Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Dan13 on August 27, 2009, 02:25:35 PM
-
Hi,
I am considering to buy an AC servo motor. This one I found satisfies all my requirements (the main of which is top speed of about 8,000RPM) but one - it has a resolver. I found that it's really hard to find motors in the above speed range, so there is no much choice really. So I am considering to buy it and fit a commutating encoder in place of the resolver (the other choice is to find a servo drive that will support servo commutation, but it seems even harder).
I have two questions:
1. Has anybody done this before and is it worth the effort? Any special difficulties?
2. Any sources for reasonably priced commutating encoders? I found some at www.encoderprod.com , just called them but their prices are much more higher than I am willing to pay for an encoder. They are selling them for $250 for a single encoder!
Thanks,
Daniel
-
I have done that on a couple of motors and it was easy enough but another motor I had the resolver seemed to be fixed to the motors shaft and no way it was coming off.
The encoder sounds about right and is cheaper than the quote I got from Renco by about $100
What I ended up doing was buying a small motor with the same amount of poles and using the encoder from it, had to machine special mount plate and an extension shaft as the motors shaft was 1/2inch and the encoder was 12mm :(
The drives I use have auto sensing for the commutation but its not the best as on power up it moves a bit while it senses the commutation, fine for a spindle I suppose but not great for an axis. Then again if the aux power is kept on the drive it will not need to do this every time the main power comes on.
What size motor you after? Could you not get a bigger motor and gear it to get the speed?
Hood
-
Hood,
I need the motor for my Emco F1 spindle. The original one is a DC motor 4000RPM, 440W, 1Nm, geared down 2:1, so at the spindle I get max speed of only 2000RPM (not really enough for most of my work) and torque of 2Nm. The torque is a bit insufficient at times, but I think an AC servo with the same continuous stall torque rating would do much better, since it can ramp up to it's peak torque much faster and stay there for longer times.
Getting a bigger motor - that was what I was thinking to do before I took apart the F1's head. The timing pulley mounts on a 35mm shaft, setting the limit to how small a pulley I can use, and the head housing limits the max pulley diameter I can put on the motor - about 70mm. So gearing down 1:1.5 seems to be the best I can do. Thus using a 8,000RPM motor I should be able to get 12,000RPM at the spindle - which is what I am aiming for.
But may be you're right, I'll need to compromise on a lower speed. May be I can find a 6,000RPM and get 9,000RPM at the spindle.
Thought using a High Speed VFD spindle, but it would limit me to a lower speed of 4,000RPM (if geared down 2:1).
Daniel
-
How about http://usdigital.com/ -- much lower cost?
-
How about http://usdigital.com/ -- much lower cost?
Do they have commutation signals?
Hood
-
Wait a sec -- that needs an absolute encoder doesn't it? That's not so cheap... <slaps forehead> doh! :-[
-
Abs may be needed on certain drives but incremental is more common I would say. I would imagine Daniel will be looking for incremental but they also have to have commutation signals in addition to the A, B and I channels. It is also preferrable if they are differential for A, B and I although you could get a line driver from US Digital to convert.
He will also have to make sure that the commutation signals match the pole pairs in the motor he gets.
Hood
-
Renco are dumping commutation encoders on ebay right now, for $20.00 and $90.00 for the differential bearing model.
Essentially you are converting an AC sinusoidal - resolver motor to BLDC. I have converted a few including the proprietary Fanuc Red caps.
As mentioned, you have to find out the number of pole pairs and align the commutation up when installing.
Renco have the pole pair designation after the resolution, 1000/4 will be for an 8 pole motor.
Nosmo.
-
Essentially you are converting an AC sinusoidal - resolver motor to BLDC
What is the difference between AC Servo and BLDC? I have searched and not really come up with anything conclusive, the best I can make out they are one in the same and its the drive that actually determines it.
I know that the motors I have are certainly described by my drives setup parameters as having a sinusoidal commutation so I would presume they would be what you are calling AC sinusoidal rather than BLDC?
Hood
-
Hood,
Once read somewhere (can't recall where exactly) :
AC servo and BLDC are basically two types of motors operating on the same principle - electronic commutation. The only difference between the two is that an AC servo doesn't have magnets at all, while a BLDC has magnets on the rotor. AC servo has windings on both the rotor and stator.
Daniel
-
FWIW:
Although BLDC motors are practically identical to permanent magnet AC motors, the controller implementation is what makes them DC. While AC motors feed sinusoidal current simultaneously to each of the legs (with an equal phase distribution), DC controllers only approximate this by feeding full positive and negative current to two of the legs at a time. The major advantage of this is that both the logic controllers and battery power sources operate on DC, such as in computers and electric cars.
Vector drives are DC controllers that take the extra step of converting back to AC for the motor. The DC-to-AC conversion circuitry is usually expensive and less efficient, but they have the advantage of being able to run smoothly at very low speeds or completely stop in a position not directly aligned with a pole. Motors used with a vector drive are typically called AC motors.
RC
-
AC servo has windings on both the rotor and stator.
Daniel
Well all the motors I have seen that are described as AC Servos have magnets on the rotor, so they should be described as BLDC?
Hood
-
RC, yes that seems to confirm what I have gathered, basically its the drive that determines whether its BLDC or AC Servo, or am I reading you wrong?
Hood
-
Hood,
That's my assumption as well.
All of my AB AC Servomotors have magnets.....as far as I know anyway.
-
Yes so do mine, also the indramat, the yaskawa, the PacSci etc, I have a nice wee collection ;D
Hood
-
The BLDC and AC servo's are constructed practically identical, three star connected stator windings with P.M. rotor, the difference comes in the commutation, Brushless DC switches two windings at any one time, like a brushed motor turned inside out.
The AC is controlled with true three phase, sinusoidal in nature.
The BLDC requires three encoder pulses or hall effect devices to commutate.
BLDC pattern http://users.tinyworld.co.uk/flecc/4-pole-bldc-motor031102.swf
The AC use resolvers to commutate but some also use the hall effect for initial rotor detection and immediately switch over to encoder resolution pulses to synchronize the three phases.
Nosmo.
-
Ok so they are one in the same and its the drive that determines what they are?
Hood
-
Well not quite, the drive AND the feedback device. IOW an AC motor with resolver will not run on a BLDC drive without changing the motor resolver to equivalent hall effect device.
The Fanuc AC Red cap motors I mentioned earlier have a differential encoder and a 4 bit binary code which they use in their systems to produce a pseudo sinusoidal pattern from the 4 bit code, similar to what a resolver would produce.
Jon Elson is selling a unit where he has found a way to convert this code for BLDC commutation.
I have generally changed the whole encoder if I get the right price.
Nosmo.
-
Once read somewhere (can't recall where exactly) :
Found where I read it. See bottom of page 26 and 27.
http://www.techno-isel.com/H835/835DownLoad.asp?Page=H835_Cat
Daniel
-
Found where I read it. See bottom of page 26 and 27.
Daniel
Daniel, Whoever wrote that article is basing his information on very old technology, there has not been a wound rotor AC servo motor used for many years, for a motor to be made synchronous to the applied frequency, it requires either a permanent magnet on the rotor or energized coils of some kind.
I think the author was mixing induction motor technology with his explanation, an ordinary AC induction motor can never be a synchronous motor, without some other tricks.
Apart from dismantling the motor to prove an AC servo motor has magnets on the rotor, is to back feed the motor shaft by some means and 'scope the three phase stator leads, you will see a generated three phase, the magnitude of which is dependant on rpm.
Generation will not occur with a solid iron rotor, only by energizing rotor coils, IF it has them.
Nosmo.
-
Nosmo,
Thanks for clearing things up. Now it makes sense to me :)
Daniel
-
Well not quite, the drive AND the feedback device. IOW an AC motor with resolver will not run on a BLDC drive without changing the motor resolver to equivalent hall effect device.
The Fanuc AC Red cap motors I mentioned earlier have a differential encoder and a 4 bit binary code which they use in their systems to produce a pseudo sinusoidal pattern from the 4 bit code, similar to what a resolver would produce.
Jon Elson is selling a unit where he has found a way to convert this code for BLDC commutation.
I have generally changed the whole encoder if I get the right price.
Nosmo.
Ok so think I have it now, the motors I use have encoders with hall signals so they can be classed as either AC or BLDC servos depending on the drive I use. My drives use the hall for startup commutation so will be AC but if the drive used the Hall continuously then they would be classed as BLDC.
I also have a motor that I put an encoder on that only has A B and I channels so it can be used on my drives if I enable Auto Sensing for startup commutation, this would be classed as purely an AC Servo as BLDC drives wouldnt be able to run it.
Hood
-
this would be classed as purely an AC Servo as BLDC drives wouldnt be able to run it.
Hood
This would pretty much sum it up, there are none that I know of anyway.
Nosmo.