Re: [CAD_CAM_EDM_DRO] running the BDI EMC on a real machine

Keith,

I am using USD E3 encoders on ballscrew ends. They are fitted with the
little inline differential drivers. I too have had
noise problems with single ended encoders in times past and
always use differential- even with short cable runs. The drivers
are real slick. just plug them in to the encoder connector and
plug the cable in to them. They are about $12.

Ordered the card and am now trying to find cables and breakout boxes. I
don't like Mouser's price very much.
STG ought to sell cables and stuff.

After christmas dinner I had a glance at the PID calculation
portion of EMC- pid.c and pid.h. I did not see anything that
took care of integral windup, differential noise immunity,
or bumpless transfer in the code. It may be there; perhaps
I missed it courtesy of Miller brewing Co or something.
Anyway I can think about that much later... I just need to
get it hooked up right now.

I have a hard limit drawn up for my machine... that is the
motors are disconnected and partially shorted when the last switch is
opened. I have to call Copley and see if I need to
snub the motor relay contacts. Opening the motor could cause
a very high voltage pulse in the amp outputs.

I have a question for you-
Did you only use 2 of the 50 pin connectors for 3 or 4 axis
EMC? I note axis 0-3 DAC outputs and encoder inputs are
all on one connector and digital i/o on another.

Les

Leslie Watts
L M Watts Furniture
Tiger, Georgia USA
http://www.rabun.net/~leswatts/wattsfurniturewp.html

----- Original Message -----
From: "Keith Rumley" <DSCadCam@...>
To: <CAD_CAM_EDM_DRO@yahoogroups.com>
Sent: Wednesday, December 26, 2001 1:02 PM
Subject: Re: [CAD_CAM_EDM_DRO] running the BDI EMC on a real machine


> Les,
>
> I'm running a BDI 2.04 (& various small modifications) with a servo setup
> on a retrofitted Bport1 Moog. Drives/motors are Emerson FX/DX490's, and I
> use the STGII card with feedback from US Digital E5D encoders mounted on
the
> ball screw ends. I'm pleased with it's performance, no motion control
> problems.
> EMC has quirks, but not in the servo motion/trajectory use. IO and
> operator/G-code differences are the main things that require getting used
> to. Currently the STGII IO is not used beyond axis limit, axis home, and
> e-stop switches.
> I'd recommend getting the 50pin IDC cable connector breakout boards
for
> the connecting the STG to amps, etc. The flat cable was not fun to solder
to
> the 25pin serial style connectors I used.
> I also highly recommend getting the differential line driver encoders,
> I've had zero problems with them (STGII takes differential encoder input)
In
> setup I crashed an axis at close to 600ipm, and the STGII/EMC combo didn't
> lose encoder counts/position. Most problems I had/have were not with EMC
at
> all. Time from start (iron, STGII card, computer) to workable
> (retrofitted mechanics, EMC 3-axis servo control) was about 6mo.
> The block processing has a noticable delay before motion starts.
(using
> a P166 with 64mb is the main reason...) It does take some time to get the
> EMC setup parameters figured out.
> Performance-wise, the FX drives have a 3ms analog command delay. This
> has kept the IPM lower than the motors are capable of. (by requiring the
> following error values in EMC to be raised beyond acceptable accuracy.)
>
> - Keith Rumley