Re: [CAD_CAM_EDM_DRO] Re: Servo Rotating Problem

Imran Jawaid wrote:

> I tried to increase the Following error but didn't work so i swap the
> leads
> which are going into the DAC's input.

DAC's input? You must mean the DAC output going to the servo amp
input. If the servo amp input is differential, then that is OK. Otherwise,
you could do this by changing the DAC output scale from
OUTPUT_SCALE = 1.000 0.000
to
OUTPUT_SCALE = -1.000 0.000

This will invert the output polarity. Otherwise, you can keep everything
consistent by reversing the motor and tach wires.

>
> With Min_Following_error= .050" I passed the jump (voila)
>
> But the motor starts to shake voilently & heats up too much.
> tried to jog but can't see cuz the encoder count was moving too fast
> (back and forth)

You have too much gain. I assume you have the servo amp
running stably with no input, so it must be the position loop that
has too much gain.

The P term is the dominant one. You probably have the default proportional
gain set to 1000, try 100, or even less, like this :
P = 100.000
I = 0.000
D = 0.000
FF0 = 0.000
FF1 = 0.000
FF2 = 0.000

> so i swaped the encoder also(as told by jon on previous post) but
> same thing happened. so I stopped it before i cook my servo motor.

No, don't change too many things at one. If the sign is wrong, you
mostly want to swap only one sign at a time. Once you get the
position loop stable, you need to get the xplot program dowmloaded
and running, so you can log position or velocity data and plot it out
for tuning the servo loop.

> BTW i've thermal relay between my amlifier & motor. But i'm not using
> it (i.e. not giving it operating voltage but they are connected) ..
> does it have anything to do with it.

I doubt it. You need to get the loop stable. If you send me your
.ini file, I can advise.

Jon