https://groups.google.com/g/sci.electronics.design/c/AGG65Xi3wac?pli=1
TL;DR
Reeds are crap. They can get stuck for being closed for too long, a bit of current or even a shock…
1 Like
Interesting read 
I’m switching to a hall sensor instead, will be more reliable.
I check the VESC output voltage and I have only 1.3V
instead of 3.3v, and my switch is normaly open, that why I need to put ADC2 low
So when I close the switch I have 0V at ADC2, and when I open the switch I have 1.3V at ADC2
I think to solve the problem I must either increase the voltage to 3.3V, but I don’t know how, or install a normally closed switch, but i think i will an issue during the run due to a false detection.
So I think I have no choice if I want to use this function I must increase the output voltage at 3.3V
I test again the voltage, in fact I have between vesc 3.3V and the GND => 3.3V
So all is ok at the VESC output.
but ADC2 and the GND => 1.2V
The question is, why I have 1.2V at the ADC2 when is disconnected ?
Perhaps it’s the same issue at this link
ADC1 problems with Firmware/Hardware issue? | VESC Project (vesc-project.com)
I found a solution
I plug a 10k pull-down resistor on ADC2 and all works fine 
Very happy
0V on ADC2 when kill switch is open
3.3V when kill switch is close
3 Likes
@FoilEngineer you can’t simply block the main battery while the motor and ESc are running full power, it will send back a high voltage pulse and destroy your electronics.
to kill propulsion power, you need to ramp down to zero RPM and then kill power.
We have not found anything that will do that.
I have four fully stuffed Maytech electronic switches that say they can switch high current, but they can’t.
@ausdroid it is not blocking the main battery, it is turning off the PWM or whatever you use to control the motor speed. It has nothing to do with the battery or high current killing
It is the same as going full throttle then taking your finger off the throttle or loosing signal so no need for any of those Maytech electronic switches. Those are used to switch on the VESC under no load to stop spark during battery connection. If you use it to connect or disconnect the battery under load it will surely blow the unit as that is not its intention.
1 Like
Thanks for the info @jeffM, found this post just at the right time for me, I’ll test the VESC kill switch feature in the coming week.
What do you use as lanyard? The leashes i’ve found are all too long to work on kill switches
I like the idea of using the VESC killswitch feature together with a hall (or similar) sensor.
Does anyone have a recommendation on which sensor to buy?
It’s not critical with the exact sensor, there are hundreds that will work but it should be a ”unipolar digital” or ”unipolar switch” sensor type.
It must not be a latching or ratiometric/analog hall sensor.
1 Like
Just a question: If your switch is normally open (open circuit with lanyard attached) then connecting a pull-down resistor from adc2 to ground will ensure low status during riding but when you pull lanyard the circuit closes and ADC2 goes high, therefore the kill switch setting in the VESC should be ADC2 high, right?
It seems to me you describe the opposite?
EDIT: tried it connected and set up like i describe it above, works as intended
Would you recommend I add a magnetic kill switch? I have been considering one as I will be using mine in the ocean, so if I fall and the board takes off I’m out of luck. I’m not sure what the advantage is of having the VESC cut power after loosing reception.
I think it’s generally a good idea for safety but i don’t know yet if the leash to it will be annoying while riding, that’s a possible downside.
Hi - so essentially your connecting the 3.3V pin in the Comm port cable to the ADC2 pin - with a switch in the middle - correct? You could even use a physical switch?
Yes and yes, any type of waterproof switch will work.
Hi ! Thanks for sharing ! I wondering where you find the screen ch1 et ch2 monitoring in Vesc-tool. I use Vesc tool last and Vesc 3.02 and wasn’t t able to see the voltage monitoring. Thanks for your help