Older Solo 60

General discussion of using Roboclaw motor controllers
Post Reply
User avatar
Basicmicro Support
Posts: 1594
Joined: Thu Feb 26, 2015 9:45 pm
Re: Older Solo 60

Post by Basicmicro Support »

1. Please elaborate on the noise. Neither the slightly larger DC Link caps nore the TVS diodes will have much effect on noise. TVS diodes only do anything when the input voltage spikes above their clamping voltage limit which is above 34v on the Solo 60. The larger caps increase the UF slightly but that has little effect. The ripple current rating is the important value. That hasn't changed significantly from the first release. We went to a larger size because we can get them more easily than the smaller caps.

If the noise is on the 5v rail the difference is most likely in the smaller caps on the board. If it is high frequency noise(sharp narrow spikes) adding a .01uf cap to 5v will probably fix the problem. If the noise is low frequency adding a 10uf cap to 5v should fix the problem.

If the noise is somewhere else(encoder lines for example) let me know.

2. When measuring noise you need to make sure you are using as short a ground lead as possible. Otherwise you could be seeing 2 or 3 times what the actual noise is. You want the loop between the probe and ground to be inches at the most. 6 to 11 voltage of "noise" is huge. So I think at least some of that is your measurement setup. Your power source could also be playing a roll. Also it depends on what you are calling noise. If it is high freqnency and near spikes that is noise. If it is the voltage source sagging at fairly low frequency(under the PWM switching frequency of 20khz) that may be a power source problem or your power wires are too long(2 feet is not too long) or too small(if you have 12aug that isnt likely either).

Shorter power wires is important. Long motor wires are less of an issue. Wire length increases inductance. Motors are already huge inductors so that isn't much of a problem. However Power wires need to deal with quick transients without larger voltage changes. The DC Link caps help with that but it is best to keep your power wires as short and fat as possible.

In your case though a couple feet isn't an issue as long as you aren't under sizing the wire gauge. With 12awg wire 3 to 6 feet in general is fine.

Assuming your noise is no probe induced noise then adding more DC link capacitance will probably fix the problem. The 2x60A is much bigger and has much more bus capacitance so its not too surprising it would have less noise, but I would consider 6v noise(not voltage droop) to be a lot.

Note that motor inductance can have some effect on that noise. If your motor was designed for a high frequency PWM(eg 60khz) then running it on a 20khz controller(like ours) would cause a lot of noise on the power input. With the bus capacitances on either controller the power voltage shouldn't be changing by even 1v(again, talking about noise at or around 20khz, not voltage droop due to inability of the power source to supply the power the motor wants).

A scope trace of the noise would help me narrow down the issue though.

Post Reply