Re: [CAD_CAM_EDM_DRO] Re: Question on connector used on servos

Mariss Freimanis wrote:

>Strangely enough gold is not always best. Gold is best for contacts
>that carry 1mA or less. Because gold does not corrode or tarnish, it
>will not develop an oxide layer to interfere with conduction for very
>low energy contacts.
>
>The situation changes drastically at higher currents. Above 10mA, a
>closing contact has enough energy to develop an arc. This arc, no
>matter how small, welds the contacts at a microscopic level and
>insures a gas-tight connection.
>
>Meanwhile, the same happening with a gold contact vaporizes the 5-
>micron gold layer leaving the just the nickel underplating. This does
>not form as good of a contact as a welded tinplate one.
>
>

There is a big difference between contacts made for breaking current and
contacts made for carrying, but never breaking, the current.

The original poster asked about connectors on motors and encoders.
Yes, gold is not good for high current relays which have to make and break
live circuits. It might actually be OK if you could afford dime-sized
contacts of pure gold, but obviously these would be relays even NASA
would have trouble paying for. But silver, and other metals have proven
best for high current relays and switch contacts that are used in the 1+
amp range.

But, for CONNECTORS that are not to be used to break the circuit,
gold contacts are good up to the tens of amps. Above several tens of
amps, silver contacts are often used, especially in circuits where
substantial
voltages are available to break down any oxide that might form.
On extremely low voltage circuits, silver connectors may still be inferior,
but that is a special case. Especially for the encoder of a CNC servo
system,
where encoder failure will cause a runaway condition, the best connectors
for the purpose should be used. Even though the encoder power terminals
will draw over 10 mA, these connectors generally are not pulled apart
under load, and the damage to the contact would not generally be much even
if it were.

And, I do know what a Gecko 320 does if one of the encoder terminals goes
open-circuit! (Neat trick, my analog servo amps are driven to the wall on
an encoder or tach runaway, and they fault on overcurrent if either happens.
I didn't really design it like that, but it is a neat feature anyway.
The machine
never gets more than 1/4" before it e-stops.)

Jon