Re: [CAD_CAM_EDM_DRO] Bridgett update

Tim Goldstein wrote:

> First a warning, this is about big iron and fast rapids so if that is
> offensive to you or generally intimidating please skip reading this post.
>
> I just bolted a big old e-bay special NEMA 42 brushed servo motor onto the X
> axis of Bridgett (Bridgeport series II). Connected it to a Gecko G320 and
> used a 62 VDC power supply. Rapids are rock solid at 150 ipm!! I know, many
> of you are tired of hearing about the Gecko drives, but every time I hook
> one of the step and direction servo units to a machine I am blown away with
> the performance improvement from a stepper. All I can say is try it, you'll
> like it.
>
> An interesting story on the E-bay servo motors. Then came with some real
> nice 500 line encoders with ball bearing shafts and one of those real slick
> factory mounts to the servos using a helical coupler. When I hooked them to
> a scope they seemed to be working fine. I then hooked the motors to the G320
> and fired it up and the servos would stabilize and then jump suddenly then
> stabilize again. Motion was the same way, smooth, then some jumps and
> lurches and then smooth. I pulled the encoders and used some duct tape to
> attach one of the Electronic Goldmine mini-servo motors with 1000 line
> encoder to the shaft where the original encoder was. Wired it up and turned
> it on. The servo stabilized just fine (considering how loose the connection
> was to the encoder) and motion was very smooth. Diagnosis was that the
> encoders that looked good on the scope have a loose connection somewhere in
> the electronics. I then ordered a set of US Digital E2 encoders with a 250
> line resolution and they bolted on very nicely with just a tiny bit of
> handiwork on the holes of the mounting ears. Fired up the drive and it is
> stable as can be and performing great. The encoders cost me as much as the
> servo motors, but for $90 including the encoder I don't know where I could
> have gotten a servo suitable for a series II any cheaper (or get another one
> for that matter).

More likely it is an electrical interference problem, from the PWM output pulses
of the Gecko drive getting into the encoder inputs. Probably a carefully selected
pullup resistor from +5 V to the encoder signals will clean this up. But, you want
to make sure you really have it cleaned up, as you will have the same sort of
misery as the people who are losing step pulses at random. I suspect the
encoders that came with the motors have open-collector outputs, and need a
fairly stiff pull-up to avoid any noise pulses getting accepted by the Gecko drive.
Anyway, I hate to see you having to REDUCE resolution by 2:1.

Jon