New Servo product pre-announcement
Posted by
Jon Elson
on 2004-01-19 20:07:12 UTC
Hello, all,
I wanted to let everyone know that I have the prototype of a new
servo system running on the bench under EMC. I modified the
FPGA firmware in my Universal Stepper Controller to generate
PWM waveforms. I then modified one of my analog servo amps
to accept a PWM input. I really had no idea how well this would
work, as it has no velocity or current feedback, as is used on the
old classic velocity servo systems. Also, it only has a position
loop update of 1000 Hz. But, in fact, it works REALLY well,
at least on the bench, and I can turn the gain up to amazing levels
without any instability. With the P,I, D and FF0, 1 and 2 coefficients
available in EMC, I can tune it to have a REALLY small following
error under all conditions. I am working with a small servo motor
from an old DecWriter printer, this is similar to some of the smaller
Electro-Craft motors.
I have to do a bit more work on this to complete the servo amp with
current limit, but it is showing a lot of promise. The servo amp is
REALLY simple, too. 4 FETS, 2 gate driver chips, 3 logic chips
and optocouplers is pretty much the whole thing.
I'm definitely going to have a demo of this at the NAMES show this
year.
This might also be of use in retrofitting some older CNC machines
which used PWM-controlled servo amps instead of analog-input amps.
I think the Bandit controls may have used amps of that type.
The PWM frequency can be set to match the amps in question.
Jon
I wanted to let everyone know that I have the prototype of a new
servo system running on the bench under EMC. I modified the
FPGA firmware in my Universal Stepper Controller to generate
PWM waveforms. I then modified one of my analog servo amps
to accept a PWM input. I really had no idea how well this would
work, as it has no velocity or current feedback, as is used on the
old classic velocity servo systems. Also, it only has a position
loop update of 1000 Hz. But, in fact, it works REALLY well,
at least on the bench, and I can turn the gain up to amazing levels
without any instability. With the P,I, D and FF0, 1 and 2 coefficients
available in EMC, I can tune it to have a REALLY small following
error under all conditions. I am working with a small servo motor
from an old DecWriter printer, this is similar to some of the smaller
Electro-Craft motors.
I have to do a bit more work on this to complete the servo amp with
current limit, but it is showing a lot of promise. The servo amp is
REALLY simple, too. 4 FETS, 2 gate driver chips, 3 logic chips
and optocouplers is pretty much the whole thing.
I'm definitely going to have a demo of this at the NAMES show this
year.
This might also be of use in retrofitting some older CNC machines
which used PWM-controlled servo amps instead of analog-input amps.
I think the Bandit controls may have used amps of that type.
The PWM frequency can be set to match the amps in question.
Jon
Discussion Thread
Jon Elson
2004-01-19 20:07:12 UTC
New Servo product pre-announcement
Kim Lux
2004-01-20 08:03:33 UTC
Rutex R990H drives: experience ?
Peter Renolds
2004-01-20 13:48:55 UTC
Re: [CAD_CAM_EDM_DRO] Rutex R990H drives: experience ?
afogassa
2004-01-20 15:40:45 UTC
Re: Rutex R990H drives: experience ?
ckhaun2000
2004-01-20 16:00:53 UTC
Re: Rutex R990H drives: experience ?
Kim Lux
2004-01-21 07:53:04 UTC
Re: [CAD_CAM_EDM_DRO] Re: Rutex R990H drives: experience ?