Re: MAX232 capacitors?

mrdarrett@xxxxxxxxx wrote:
Chris wrote:
mrdarrett@xxxxxxxxx wrote:
I'm thinking of building something like this:

Now, the MAX232 needs 1.0 uF tantalum capacitors.

Mouser stocks a dizzying array of Ta capacitors. Any suggestions on
which ones would suffice? (I'm looking for the cheapest ones that will
do the job.)

Now, does the MAX232 need 1.0 uF capacitors? Or will 1 uF be good


Hi, Michael. The MAX232 uses the 1uF tantalum caps as a charge pump
voltage doubler/inverter to change the +5V into a +/-10V for the
RS-232. Tantalum is used because it has low ESR at the charge pump
frequency (above 10KHz AFAIK). Any solid tantalum cap will do, cheaper
is OK. You didn't specify whether you wanted radial or axial through
hole leads, or whether you wanted SMT caps, so you don't get a
definitive answer, but you can look yourself.

I took an opportunity to look at the Mouser selecttion using their
product selector, and found that most of the items listed are
non-stock, or require a large minimum order. If you want a cheapie
onesies part, and you can live with a radial through-hole part, their
P/N 80-T350A105M25AT (1uF 25V molded radial lead, Kemet) goes for $0.31
ea. in single quantity. It will work, and you could do worse.

Make sure to watch the polarities of the caps -- they *will* explode in
quite startling fashion if you install them bass-ackwards.

Good luck

Ok, thanks for the information.

Is tantalum actually required? I later looked at a TI MAX232 spec
sheet (again from Mouser), and there's a Note 2 there: Nonpolarized
ceramic capacitors are acceptable... (page 7)

Is it ok to take this advice, or am I better off just getting the
tantalum...? (reduced noise, etc.?)



The charge pump droop is proportional to the ESR of the caps. I've used many of the x232 parts (now second sourced by TI) and ceramics work great without the problems of tants.

1uF ceramics rated at 2x the expected peak voltage should be just fine.