Emergency braking concept for VESC

General topics and discussions about the VESC and its development.
polo
Posts: 8
Joined: 24 Feb 2016, 09:53
Location: Lyon, France

Emergency braking concept for VESC

Postby polo » 03 Nov 2016, 13:21

Hi there,

The idea behind this concept is to allow a safe braking mechanism in case of disconnection of the controller or any issue that could lead to the board being stuck in full throttle with no one on board.

The point is to disconnect the motor from the VESC to protect the VESC from any charging current caused by the spinning motor. And when the motor is disconnected from the VESC it is connected to resistors between the 3 phase that will act as an energy dissipation and reduce the current flowing in the winding.
The disconnection between the motor and the VESC will be handled by 3 mechanical relays (48V 30A) that will be triggered by the kill switch.

The expected behavior when the switch is closed is safely braking of the e-skate.

What do you think about that and what impact could have the relay on the motor detection for BLDC tool ? I guess it will add resistor to the motor so is that a big deal ?

Thanks ;)
Attachments
emergency breaking vesc.png
emergency breaking vesc.png (119.22 KiB) Viewed 3524 times

rew
Posts: 937
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Emergency braking concept for VESC

Postby rew » 04 Nov 2016, 08:37

Reverse the contacts on the relays (NO contact goes to the VESC) and use a NC switch. That way if you fall off taking the switch with you the circuit is broken and the VESC should stop.

You have drawn 1000 ohm resistors. At nominal 40V, that will create a braking current of say 40mA or 1.6W. I guess I'd use 50W resistors and aim for 400W of dissipation at full speed. You do the math.

I'm not sure how long the relays will last if you switch them at 50A (or more).

Consider leaving the resistors attached to ONE phase. disconnecting both the other phases will take it out of the circuit. That leaves just two relays, you might be able to find a dual-relay....

polo
Posts: 8
Joined: 24 Feb 2016, 09:53
Location: Lyon, France

Re: Emergency braking concept for VESC

Postby polo » 04 Nov 2016, 09:21

Hi rew,

Yes I understand you point concerning the resistor value. Actually I planned on using 100 Ohm resistor which will give me a 16W braking force. It's obviously quite low so I might decrease these resistors to have more braking.

About relays, I've bought some cheap Songle relays on ebay, I will see how long they last ^^

Can you clarify your last sentence, do you mean using only 2 relays and shorting the 2 motor phases on 1 resistor ? Won't this lead to an unbalanced current on the motor winding ? I think it's not great for the motor, isn't it ?

And about the motor detection in BLDC tool, should I do it with the relay attached (and their NO contact impedance) or just with the motor naked ? I guess it will change the way the VESC is driving the motor ...

pf26
Posts: 293
Joined: 28 Mar 2016, 14:37
Location: FR Valence

Re: Emergency braking concept for VESC

Postby pf26 » 04 Nov 2016, 10:45

I fear you are likely to kill your Vesc if you switch your relays while powering your motor.
Relays are slow devices, they open/close within a few ms, and not all at the same time and bldc motors are inductive. While opening the first relay circuit the inductor wants the current to keep flowing and rises the voltage accordingly. Very large overvoltages or sparks can occur, and I doubt your FETs will sustain this.
To protect agains VESC going furious and powering your motor (I had this on a bike due to short on the throttle wiring) I use an easily removable 30Amp car fuse between battery and Vesc. I added an antispark made of a 47R resistor in series with a switch that I depress (so the Vesc is powered via 47R) before inserting again the fuse.
Now protection against motor powering the Vesc while not connected to battery - I think it would be better to have this implemented in firmware (a detection of very large input voltage variations should prevent from braking, and thus avoid possible overvoltage) -

akira9a
Posts: 15
Joined: 23 Mar 2016, 13:00
Location: Paris

Re: Emergency braking concept for VESC

Postby akira9a » 04 Nov 2016, 12:05

pf26 wrote:I fear you are likely to kill your Vesc if you switch your relays while powering your motor.
Relays are slow devices, they open/close within a few ms, and not all at the same time and bldc motors are inductive. While opening the first relay circuit the inductor wants the current to keep flowing and rises the voltage accordingly. Very large overvoltages or sparks can occur, and I doubt your FETs will sustain this.


Would SSR relays solve this issue ? There are some with heavy current specs !

polo
Posts: 8
Joined: 24 Feb 2016, 09:53
Location: Lyon, France

Re: Emergency braking concept for VESC

Postby polo » 04 Nov 2016, 12:20

akira9a wrote:
pf26 wrote:I fear you are likely to kill your Vesc if you switch your relays while powering your motor.
Relays are slow devices, they open/close within a few ms, and not all at the same time and bldc motors are inductive. While opening the first relay circuit the inductor wants the current to keep flowing and rises the voltage accordingly. Very large overvoltages or sparks can occur, and I doubt your FETs will sustain this.


Would SSR relays solve this issue ? There are some with heavy current specs !


Yes I'm wondering also if SSR could be suitable....

Also maybe a solution would be to disconnect the VESC from the battery at the same time the motor get disconnected from the VESC ?

polo
Posts: 8
Joined: 24 Feb 2016, 09:53
Location: Lyon, France

Re: Emergency braking concept for VESC

Postby polo » 04 Nov 2016, 12:59

So I'm wondering : if I put an SSR (Solid State Relay) to disconnect the VESC (which will be faster than the relays) would this be okay for the VESC ?
(no overvoltage ...)
Attachments
emergency breaking vesc 2.png
emergency breaking vesc 2.png (37.62 KiB) Viewed 3458 times

rew
Posts: 937
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Emergency braking concept for VESC

Postby rew » 06 Nov 2016, 08:40

Assume you're going down the hill. Assume you think you're going to fast, you might fall off at this speed, so you are already applying the brake. THEN you press the emergency button.

SSR shuts off the battery, The VESC starts charging the capacitor until the relays remove the motor from the VESC.....

Assume you're braking full-throttle, 40A motor current at near the max speed (95% pwm) of the system.

So now you have about 40A of current flowing into the capacitors. At 2000uF (2mF), that will create a dU/dt of 20kV/sec, or 20V/ms.

Thus with your SSR shut off, after 10ms when the relays actually switch there has been enough time to increase the voltage by about 200V. Quite enough to blow FETs, DRVs and capacitors. (at such a big voltage change, just doing the dU/dt at the start voltage is inaccurate. In reality only enough energy for about 100V of voltage increase is available. But even the first 1ms of voltage increase is quite enough to fry everything).

The popular songle relays from ebay are rated at 10A. I would not dare put 50A through those....

polo
Posts: 8
Joined: 24 Feb 2016, 09:53
Location: Lyon, France

Re: Emergency braking concept for VESC

Postby polo » 06 Nov 2016, 15:24

Yes, what you say actually make a lot of sense :)

That is why at first I didn't want to disconnect the batteries from the VESC, I know now that i will definitly not do that.

I guess that with 3 phases SSR between the motor and the VESC it could be a solution to disconnect in case of emergency. There is preaty heavy duty SSR out there : http://www.ebay.com/itm/Three-Phase-3Ph ... vX9wORZj-w

But what I don't know is if there will be impact on the VESC for driving the motor through the SSR ...
Any idea on this ?

rew
Posts: 937
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Emergency braking concept for VESC

Postby rew » 06 Nov 2016, 23:00

And you think that the VESC won't mind that the measurements of the phase voltages are off by say 1.5V to 2V? I expect that the VESC WILL mind and that you won't like the extra power consumption caused by the extra voltage drop. SSRs are usually meant for AC and line voltages. So if you lose 2V in a 240V application, you're loosing less than 1%. But when driving slowly at say 10% of the maximum speed of your motor (5V effective) you're losing about 28%.....


Return to “General”

Who is online

Users browsing this forum: No registered users and 6 guests