Some advice on a motor controller capacitor placement!

Thread Starter

bytraper

Joined Sep 28, 2010
127
Hi people!

I'd like to get some advice on what is the correct way to hook a main capacitor up to a speed controller.

In the past I've been just hooking the main capacitor after the switch to turn it on, the same as the image on the right. However, recently Ive been told that its not correct, and that the setup on the image to the left is correct. That the main capacitor must go between the motor + and - and a smaller capacitor should be used on the beginning of the control side.

Can anybody advise me on what the correct method should be ?



Thank you for any and all advice. It is appreciated!
 

Kermit2

Joined Feb 5, 2010
4,162
The 'correct' way to do it depends on the final application and situation.

The cap can filter RF range noise out of the circuit, and/or be used to decrease arcing, and sparking in the commutator, brush portion of the motor. Switch debouncing is also not to be overlooked

It acts much like a decoupling cap would when used with IC devices.
 

thatoneguy

Joined Feb 19, 2009
6,359
The 2200uF cap is electrically in the same place on both schematics, on the first one, you add a 470uF in parallel with the 2200uF when switch is on and MOSFET is conducting.

For brushed "hobby motors", it is common to put a 0.1uF or 0.01uF cap across the terminals to reduce RF interference.

That large of capacitor wouldn't react fast enough at the frequency the motor runs, Electrolytic capacitors are higher impedance than ceramic or metalized film, and don't charge/discharge as quickly.
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
Hmm,

I didn't show it in the example above, but I run a snubber in the form of a 0.1uf in series with a 33 ohm resistor across the motor output terminals, but what I've found from experience is that if people have the motor wires too long then they need to put another 0.1 capacitor on the actual motor. I guess the wiring has inductance.

My main concern with the way I did it (the image on the right) was that I'd have to run quite thick traces to wherever the power switch is, meaning if there's spikes that aren't filtered, then they go through the switch. In my limited experience, this causes the switch to jam by arcing.

The other way, with the large capacitor right on the battery input side and the other capacitor on the control side I don't have that problem.

But either way I do it, if I put a coil on the output of the controller and run it up to 30a and leave it for about 5 minutes the capacitor is really really hot, so I wasn't sure if I was doing the right thing.
 

thatoneguy

Joined Feb 19, 2009
6,359
You are correct on capacitors reducing arcing, but if you capacitor is getting hot, that means it is passing a LOT of AC ripple to ground.

Could you draw a schematic of how it was hooked up when the capacitor was getting hot? BTW, How Hot? Uncomfortable to touch for more than 5 seconds, or uncomfortable to touch at all?
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
Its basically like this (see diagram below). At first I had only a 100uF 63v (where the 2200uf 63v now is) on the front end, but when I put an inductive coil on the outputs of the controller, I noticed that at first, it blew up the 100uf 63v cap, so then I stuck a 2200uf 63v on there, and then it burned out the track leading up to the cap and left the cap very hot.

Then I removed the 2200uf 63v and put it directly on the battery input terminals and put the 100uf 63v where it was originally (where the control side power enters). This saved the control side and that cap on that rail didn't blow up or get hot or burn out the trace any more, but when I left it (for the 5 minutes I was talking about in the previous post), the cap I put across the input battery terminals (2200uf 63v) was very hot, it was touchable, but not for 5 seconds, maybe 1.

Its max current is 30A at 12V. When i tested it, I had 20A running through a home made coil which acts like a heating element
I was thinking to add another cap in parallel because lot of designs I see use 2x capacitors at the front end and there must be a reason for this, maybe they share the pain :)



But throughout all the torture I've given this prototype controller, the control side hasn't blown up which im thankful for because this is my last board!


 
Last edited:

thatoneguy

Joined Feb 19, 2009
6,359
What is this supposed to do? I don't see a motor or inductor in the schematic above, though it may be the yellow on black confusing me.

When you are burning traces on a PCB and caps are exploding, something isn't designed right.

I see the control circuit made from the LM317 Regulator, and diodes, but no specific load connections/positions.
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
I've updated the drawing and changed it to colours that might be more suitable for you?

Either way, running the 2 caps in parallel solved the problem! But I'm not entirely sure why.
 
Last edited:

thatoneguy

Joined Feb 19, 2009
6,359
Which two caps did you put in parallel that solved the problem?

Can you make a new diagram showing that layout, leaving the diagram above for comparison purposes?
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
I ended up just ditching the switch altogether, and doubled up the 2200uf 63v.

I'm still not sure why its working properly now, but it is. I think in light of what I've been playing with over the last few days, my preference would be to put the switch after the main capacitors.
 
Top