage), it could just be a victim of vibration. Most ECU manufacturers use potting epoxy to help guard against components cracking and vibrating loose - your ECU could have a bad internal potting job, which allowed something to crack and move just a hair when it gets hot, causing your intermittent issue. It's hard to tell without autopsying the ECU to look for any loose or burned components; or for the presence of any hairline cracks in the PCB circuit traces.

Now that the bike is running I would recommend that you make a log file then if you have a problem it will be easier to see what is wrong by comparing the log file.

Yeah. So far it's been a mix of hot and cooler weather, and the bike has not once exhibited idle issues. Nor has it thrown a code. So I'll buy a new ECU and send the loaner back.

Charles.

Good deal, glad that you didn't get too discouraged and give up on it. Electrical issues are usually the hardest to diagnose, often times due to the part that is bad being very non-obvious. Your situation in particular, with the one component that you are relying upon to give you an idea of what is wrong being bad itself, which sends you off on wild goose chases, is pretty much the worst case scenario.

I hate to say it, but you seem to be a magnet for the most improbable crap happening. Your RX3 ownership is very atypical, to say the least. On the plus side, it has been a very good learning experience for all involved and those following along.

It's weird tho, it's only the RX3 that gives me fits. I have an SSR Buccaneer with 13k miles and a Benelli Leoncino with 6000 miles and neither has given me as much trouble as the rx3.

I didn't even mention the fact that it takes a super deliberate shift into 6th gear or it will pop out into a false neutral. But every other gear shifts like it's a Japanese bike. I even went to a shorter lever so there was less throw, but it only helped a little bit.

So yeah, the RX3 has not been the best ownership experience. But I still gave the SG400 a shot... glutton for punishment? Or hopeful romantic? Time will tell.

Charles.
Charles.

You have all the hardware, you can download the map and send it to oldschool tuning and upload a map like I have, you can write Mauricio and he will give you the same modification or I can send you my map and in the HOD ECU HACKER you can compare my bin. file with yours and the program will make you a patch that you can upload together with your original bin. file, your bike will have a better bottom and you will not have to drive at high revs, it will have less fuel consumption, less noise, less moto hours, so only the pros.

That sounds really interesting. Can you provide any more information? Heh, I wonder if I could flash the loaner ECU, and make it a fully operational ECU instead of one that throws a code for a missing sidestand safety switch (Which the US RX3 does not have).

Charles.

You might want to try with your broken ecu first to be safe, as it only has a slow RPM problem, as far as the side foot is concerned, if it can't be done programmatically, then you would have to check in the schematic and make a jumper The HOD ECU HACKER program has options to compare bin files. and create patches based on that I played with it, but I did not check exactly generally from what I know, it is changed injection time and dose of air and fuel to it should be changed ignition time, and that's all.
I will be pulling the engine out of my rx3 soon to replace the cylinder and will study these patch option carefully, and did you at your place pull the engine out to replace the cylinder?

Yes, engine has to come out of the frame to get the cylinder off. It's really easy though. Tip: Put the bike on the center stand, and put a book or a block of wood under the rear tire. Now when you remove the swingarm bolt, nothing will move. In this way, you don't have to remove the rear wheel and swingarm first. Saves some time. Also get the bolt-on downtubes that run under the engine off before you unbolt anything else. I put a jack under the engine and balanced it as I took the last bolt out. Then I just picked it up and put it on my workbench. It's not very heavy.

Charles.

Thanks a lot I was worried about the swingarm, now I know that the bike will not turn over by the way I contacted the Chinese guy about the rear shock absorber I want to replace it with a longer one, but it turned out that the spring is 11.5 mm and such a shock absorber is quite expensive about 150$ I have to pull it out and give the exact dimensions to the Chinese guy.

I cut my dog bones and removed 6mm and welded them back together to get a little more ride height.

Charles.

I think a longer shock with a longer spring would be better than the stock one, but if it turns out the spring is the same length but the shock is longer then I'll buy a $60 dollar shock with a 7mm spring only and put mine on. Generally a spring like the one in the rx3 costs about $100.