Re: [CAD_CAM_EDM_DRO] Encoder options

Doug Harrison wrote:

> The Gecko G320/340 servo amps are wonderful to work with. One problem many of us have encountered is the limitation on cable length imposed by single ended encoders.
>
> Automation Direct sells a nice encoder with totem pole output that will run on 10-30 volts. It seems that the higher voltage/current figures of this encoder will make it less succeptable to noise.
>
> If I drive these encoders with a 24VDC supply and use a voltage divider to get the 5V required for the G340, can I expect better noise rejection than possible with USD or similar low voltage/current encoders? The size of my router table will require cables up to 15 feet long.

Not necessarily. It depends on how the noise is getting into the system.
There is no reason to be absolutely certain that the noise is getting into
the encoder signals. it could be getting into the logic from the power
supply lines, the motor wires or the encoder power lines. Powering the
encoders with a separate power supply might help, but that power source
still needs to be connected to the Gecko's ground. The only way to
avoid that is to opto-couple the encoder signals to the Gecko, which
might be a good idea. It is possible that the high voltage signal, when
divided down will reduce the noise at the Gecko, but it can't be assumed
to work until tested.

A better approach would be to put a differential driver on a tiny PC board
right at the encoder, and use a differential receiver at the Gecko.
The AM26LS31 and 32 devices were the original chips for this purpose,
but there are now some chips like the 75ALS191 series that are better.

Jon