# ripple current rating. Low vs high frequency...

#### Gibson486

Joined Jul 20, 2012
344
I am trying to figure out the bulk capacitance needed for a stepper driver circuit. I calculated that i need at least 6 uF of bulk (i = C dV/dT).

If my motor takes 1A, then I should get a ripple current of at least 1A, correct?

When I look at digikey, there is a low frequency ripple and a high frequency ripple. When I pick this, do I pick the rating based on my step speed? So, if I am giving the driver pulses at a frequency of 200 Hz, then I care more about the low frequency ripple as opposed to the high frequency? From what I am seeing, low frequency ripple is always much lower than high frequency unless I really get a high cap value (like 100uF). Or am I thinking about this wrong?

#### wayneh

Joined Sep 9, 2010
17,174
This might give some perspective:

#### Attachments

• 1.2 MB Views: 24

#### Papabravo

Joined Feb 24, 2006
19,262
A schematic would help us understand what you are talking about. I for one have never used capacitors anywhere near a stepper motor driver. @wayneh is trying to point out that the actual capacitance the you get is not very reliable when it comes to tolerance. In some cases the uncertainty can be as bad as -80%, +20%. Take your value of 6 uF, apply the tolerance, and tell me what ripple value(s) you come up with.

#### wayneh

Joined Sep 9, 2010
17,174
The reference also describes some of the issues, ie. heat. The way I see it, the problem is ohmic heating I^2•R, where R is the ESR at the frequency of the ripple and I^2 is the time-averaged square of the current Charge moves in and out depending on the frequency of the ripple, and the ripple size. Integrating that over time gives the average I^2•R and that's what the cap has to dissipate. The specs give an example of a ripple and a frequency that are acceptable, and you're left to make a rough approximation of how close to your own needs that one point is. Things like ambient temperature and the temperature rating of the cap come into play.

Personally, I'd throw in a big safety factor so that my ripple current was no more than about half of whatever the spec says.