too long battery wires will kill controllers

[spinningmagnets]

For new members, it costs the same to put the controller near the battery, as it does to put it near the motor. Put the controller near the battery and make sure the battery-to-controller cables are short and fat.

Copper wire is pretty cheap in the lengths we use, so...if the controller-to-motor phase wires are getting hot, make them fatter too...

 

[fechter]

Keeping the battery wires together or twisted will help minimize the inductance. The "loop area" largely determines the inductance. By keeping the wires close, the loop area is minimized. By mounting the controller close to the battery, and further from the motor, the stress on the controller caps and FETs will be reduced, but at the expense of more resistance losses in the phase wires.

 

[zackclark70]

I am bored of seeing this thread

For the average user this will make near 0 difference and there is much more important things for people pulling less than 40 battery amps ( 2kw on a 48v setup) a new controller for a low power setup is not expensive and functionality outweighs the cost of a controller


( just think people buy a new phone every year when they dont really need to and thats £600+ a controller for a bike is less than £150 and will last years )

For the advanced user that pulls over 40 battery amps you will find that the v drop on the battery wires and connections along with battery sag will cause much more damage to the caps than the inductance will because the caps will have far more ripple and to be honest you could just uprate the caps and run bigger battery wires

having long motor wires that are effectively AC will have more inductance and you will need to extend your hall wires and that can also cause issues so in my opinion its better not to substitute 1 issue with another ...

in the perfect world we would have superconducting motors battery's with a crazy low IR and huge capacity and controllers the size of a grape but in the real world we balance functionality and cost

 

[billvon]

The example serves to show there is considerable energy developed by changing current in the inductance of even short cabling. Circuit configuration is different, but the motor controller's body diodes do not absorb voltage transients and ripple from battery circuit inductance.

Right. That's what input capacitors are for. As long as capacitance is significantly higher than cable inductance, the input will present as a voltage source - which is what the controller needs to operate correctly.

Take the case of two 5mm diameter leads, 5mm apart, 1 meter long. You are going to see an inductance of about .3uH. Input capacitance is going to be on the order of 100-1000uF. So in that case capacitance dominates, and the controller works fine. If your cables are >100m long you might stat having trouble in a setup like that (although of course you will run into other problems long before that.)

The body diodes of the FETs capture the flyback from the motor inductance (that's why we can add wire on the motor side with little concern), and handle it nicely. This is not the case for the energy in the battery side inductance, which adds to the battery voltage and increases the system voltage when the battery current is reduced.

Again, that's only true if cable inductance is close to or greater than input capacitance value.

 

[liveforphysics]

In practice, all the noise and spikes come from the rising falling edge of the switching transition. At these speeds the total capacitance is essentially irrelevant to the circuits function, and ESR and ESL of the cap bank becomes all that matters.

I've seen 60v of bounce across a 6in length of 0.25in thick by 1.5in thick copper buss bar.

 

[billvon]

In practice, all the noise and spikes come from the rising falling edge of the switching transition. At these speeds the total capacitance is essentially irrelevant to the circuits function, and ESR and ESL of the cap bank becomes all that matters.

Remove all the input capacitance of your motor controller and see if the performance of your controller changes. I suspect you will find that it does.

 

[liveforphysics]

I didn't say the cap bank isn't critical, merely that the amount of capacitance it holds is not what matters for its function. To actually use the capacity of the caps, the bus voltage would need to swing so greatly it would defeat the purpose of the cap.

ESR and ESL are the metrics that matter with respect to keeping the DC bus voltage stiff. For this reason a 10uF film cap designed around minimum ESR/ESL can do a better job controlling the DC bus for a motor controller than say 10,000uF of energy storage dense electrolytics.

 

[Alan B]

The electrolytic caps used by most ebike motor controllers are not very effective at stabilizing the bus voltage above DC due to their poor ESR and ESL characteristics. It is this higher frequency ripple that causes the problems. The values of these parameters for electrolytic is often greater than those of the input cables, preventing the capacitance of the device from effectively smoothing the bus voltage. Proper controller design uses capacitors of different construction that have very low values of ESR and ESL. Using electrolytic caps is a low cost value engineering technique for modest power levels and results in compromised bus voltage stability. Adding length to the battery cabling makes it worse. You never know what those margins are, so you don't know how much cable you can add and stay within margins (until a few controllers are sacrificed to learn from). With a low power controller, a few inches of cable is unlikely to be a problem (as long as it is heavy enough). But try adding 100 meters and see what happens. Preferably post video and scope traces for public enjoyment. Since we don't accurately know how much cable is safe, why add any?

More info on electrolytic capacitors:

http://www.cde.com/resources/catalogs/AEappGUIDE.pdf

 

[billvon]

I didn't say the cap bank isn't critical, merely that the amount of capacitance it holds is not what matters for its function.

?? Of course it matters. That's why it's there, and why it is as large at it is.

To actually use the capacity of the caps, the bus voltage would need to swing so greatly it would defeat the purpose of the cap.

You are not using very much of the energy storage of the caps. You are using them to manage impedance; to make sure the input to the motor controller remains "stiff" (i.e. a good voltage source) at all frequencies.

ESR and ESL are the metrics that matter with respect to keeping the DC bus voltage stiff.

ESL has no meaning at DC; it only kicks in at AC frequencies. ESR is indeed an issue; this presents as voltage droop under a DC load, which is not an inductance effect.

For this reason a 10uF film cap designed around minimum ESR/ESL can do a better job controlling the DC bus for a motor controller than say 10,000uF of energy storage dense electrolytics.

Right - but that is an AC consideration. Which is why controller input stages have both cheap electrolytics (to ensure low impedance at low frequencies) and film/ceramic caps (to ensure low impedance at high frequencies.) Both are important.

 

[liveforphysics]

What components of the 'noise' or 'ripple' isn't AC again?

 

[granolaboy]

Guys, thanks for the thread...I'm currently assembling a plan to put an AC-75 and a zilla controller in my Toyota 4x4, and I hadn't even considered inductance on long battery leads hurting the controller. Seems obvious after reading it, and i'm a bit embarrassed for it not being on my radar in the first place.

I suppose if you are running a 40A controller at 48v you probably don't need to worry about any of this, or probably don't care because you are looking for an excuse to upgrade your $50 controller anyway. I could have been 50 miles deep on a decommissioned logging road staring at an expensive fried controller and a long walk home.

Threads like this remind me to step back a second and do a reality check once and a while...

 

[argo]

Hi
Can someone please comment? Would these capacitors (5 in parallel) be suitable for smoothing the ESC input to a 14S 140A system...?
https://nz.rs-online.com/web/p/aluminium-capacitors/0571268/

Panasonic FC Series Type A electrolytic capacitors offer a high endurance, low impedance solution. These aluminium capacitors feature endurance between 1000 and 5000 hours.

Specs:
Capacitance 560 µF
Technology Electrolytic
Voltage 100 V dc
Ripple Current 1.77A
Leakage Current 3 μA
Lead Diameter 0.8mm
Equivalent Series Resistance 41mΩ
Lifetime 5000h