ESC braking causing spike?
 

I'm interested to know peoples thoughts and if you have collected any data on what seems to be a technical wide issue with braking on BL ESCs.

It seems, that when you brake, there is a voltage spike. When running on the limit, such as the Castle MMM on 6s, the spike can quite easily damage or break the controller.

I recently broke a Mtroniks Brushless setup (a UK band) and was interested to find this problem here too, this is what they said:


This makes me want to know more. What is the actual process in braking that causes this spike?

And how can it be resolved, other than using mechanical brakes... or not braking at all.

Would lowering or even increasing the brake curve reduce this risk?

Thanks,
Nick.

Instead of using the motor as a "motor" under braking it is used as a generator. hence regenerative braking. your best bet is to add a capacitor array to absorb the spike.

How many capacitors and what sort of spec would I need?

lots of info and discussion in this thread:

http://monster.traxxas.com/showthrea...=489081&page=3

Caps will help smoothen the spike, which will help reduce the peak somewhat, but what you want is a transient voltage suppressor, or TVS. It basically shunts voltages above its rating to ground. As you can imagine, a single device has limits, and there is a LOT of kinetic energy when stopping a fast moving vehicle on a dime, and so it can;'t handle the power and blows.

This is especially important when running at the ESC voltage limit because there is not much headroom. If a spike is, for the sake of argument, 50% higher than the battery voltage applied, a 6s setup will generate over 33v. A 5s setup will generate over 27v. As you can see, the 6s setup will work the TVS harder since it is already closer to the ESC rating.

You can run several TVS devices in parallel to help share the load, but even then, there will still be limits. A combination of running a little less voltage than the ESC rating, more gradual braking, and the addition of caps and TVS devices will go a LONG was to curb those pesky spikes.

^ That is why I stick to 5s lipo, and pump my brakes to slow down, rather than stomping on them quite hard. Works for me, and I have the old V1 MMM that is 'a bit flakey' to say the least :smile:

Agreed. The biggest difference you can make, and easiest, way to reduce braking spikes is to just go easy on the brakes, slamming on the brakes is not good for anything, ESC, motor, drivetrain, driveshafts, hubs, etc. I am personally running a Quark Monster at it's limits of 6s LiPo for the last 2 years now with a modified cap bank added directly onto the PCB power input, the controller is known to be weak when running 6s, but I have been running hard for the lat 2 years now with no problems, both racing and bashing. Check sig. down below.

This is how I run my setups.

Will the setup of 6s cells with a lower kv motor generates more spike volts compared with the setup of 4s cells with higher kv motor?
I thought since the current requirement is less in the 6s setup, the spike volt will not higher than that of 4s setup. Is that right?

I've been thinking...

The problem is when you brake/reverse, the motor generates additional voltage.

So, the capacitors need to be inbetween the ESC and motor? The problem is, when you reverse the motor, won't the voltage also reverse down the pos/neg wires? Which means the capacitors have to be non-polarised or they'll go bang.

And which of the 3 wires would I have to connect capacitors too?


I don't understand how putting capacitors on the battery side of the ESC will reduce the spike...?

Putting caps on the battery side gives that voltage somewhere to go, since the internal resistance of your lipos is higher than the internal resistance of the ESC (energy follows the path of least resistance) a lot of energy is contained within the ESC causing serious heat issues as well as a lot of wattage running through circuits and components not built to handle it. By adding caps you are giving some of that ESC-killing wattage somewhere to go and they will also help protect against ripple current (a major killer of ESCs). $5ish worth of components can save you hundreds worth of ESCs!

I think this is something i should do with the muggy perhaps

If all else is equal, the spike on 6s will most likely be higher than on 4s. A lot depends on how you apply the brake (jamming it on vs smooth gradual application).

The motor is not reversed when braking. Basically the ESC pulse-shorts the motor windings (ever feel the turning resistance if you take a raw motor and connect the phases together?) The spikes come between the PWM pulses.

Do NOT put caps on the motor wires!

As I said before, caps won't do a whole lot to reduce the spike amplitude except the fact they take a hair longer to charge does reduce the peaks somewhat.

Well I've put caps on, even if they do very little, it's still something!!


http://www.msuk-forum.co.uk/uploads/..._714_97971.jpg


I've turned the brake power to 30% on the MMM and lowered the curve. So hopefully I don't blow it this time.

(The caps are soldered on, that's silicone instead of heatshrink, incase you thought I've glued them! lol)

I guess we should have told you this before, but are those low-ESR caps, and are they rated for 105 C? If not, then they wil lnot do anything for you, and they will most likely fail if pushed hard at all.

Please tell me they are low-esr. :oops: