Failsafe shutdown of VESC

General topics and discussions about the VESC and its development.
Luke
Posts: 2
Joined: 27 Nov 2017, 20:43

Failsafe shutdown of VESC

Postby Luke » 27 Nov 2017, 21:09

I am using a VESC in my sons go kart project that we are working on. It is a kettler kart fitted with a SK3 6374 motor, geared for around 15mph on 12s, but I will reduce this with the RPM limit, so far the torque seems more than enough for some drifting fun :) At the moment I am just testing using a remote control and the PPM input to the VESC while the mechanics are sorted out. After testing I would be looking to use an analogue throttle pedal for him to control himself.

I will need some way to disable the drive of the motor in a few circumstances:
I will have a "dead man" remote control with me while he is riding it so that I can remotely shut it down
The kart will have a foot operated mechanical brake, I will have a switch on this so that he cannot use the throttle and brake together
There is an emergency stop button on the kart in case of an unexpected failure of some kind (throttle pedal, VESC going mad etc)

I am not looking for an emergency stop where electrical braking would be needed, just that the drive is certainly disabled.

I have a large contactor on the battery side which is part of the anti spark startup (contactor bypassed by 100 ohm resistor which closes 5 seconds after batteries are connected) - I do not really want to use this because I think it would not be good for the VESC, but it would at least be failsafe.

What I really need is just to ensure that the FETS are no longer driven. I understand that they are driven from the GVDD provided by the DRV8302 which in turn comes from PVDD1 (pin 29)? Is this correct? Is there any way to interrupt this using a small relay? Would it also be an option to de-solder the gate pins of 3 of the FETS and use a 3 pole relay to switch them off?

Something along the lines of what was being mentioned in this thread:
http://vedder.se/forums/viewtopic.php?f=14&t=224


What would everyone else do for a safety interlock?

Thanks in advance, Luke

pf26
Posts: 309
Joined: 28 Mar 2016, 14:37
Location: FR Valence

Re: Failsafe shutdown of VESC

Postby pf26 » 28 Nov 2017, 11:15

I think you could simply use a switch to force the STM32 in reset in case VESC goes mad, this can be done by connecting NRST (5) and GND (1) on P2 connector. Obviously not 100% failsafe.
Disconnecting PVDD from DRV should work too, and safer. It was shown here how to do http://vedder.se/forums/viewtopic.php?f=7&t=363#p2214.
Disconnecting the VESC from the batteries at speed will make the VESC powered from the motor. Applying braking in this conditions may result in voltage rising very fast, and potentially destructive overvoltage. Not sure if it will actually occur in reality, and don't really want to test myself..

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Failsafe shutdown of VESC

Postby rew » 02 Dec 2017, 19:49

I have had my lab powersupply backfed from braking the motor. Once with the powersupply at its max and a few times less agressive with some margin on the powersupply. But yes, the VESC will remain powered even if it cannot drive the fets as long as the motor turns....

Luke
Posts: 2
Joined: 27 Nov 2017, 20:43

Re: Failsafe shutdown of VESC

Postby Luke » 04 Dec 2017, 18:14

Yeah I've triggered the over voltage on my bench power supply a few times as well, I really should be more careful!

Anyways, the view is that it should be possible to interrupt the power as described often, for example connecting it to a relay triggered by a brake pedal switch?

Also because I'm running 12s I am a little worried about reaching the limit of the drv chip. I have an on board 12v supply from a dc-dc converter, could I supply the PVDD from this instead or does it need the full battery voltage?

Thanks for the help guys, will report back when this is done for future reference

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Failsafe shutdown of VESC

Postby rew » 05 Dec 2017, 21:46

I asked TI about providing the PVDD externally. They answered! The response: DO NOT DO THAT!

That might be (and probably is) just CYA policy, but that's the official word....

(I use a DCDC converter in an application. The datasheet says: "min load: 10% of max load". In my application the load can be very light at times. So I asked the Chinese manufacturer. Their response: We recommend you add a resistor to cause 10% load at all times. But as you suspect the voltage will rise a bit. We tested four production samples, and they came out at 5.8, 5.8 5.9 and 5.8 (I don't remember the numbers). This, in my case is good news as everything on that "officially 5V" rail not only doesn't get damaged but also WORKS up to 7V without any problems. No resistor it is... :-) )


Return to “General”

Who is online

Users browsing this forum: No registered users and 6 guests