Re: Capacitor sizing question

--- In CAD_CAM_EDM_DRO@yahoogroups.com, "turbulatordude"
<davemucha@j...> wrote:

>
> Hi all,
>
> I'm a little on the fence here and hope someone can offer some
> guidance.
>
>
> With a 40 volt transformer and a bridge rectifier I should be

getting

> 38.8 volts before the cap. After the filter cap, the voltage would
> be 54.8 volts.
>
> For the motors, the calculation says 8,000uF so with a cap of

5,600uF

> at 63 volts I can parallel two of them and get more than enough
> capacatance, but am I cutting the voltage too close ?
>
> I'll be using Gecko's with a 470uF-100v on each Gecko.
>
> Typically the rule of thumb is 20% over for caps and this is only

10%

> over.
>
> This is for a plasma table and hopefully high rapids as the norm

for

> cutting.
>
> Any recomendations for the caps ? am I being greedy for trying to
> use a $1.50 capacitor (on sale) or should I get a much larger

voltage

> capacitor ?
>
> Dave

There are two things you need to worry about Dave. The voltage from
your secondary is of course porportional to the input and you need to
design using "worst case" numbers, i.e. high line conditions. While
the nominal AC voltage is ususally 120VAC it can be as low as 110VAC
and as hign as 125VAC (with possible surges to 130VAC). Caps have a
breakdown voltage that is a function of the dialectric and spacing
across a temperature range and like all electronics the manufacture
will build in some safety margin but experience over time has taught
us that running any electronic device (passive or active) right at
the edge of its ratings will shorten its live span.

The other issue is the dreaded back EMF from the motors. If you were
to incorporate an active voltage clamp on the motor DC then you might
be able to use those caps.

I don't know if you will be using steppers or servos but its nice to
have the option to raise the DC voltage if you find your upper end
performance needs a little boost. If your caps are the 100V species
then you can easily raise it to the driver limit without a lot of
retrofit.

Given a choice I would sacrifice capacitance for breakdown voltage in
a motor application.