Re: CNC interface PCB

> Subject: CNC interface PCB
>
> Hi All -
>
> I'm working on a custom interface board for my controller, and I
> have some questions regarding power capacity of PCB traces. I
> would like to use an isolated corner of the PCB for some relays
> for energizing external devices (coolant pump, air cylinders,
> etc.) as well as possibly cutting power to the Gecko's under
> estop. The worst case is the Gecko's, rated for 80VDC and 20A.
> Obviously the Gecko's are supposed to handle this on a PCB, and
> I'm wondering what it would take in terms of trace geometry to
> handle the same ratings going from screw terminals to relays and
> back on my PCB. Has anyone fried the traces on their Gecko's?

Carlos;

A .010" trace will handle that current nicely .. at zero degrees
Kelvin.

The actual figure for current as a function of trace width is related
to the AMBIENT TEMPERATURE.

I'd try to hunt down Printed Circuit Design by Ginsberg.
ISBN 0-07-023309-8 Fabulous reference.

Also, you may want to read ANSI/IPC document D-949.

For power applications, I've always used a 20 degree C rise as a VERY
conservative figure. From the charts in the book above, to keep a trace
carrying 20 amps from rising only 20 degrees C, a trace of 700 cross-
sectional mils is required. With two ounce copper plating, this is a
trace .250" wide.

If a person is willing to tolerate a 60 degree C rise, only half that is
required.

These figures are for INTERNAL traces. Strangley, external traces, as
would be found on a simple two sided PCB, require about 2.5 times the copper.

Note that 2 ounce copper is unusual, and normally must be specially
requested. Also, the above rules only apply to COPPER plating.

Alan

--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta