Can I wire temp sensors from two hub motors to Cycle Analyst

I am running two matching QS Motors hub wheels in a three wheeler. The controllers are both Kelly KLS7230S. From what I can gather here on ES, most folks running dual motors only use the sensor in one of the motors hooked to their Cycle Analyst 3. My concern is that the non-wired motor might, for some unforeseen reason, overheat and cause damage. Is it possible to either:
a) wire both temp sensors to the CA3 or
b) wire one motor temp sensor to the CA3 and the other motor temp sensor to its respective controller. The Kelly controller would then shut down power to the one wheel in an overheating situation and the CA3 would shut down power to BOTH wheels if it senses overheating in the wheel it is wired to.

B is your only option AFAIK.

But then again, have you even tried to overheat them? I ran a qs1500W hub on my scooter at 5Kw and I never managed to overheat it, even on steep hills.

eee291 said:

But then again, have you even tried to overheat them? I ran a qs1500W hub on my scooter at 5Kw and I never managed to overheat it, even on steep hills.

Click to expand...

I'm also thinking about overheating due to some sort of electrical or mechanical malfunction in the wheel...something beyond putting the motors to a tough test. I guess there is also a third option, wiring each hub wheel temp sensor to its respective controller. I also think "b" in the OP is my best option providing it wouldn't create some unwanted electronic side effect or danger.

There's another option: wire the sensors to a switch, to toggle between them from the CA.

You can use a mechanical one that you manually switch, or you could make an automatic electronic switch that toggles between them every so often.

cboy said:

I'm also thinking about overheating due to some sort of electrical or mechanical malfunction in the wheel...something beyond putting the motors to a tough test.

Click to expand...

JLE offered some suggestions on this way back in early CA3 development:

justin_le said:

...
If you get two 20K thermistors and wire them up in parallel, then it behaves like a single 10K thermistor that (due to the high nonlinearity) is mostly weighted to show the higher of the two temps.

You could also have two semiconductor resistors that have a voltage output and link them up to the temp input via a diode, and then just the highest one would show.
...
-Justin

Click to expand...

amberwolf said:

There's another option: wire the sensors to a switch, to toggle between them from the CA.

Click to expand...

Interesting approach...thinking outside the box.

teklektik said:

justin_le said:

...

You could also have two semiconductor resistors that have a voltage output and link them up to the temp input via a diode, and then just the highest one would show.
...
-Justin

Click to expand...

Click to expand...

This sounds like it might be the direction to go. Unfortunately electronics are well beyond my pay grade...but I can follow directions pretty well. I'm running QS Motors hub wheels with KTY83-122 sensors (1000-1020 ohm, Positive Temperature Coefficient). What size semiconductor resistors would I use and what diode? Also, where should these components be placed in the wiring set up?

cboy said:

This sounds like it might be the direction to go. Unfortunately electronics are well beyond my pay grade...but I can follow directions pretty well. I'm running QS Motors hub wheels with KTY83-122 sensors (1000-1020 ohm, Positive Temperature Coefficient). What size semiconductor resistors would I use and what diode? Also, where should these components be placed in the wiring set up?

Click to expand...

Some notes:

For this approach - "If you get two 20K thermistors and wire them up in parallel, then it behaves like a single 10K thermistor that (due to the high nonlinearity) is mostly weighted to show the higher of the two temps" - then all you do is replace the thermistors and parallel them. You do have to open up the wheel.

For this approach - "You could also have two semiconductor resistors that have a voltage output and link them up to the temp input via a diode, and then just the highest one would show." - this takes some electronics, because you have to match the voltage input to what the sensor expects. Most systems, for example, use a 10K pullup resistor with a 10K thermistor, and use a bias of 5V - so at 25C you get 2.5 volts.

To do this, you'd replace the thermistors with 'semiconductor resistors that have a voltage output' (like the LM34) then use two diodes to OR the output. Then you'd have to bias it (and potentially amplify it) so that when you hit your max temperature, the voltage at the output is the same as what the CA thinks that temperature is.

If you don't want to open the wheel you have a harder task ahead of you. You could build something external to bias the thermistors, then buffer them, then use diodes to OR the result, then probably buffer it again to go back to the CA. This would require a quad op amp, a bunch of resistors, two diodes and a few caps.

billvon said:

For this approach - "If you get two 20K thermistors and wire them up in parallel, then it behaves like a single 10K thermistor that (due to the high nonlinearity) is mostly weighted to show the higher of the two temps" - then all you do is replace the thermistors and parallel them. You do have to open up the wheel.

For this approach - "You could also have two semiconductor resistors that have a voltage output and link them up to the temp input via a diode, and then just the highest one would show." - this takes some electronics, because you have to match the voltage input to what the sensor expects. Most systems, for example, use a 10K pullup resistor with a 10K thermistor, and use a bias of 5V - so at 25C you get 2.5 volts.

To do this, you'd replace the thermistors with 'semiconductor resistors that have a voltage output' (like the LM34) then use two diodes to OR the output. Then you'd have to bias it (and potentially amplify it) so that when you hit your max temperature, the voltage at the output is the same as what the CA thinks that temperature is.

If you don't want to open the wheel you have a harder task ahead of you. You could build something external to bias the thermistors, then buffer them, then use diodes to OR the result, then probably buffer it again to go back to the CA. This would require a quad op amp, a bunch of resistors, two diodes and a few caps.

Click to expand...

Sounds well beyond my meager grasp of electronics. Plus I'm not anxious to be pulling apart the hubs at this juncture. Perhaps I am better off simply wiring each motor temp sensor to its respective controller and call it a day (this at least provides protection for both motors and is straightforward enough for me to wire and understand) I give up the great feature of the CA3 to monitor and display motor temps and and to limit power to the motor(s) should they overheat. I am still curious, however, if one motor temp sensor can be wired to the CA3 while the other is wired it's respective controller. This allows display of temps in one wheel while having power limiting protection in both wheels in the event of overheating. It also seems simple and doable. I just don't want to screw something else up by wiring this way.

cboy said:

Interesting approach...thinking outside the box.

Click to expand...

Oh, I reused the box for something else a long time ago.



As far as one motor > controller and one motor > CA , that should work fine, and it's likely that both motors would be the same temperature unless you happen to be doing donuts at high power, or spiralling up a mountain road to the top.

amberwolf said:

As far as one motor > controller and one motor > CA , that should work fine...

Click to expand...

Thanks amberwolf. May not be optimal but it does provide a reasonable degree of protection for both motors while allowing pretty good visual monitoring of the temperature situation...and it is simple. I like simple.

I run 2WD with only a single motor temp sensor going to the CA. This avoids mechanical contacts (switches) in analog signals where tiny fractions of a volt are significant and relies on the supposition that with equal wheels and controllers, identical motors are doing about the same thing. Since the single CA/throttle controls both motors, they both get thermal rollback protection. Easy peasy.

Don't turn this into a Science Project....