Choosing a Capacitor

I need some help picking out some capacitors. I got an arduino board hooked up to this motor driver as well as a xbee shield. All of it sits on Tamiya Bulldozer base. The problem is they are all hooked into the same power supply. I have blown up 2 xbees (yes, an expensive lesson learned) before realizing why you have a separate power supply for the motors. I've hooked it up to an o'scope and have seen spikes at every point I could possibly check. And of course they disappear when the motors are off.

I've been reading and I've learned to put a large capacitor across each motor as well as a large one on the output of the power supply. I am just not clear on what size and what kind of capacitor to use!

I got a 9.6V RC battery powering the Ardunio board. It is also stepped down to 5V which feeds the motor control board, the xbee shield, and motors. Thanks for the help.

Not sure caps alone would solve it, but only some testing can say for sure. You might consider separate 5vdc regulators for the different modules and be sure that all the ground wires from each module are run on their own separate wires all returning to the ground side of the battery.

Lefty

Man, that was the one answer I didn't want to hear. All 3 modules have a 5V or 3V regulator on them. I hooked up the o'scope from the ground to the main 5V regulator, 3V on the xbee shield, 5V on the Arduino Board, 5V regulator on the motor board, and even to itself (basically the two leads were just touching the ground on the battery) and I could still see spikes.

The Xbee shield is the only one that doesn't have it's own ground straight to the battery, it goes to the power supply. I will try changing that tomorrow.

1) I have trouble believing those dinky Tamiya motors were generating enough nastiness to take out your XBee.

2) The fact that you were seeing spikes even when the oscilloscope lead was connected to itself suggests EMI is being generated (not a big surprise) but not necessarily large voltage spikes that would damage anything.

If all your modules have regulators on them already, then they already have capacitors. I don't think this is your problem. I think something is wrong with the XBee hookup. I'm not entirely sure what you mean by "doesn't have its own ground straight to the battery, it goes to the power supply". I thought the battery was the power supply.

Have you read this on decoupling :- De-coupling

Very interesting article. I remember learning about all this years ago, but never got a chance to use it and of course never stuck in my head. I understand the formulas enough to figure out XC or XL if I had the other values. But how to apply it to this real life problem? I just can’t see how XC or XL will tell me which capacitor to use.

RuggedCircuits if you don’t believe the Tamiya motors are the problem, what else do you think could be wrong? There isn’t much to the xbee shield. The 3V regulator is working. Maybe that chip?

RuggedCircuits if you don't believe the Tamiya motors are the problem, what else do you think could be wrong? There isn't much to the xbee shield. The 3V regulator is working. Maybe that chip?

I don't know...I don't really understand how you have things wired up:

I got a 9.6V RC battery powering the Ardunio board. It is also stepped down to 5V which feeds the motor control board, the xbee shield, and motors.

A sketch of all the hookups would really help. It's possible that 9.6V is somehow finding its way into your XBee...really just a stab in the dark.

Perhaps add a 12V transient voltage suppressor (TVS) from the battery to ground (right near the XBee) and monitor that with an oscilloscope again and see if the spikes are limited to 12V. If so, then they're real (rather than just EMI) and those dinky Tamiya motors may well be the problem.

I've created a small powerpoint with what I got. It breaks it down and has some pictures. The last slide shows the spikes. The biggest one was 9Volts. This was measured at the power input for the Motor Control Board. Download Here

Again I would suggest every module should have it's own unique ground wire run to the battery's ground side.

I would also wire the xbee +5vdc input from the Arduino's 5vdc connector pin rather then from the 5vdc regulator that is powering the motor stuff.

Lefty

Nice diagram.

I guess I don't understand the purpose of the 5V regulator. The Motor board has its own regulator and expects 6-26VDC in. You say your XBee shield has its own 3V regulator, so why not power it from the 5V output of the Arduino.

The farther away you can keep the XBee (electrically) from the motors, the better.

Lefty, I've got all the ground wired to a switch right before the battery ground. That's the kill all switch. Each module is wired the same.

RuggedCircuits, the reason I have the regulator is because of the motor/gears. If I let the battery power the Motor Board, then both motor and gears will get the full 9.6V. I did that once. It required a new gear. That regulator only regulators the L289 chip. That's how I understand it at least.

Does that o'scope picture look normal? Am I just chasing a problem that doesn't exist and I should be looking somewhere else for this xbee failure problem?

OK, I can see dropping down the 9.6V for the motors, but there's still no reason to power the XBee from that supply. I'd still try powering it from the Arduino.

That oscilloscope trace looks like standard switching noise from a motor driver. I'm not sure there's enough energy in those spikes to damage anything, but why take the chance.