The gecko 320x is apparently a pretty bulletproof and popular option for DC servos
What is available for AC servos that is similarly bulletproof and has lots of support?
I have some AC servos that I’d like to experiment with.
Thanks
Duke
Printable View
The gecko 320x is apparently a pretty bulletproof and popular option for DC servos
What is available for AC servos that is similarly bulletproof and has lots of support?
I have some AC servos that I’d like to experiment with.
Thanks
Duke
Usually easiest to buy a servo and driver as a set rather than trying to match a driver to an existing servo.
Certainly you can find a driver for an existing servo but it can be a pain.
G320X is a primitive and limited drive. Nothing like a modern AC servo drive.
Need more details than "some AC servos"
You can not compare a Gecko Drive with an Ac servoQuote:
Originally Posted by jduke
Not all are created equal what are the AC servos motors you have, you normally have a servo drive that is a match for the servo motor, there are many things to consider when going with a true Ac Servo system
Quote:
Originally Posted by mactec54
I have 2 Mitsubishi HA40CS ac servos that are apparently 2000rpm 750w to1Kw motors
and one HA80CS thats probably a 1.5kw to 2kw motor.
these are from a dyna mechtronics dm4400m with mitsubishi melda control.
apparently the Melda doesn't have a great deal of documentation currently available since its 30 year old technology.
what is it that makes servos such a weird beast that you have to Match the drive to the motor.
I understand that they are essentially a three phase motor with an encoder on the back and that feedback loop from the encoder is what tells the drive/amplifier where the motor is.
what is such witchcraft about this that you have to have a matched driver/amplifier to make it work?
I can understand, if you have to have a specific encoder to be able to get the right input from the motor, but If your willing and able to change the encoder, why can't you get a matched encoder and drive/amplifier and put that on an AC servo to get a working motor?
Is there enough difference in the windings on these that they have to have a matched driver/amplifier to make them work?
I am probably going to use newer AC servos to update the MIll once I get the spindle sorted out and make sure the project is worth doing, but I am interested in the why as well as the how Hence my questions about what the witchcraft is in old Servos.
Thank you for your patience and answers as well.
Duke
Being that old, unless you have the drives, they would not be worth messing with, no witchcraft involved, technology has advanced, but even 30 years ago manufactures had matched servo motors to drives, they would also be using Analog +/- 10v for the control so you would only have this option when get a new Machine ControlQuote:
Originally Posted by jduke
AC Servo motors are nothing like a regular AC 3 Phase motor, so not sure where you get that false information from :)
There are Dumb Servo Drives that you may be able to use, depending on the Encoder output you have, but the setup would not be worth the trouble, with something this old, when you can buy plug and play AC Servos like these from DMM with up to 20Bit Encoders, DMM | AC SERVO DRIVE | AC SERVO MOTOR | ROTARY ENCODER
There are several reasons, servo systems such as Fanuc, Mitsubishi, have proprietary methods of closing the servo loop.Quote:
Originally Posted by jduke
Using systems such as Mach etc, the controller does not close the loop, so you are essentially committed to using drives that close the loop, and preferably are matched.
But where the servo loop is closed in the controller such as Dynamotion/Kanlog, Galil, Linux etc, then simple non-intelligent transconductance (torque mode) drives are easier to mix and match.
Also where the controller closes the loop, multi-servo electronic gearing and Cam can be done.
For example for many years, Galil re-labled the transconductance A-M-C drives for their products.
I picked them up on ebay many times, also the likes of Copley and Aerotech drive products to mix and match very easily.
.
Between Galil (local to me) and AMC, the AMC are a little easier to work with, per the applications engineers at the motion company I used to do motor design for.
Servos are motors with position feedback, however what industry usually accepts as a servo in precision motion is low cogging, low ripple torque, and precise enough alignment of the encoder to get consistent behavior full circle. That said, any motor of any phase count that has 4 quadrant control and position feedback (not even absolute!) is considered a servo. There are plenty of garbage motors that are considered servos just by way of an encoder being bolted to the end.
Consider Odrive for a controller, it seems to be a solid basic servo drive and I know a number of people are using it in precision ag projects with bike wheel hub motors with great reviews.
As for the magic sauce for pairing controllers and motors, it's usually tuning, filter sizing, and sizing switches for the motor inductance. Servos are often very low phase inductance [EDIT: Not induction motors, motors with low phase inductance] motors which adds a switching challenge, ironic since it makes torque ripple worse, but switching nowadays is fast and cheap. A paired controller will have presets for the motor tuning and filters sized for the particular harmonics of the motor EMF. It will also switch fast enough on a non-harmonic frequency to avoid ripple and noise injection. Contrary to popular belief they are not usually co-developed, that matched controller is usually just an available product that is adjusted to "match" the servo. Aside from the analog filters any good drive can be tuned to do the same matching, it's just more work.