I've already found a generalized formula for calculating ripple current. This is for switching type applications.
I = Vcc / (2xFxL)
However the missing factor is the inductance of the motor.
I'm using Pololu 12V brand motors, 300mA free run, 5A stall.
Anyone have any experience or test data that would give a general indication of winding inductance in that spec range?
Obviously its not very practical to have 5A+ of ripple caps just to cover the full spectrum.
I = Vcc / (2xFxL)
Are you sure the "L" refers to motor inductance. You said it applies to switching applications. Are you sure it isn't a reference to the inductor in a switching power supply ?
raschemmel:
I = Vcc / (2xFxL)
Are you sure the "L" refers to motor inductance. You said it applies to switching applications. Are you sure it isn't a reference to the inductor in a switching power supply ?
In a motor driver the motor inductance serves as the inductor of a buck-converter whose
output is used directly in the winding itself to overcome rotational back-EMF!
Its the same problem. Basically that formula is a restatement of the differential
equation for an inductor, namely:
dI/dt = V/L
There's no easy way to know the inductance of a DC motor since manufacturers
rarely quote them (stepper motors are an exception). Try 8kHz, and if that's too rough
or the acoustic noise is objectionable, double it.
Actually ripple current is usually quoted as r.m.s., but that formula is peak-to-peak...
Actually ripple current is usually quoted as r.m.s., but that formula is peak-to-peak...
They use mean average, which is more accurate when describing a square wave, r.m.s is for sine waves.
:~
Current ripple and voltage ripple are not the same thing, at least in this context. I really don't care what the capacitance of the cap is, what I need is for it to be able to supply a large amount of current during the switching process on a h-bridge so as to not over stress the battery on random spikes and keep the other electronics from being affected by the large current draw/ripple.
Do you really have to calculate the ripple current. ?
How long would it take to buy a cap rated for 1000uF to 4700uF ?
Wouldn't that make your ripple calculations unnecessary ?
harddrive123:
Actually ripple current is usually quoted as r.m.s., but that formula is peak-to-peak...
They use mean average, which is more accurate when describing a square wave, r.m.s is for sine waves.
The mean ripple current is zero by definition!! You probably mean mean of absolute value.
Ripple current is usually a triangular or sawtooth wave in power-conversion, not a square
wave.
Electrolytic capacitors are rated by r.m.s. ripple current because its the heating
effect of the current that matters. I suppose its really a heat-dissipation rating
(in conjunction with the ESR).
How long would it take to buy a cap rated for 1000uF to 4700uF ?
Wouldn't that make your ripple calculations unnecessary ?
No, I could buy 2 caps both at 1000uF, one can supply 250mA ripple current, the other can supply 15A of ripple current. The 250mA ripple cap is utterly useless in this type of application. The caps are acting like a temporary power supply for the initial current draw, if they aren't large enough the surge will need to be fed by the battery/supply directly and cause a ripple on other electronics being powered by the battery or supply.
The ESR value is more or less proportional to the ripple current of a cap, capacitance in this application is a side point. I my search I've seen 3A 680uF caps and 2.5A 2200uF.
I don't know where you get you information but caps are never rated by amps of current. They have a capacitance value and a voltage rating on the label and that's it. In 30 years of working as a technician with every possible type of component , I have never seen a current rating on a cap. If you can post a photo of a cap with such a rating on the label I would like to see it. (there is also a temperature rating but that is not relevant here).
see attached.
So where are you getting the rating in amps that you are giving ?
Go to digi-key this second, aluminum caps, there will be a spec "ripple current" Pick a cap value, you will be able to see many different choices for the ripple current.
People use caps as power supplies all the time, depending on the load and capacity of the cap, they are just as good as a battery for 0.00001 sec or 2 years.
Ok. I see it. I never had a reason to look that up and since it isn't on the label I never thought about it before. I generally only look up the specs the engineers request. Thanks for the info. I guess old dogs do learn new tricks.. XD
One of the difficult problems designing power electronics is finding capacitors with
high ripple current ratings, its a key parameter in any switch-mode supply.
