This is amazing, thank you so much. If you do not mind, there should be another data point for 20°C/68°F, under the "Fuel System (PGM-FI)" section, "DTC Troubleshooting" subsection, "ECT Sensor Resistance Inspection". For me this is under "5-20", page 112 out of 430. If you give me that datapoint, I could easily interpolate the entire curve, thus allowing to theoretically predict any other value.

Thank you very much for all you've done so far, you're very kind.

 

Yes, under section 5 MGM-FI MIL 7 Blinks.

Standard: 2.3 - 2.6 K Ohms at 20 degrees C / 68 F

 

You sir, are awesome.

That's very interesting... So, the reference value for 20C is allegedly the same as for the old probe. But then the values for 20C are now within range of the values for 80C? (80C corresponds to 2.1-2.6 kΩ, while 20C would correspond to 2.3-2.6 kΩ?). I find that hard to believe. I think there was an edition error in the manual edition, and that they only changed the values for the (80C, 120C) pair; while keeping the old value for the 20C data point.

If that were the case, the ECM could not possibly distinguish any temperature in the 20C-80C range, and I believe it does.

Another point: "16 - 20 psi of pressure" is probably for bigger engines, as they potentially generate more heat. The CBR125 is reportedly using pressures between 13.5 and 17.8 PSI, as per the service manual. 13.5 PSI is lower than 1 ATM, so that seems unlikely. 17.8 PSI is 122.6 KPa, which is 3.13 PSI (21.6KPa, or 0.26 bar) more than 1ATM. So, while I still would not recommend opening the radiator cap when the radiator is hot, 0.26 bar should not be enough difference for the sensor to measure different values when in the pressured loop or at 1ATM. Consequently, testing the probe at 1ATM should be just fine.

So, in conclusion, I will now buy a new ECTS and do measurements when I get it. Thank you very much for your contribution to my thread!

 

Can you point me in the right direction for getting a shop manual for the CBR125? They're not in the US, so finding something for EU/UK based models is challenging.

 

Sorry about the delay, I got side tracked in the physical realm. I think this is the one I use.

 

Wow, who produced that one? I've never seen a Honda PDF so clear.

 

My guess would be Honda. It's quite literally the ebook

Too bad it's for the previous generation and not my model.

 

Do you know where it came from?

 

I looked at the metadata in the file, and it confirms what I wrote above: the seemingly file comes from Honda, and is likely the same file that was used to print the official manual for 2007-2010 CBR125R.

 

So, I know this has been a long time, but I finally got both the necessary tools and parts, and the time to finish this.

I took the NTC probe I ordered (and finally got!) for replacing the ECT sensor on the CBR125R (OEM number 37870KRJ901), and I did a whole bunch of measurements with a multimeter and a digital cooking thermometer.

As I suspected, and as multiple users here also hinted at, my ECT sensor was faulty. As I said in a previous post, the resistance I measured on the old (faulty) ECT sensor was around 2.3-2.6kΩ, at around room temperature (~20°C).

Now, here are the values I got with the brand new one:




I am sharing this information in case someone else has the problem I had.

The approximations I got for interpolating/extrapolating continuous values based on the discrete samples I measured are:

• Exponential approximation: 53.7257 × e^(-0.0357322 × t)
• Quadratic approximation: 47.1482 - 1.2477 × t + 0.011309 × t² - 0.0000328269 × t³

Where t is the temperature of the probe is set at. As a side note, the quadratic approximation only gives valid values for temperatures lower than 92°C (the highest temperature point I was able to go to and remain stable for long enough for a correct reading), as the approximation was computed for 92°C as a minimum, and the values raise after that point (which does not happen with the NTC sensor - the resistance always goes lower as the temperature rises).

The value at 20°C was around 26kΩ, and the value at 80°C was around 2.9kΩ.

I have also attached a zip archive (data.zip) to this post in case someone wants to get all the data (in LibreOffice .ODS format) I used to make the plot I included above.

I hope it helps someone!