Emergency braking concept for VESC

General topics and discussions about the VESC and its development.
tlonsky
Posts: 4
Joined: 26 Apr 2017, 12:59

Re: Emergency braking concept for VESC

Postby tlonsky » 07 Jun 2017, 07:42

Hi, so is there some possibility to unplug the motors from vesc? I am building a robot with vesc and i have to solve some emergency button. If there will be some failure in my program (which is giving the commands to vesc) the worst case is that there will be command full gas and i need to stop the robot. So i was thinking on the relays as mentioned in previous posts controlled wirelessly. Or is it better to put relay in between driving signal from processor and vesc. For example to unplug the broken processor and plug different cpu with simple command "stop".
Also i will control 4 vesc via can. So if the can communication will mess up can i plug also different processor to RC input on all vesc and simulate braking signal. What will happen to vesc if he receives two signals from can and from RC?

kitewinder
Posts: 16
Joined: 29 May 2017, 20:47
Location: Bordeaux

Re: Emergency braking concept for VESC

Postby kitewinder » 11 Jun 2017, 11:40

Hi there,

I'm working on braking modes with the vesc as well.
There is two ways to break a bldc: regerative braking and direct coil current sink through power capacitor as depicted in this topic.
There must be a way to control the mosfet so that you could either use regenerative braking or shorten the coil . if you can do that, you can also modulate the triggers of the mosfet so that you have a dissipative braking trough the mosfet without adding any resistor.

see attached files :
direct coil shortening will be to trigger MOS D, E and F.

It might even be possible to balance between regenerative braking and coils shortening.

alim moteur brushless.jpg
alim moteur brushless.jpg (27.72 KiB) Viewed 2262 times


I will investigate on this issue and post later but if anyone has already some infos on that, go for it.

olivier

rew
Posts: 926
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Emergency braking concept for VESC

Postby rew » 11 Jun 2017, 21:46

Assume for a moment that I'm on my E-bike, going 7m/s with 100kg. My motor has 30mOhm of resistance.

The energy available for braking is 0.5 * m * V^2 = 0.5 * 100 * 49 = 2500J.

Now if you brake without regenerating that energy has to go somewhere. If your mosfets have 1.5mOhm resistance and the motor has 30mOhm resistance, the energy will be dissipated 95% in the motor, 5% in the mosfets. So you'll end up with about 2375J of heat in your motor.

kitewinder
Posts: 16
Joined: 29 May 2017, 20:47
Location: Bordeaux

Re: Emergency braking concept for VESC

Postby kitewinder » 12 Jun 2017, 07:05

So what ?

Assumes your motor is 0.5 Kg of steel . It has a thermic capacity of 500 J / Kg / K. You will finish with a 10 degrees increase in motor temperature.
If you have a good thermic dissipator for you motor, that will be in no way an issue.
Moreover you can control the Mosfet so that it is in between On and Off State so that you could balance impedance in a much better proportion that 95 / 5 . let's say 50 / 50 .

As I say, I will investigate on this issue this week and let you know about my results. What is for sure is that the Mosfet bridge configuration is enough to deal with braking issue . Extra components are not needed.

arvidb
Posts: 223
Joined: 26 Dec 2015, 14:38
Location: Sweden, Stockholm

Re: Emergency braking concept for VESC

Postby arvidb » 12 Jun 2017, 11:45

Well, yes, with good cooling (i.e. VESC6) you might get away with dumping 5 % of 2500 J into your MOSFETs, if you don't do it too quickly. No way you will be able to dump 50 % into them though!

kitewinder
Posts: 16
Joined: 29 May 2017, 20:47
Location: Bordeaux

Re: Emergency braking concept for VESC

Postby kitewinder » 12 Jun 2017, 14:42

Mosfet are meant to be mounted on a massive heatsink. So yes, you had better deal with this extra power otherwise you will get in trouble !
IRFS730 can drown up to 375Watts !

rew
Posts: 926
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Emergency braking concept for VESC

Postby rew » 12 Jun 2017, 17:49

Where did you get THAT idea? Oh... the datasheet?

That is under ideal conditions. Usually things like: "if you keep the case at 20 degrees". So if you connect a 0.5K/W heatsink with 0.2 K/W of heat-paste, you'll be keeping the outside of the heatsink near absolute zero to make that happen....

devin
Posts: 244
Joined: 08 May 2017, 01:55
Location: San Francisco, California, US

Re: Emergency braking concept for VESC

Postby devin » 13 Jun 2017, 17:50

kitewinder wrote:I'm working on braking modes with the vesc as well.
There is two ways to break a bldc: regerative braking and direct coil current sink through power capacitor as depicted in this topic.
There must be a way to control the mosfet so that you could either use regenerative braking or shorten the coil .


rew wrote:Assume for a moment that I'm on my E-bike, going 7m/s with 100kg. My motor has 30mOhm of resistance.

The energy available for braking is 0.5 * m * V^2 = 0.5 * 100 * 49 = 2500J.

Now if you brake without regenerating that energy has to go somewhere. If your mosfets have 1.5mOhm resistance and the motor has 30mOhm resistance, the energy will be dissipated 95% in the motor, 5% in the mosfets. So you'll end up with about 2375J of heat in your motor.


I'm imagining the deceleration of the rotor from the coil current sink via winding resistance would be so abrupt and violent that it would lock up the motor essentially immediately (since the coefficient of friction between the tires and road would not be sufficient to maintain kinetic energy transfer from the rider directly to the rotor...). Once the motor locks, the wheels can no longer turn so, I'd imagine a good portion of that riders kinetic energy would actually transfer into the burning/skidding of tires, and the shredding of flesh as the rider disconnects from their vehicle and makes contact with the asphalt.

Also I would imagine dumping all that kinetic energy into the coil would be extremely mechanically violent from the stator's perspective as well since that initial huge deceleration force torque load would then have to transfer via the stator to the motor mounts to whatever structure is mounting the motor. If these structures were not designed to handle this amount of torque it could lead to some sort of explosive mechanical event with the vehicle as well (imagine at a high rpm if the rotor and stator were to almost instantaneously become mechanically coupled/engaged) -- something would deform or break.

rew
Posts: 926
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Emergency braking concept for VESC

Postby rew » 13 Jun 2017, 19:12

As "stopping the wheel and burning rubber" instead of the motor is not a way to stop in a controlled manner. So the idea would probably include PWM-ing the "brake" to a level that prevents the wheels from going in a slip.

I suspect that there is a limit to the amount of force the magnets can excert on the wheel, no matter how strong the current is in the stator coils. So, if you have a 1000W motor, meant to run 15m/s (54km/h), it will be excerting 66N on the road at that speed. I wouldn't be surprised if the magnets were calculated for say a max 100N force onto the road. At 100kg "all up" weight, that would give you a 1m/s deceleration. The friction coefficient is likely 5 or 10 times higher.

This limit does not influence where the energy ends up, except: not in the burning rubber.

devin
Posts: 244
Joined: 08 May 2017, 01:55
Location: San Francisco, California, US

Re: Emergency braking concept for VESC

Postby devin » 13 Jun 2017, 19:34

rew wrote:As "stopping the wheel and burning rubber" instead of the motor is not a way to stop in a controlled manner. So the idea would probably include PWM-ing the "brake" to a level that prevents the wheels from going in a slip.

I suspect that there is a limit to the amount of force the magnets can excert on the wheel, no matter how strong the current is in the stator coils. So, if you have a 1000W motor, meant to run 15m/s (54km/h), it will be excerting 66N on the road at that speed. I wouldn't be surprised if the magnets were calculated for say a max 100N force onto the road. At 100kg "all up" weight, that would give you a 1m/s deceleration. The friction coefficient is likely 5 or 10 times higher.

This limit does not influence where the energy ends up, except: not in the burning rubber.


Rider's 2500J = 2500 Watt Seconds

Assuming the rotor is generating 40V of Back EMF initially at cruising speed

40V is at the start of deceleration but 0V is at the end, so average voltage during deceleration is 20V

(Ohm's law: 20V, 0.03ohm, 13333.33W, 666.66A)

Since an average of 20V applied to the winding transfers an average of 13333.33 Joules per second to the 0.03ohm winding, and the rider has 2500 Joules

2500 Rider Joules / 13333.33 Joules transferred per second = 0.1875 seconds

Therefore assuming ALL of the 100kg rider+vehicle's kinetic energy at 7m/s (15.66miles per hour) transfers efficiently to the rotor during deceleration I think the rider should come to a nice even stop from cruising speed in about 0.1875 seconds.

:(

that's about -3.8g of acceleration on the human body, almost as much as experienced by the world record holder in top fuel drag racing.

sources:
http://www.smartconversion.com/unit_calculation/Acceleration_calculator.aspx
https://en.wikipedia.org/wiki/Orders_of_magnitude_%28acceleration%29


Return to “General”

Who is online

Users browsing this forum: No registered users and 3 guests