Kirk Wallace wrote: > On Mon, 2007-10-22 at 23:20 -0500, Jon Elson wrote: >> >>Looking at your link, it says they are sinusoidal. > > > Shoot, I thought I looked, but I guess not hard enough. > > Well hidden, I just got lucky! >> >>In theory, yes, but it would require replacing the "catch" >>diodes, power FETs, bootstrap diodes, snubber caps, main DC >>filter caps (both film and electrolytic). I think that covers >>all the parts. The FET drivers are good to 600 V. > > > That seems reasonably doable. > Until you've had one of these amps explode in your face on the bench. > > > So, with a trapezoidal drive and trapezoidal motor, how is the drive > able to control the output shaft position to fractions of a degree? it applies current until the encoder says it is "there", then it stops applying current. This is all managed by the PID loop in the PC, the drive just turns a 5 V PWM signal from the UPC into a pulse at the supplied DC voltage, and uses the Hall signals to decide which windings to apply it to. > Opps, in thinking about this question and checking the brushless amp > documentation, I may have just answered it myself. The brushed and > brushless amps function in the same way, and have the same subsystems, > but with one the activated windings are determined by carbon brushes and > copper fingers, and in the other, magnets and Hall sensors. Yup, exactly! Either drive > could not care less what the shaft position is. I viewed trapezoidal and > sinusoidal motors as slightly different variations of each other, but > they are very different Well, they really are NOT all that different. The way the ends of the winding of each phase overlap is where the difference is. . With that in mind, I am guessing, the > sinusoidal motor was invented so that a stepper-like driver which does > care about shaft position could be used. No, by tapering off the current in one winding while tapering ON the current in the next winding, they eliminate a little bump when the drive commutates to the next winding. The trapezoidal drive abruptly switches all current from one winding to the next, so it is always driving + to one motor terminal, - to another, and leaving the 3rd completely open circuit. The sinusiodal drive provides current to all windings all the time, much like a VFD does for an AC motor. So, all or most sinusoidal > drivers have step/direction inputs and trapezoidal drives PWM/direction > and the similarities in the motors is coincidence? > No, not true. MANY servo motor drives have +/- 10 V velocity command inputs, some have network connections or proprietary command languags, some have step/direction or "single pulse" which is one wire for CCW, one for CW. Others have quadrature inputs so they can be directly driven by an encoder for electronic gearing. There is NO relationship between sinusoidal vs. trapeoidal and step/dir vs. PWM. In fact, fairly few commercial servo drives accept the PWM input, I did mine this way because it was the simplest and cheapest. There is no computer in the servo amplifier at all. > If this is true, then I am way out of luck as a far as drivers/amps for > my PacSci motors. Well, you could ship me one and I could see how bad the vibration is. It might not be bad at all, I just can't predict. The only sinusoidal motor I've tried is a 22 Lb API motor, and it wasn't bad, but it would have been quite noticable on a machine. I just got a Giddings & Lewis motor which also looks to be sinusoidal, and I will try it by the weekend. It is about 10-12 Lbs, and might work fine. The tiny Pittman motors I started with also appear to have sinusoidal windings, although the motors are specified for trapezoidal drive, and there's no vibration there. The API motor may have been dropped, so all results with that one may still be suspect. The G&L turns like there's nothing inside, no drag, lumps or anything manually, so it may be in better shape.
Oh, the PacSci, you say it doesn't have an encoder, just Hall outputs. I'd have to hook an encoder to it for evaluation, if that is the case. I do have Rutex drives for my Sanyo's, but I don't > have a good feeling about support, let alone getting creative with > applications. Any one have thoughts on this drive? > > http://granitedevices.fi/index.php?id=8 > > It too is a little shy on current and voltage. > Very interesting! I was going to say it was a great deal, but I'm not sure what 219 Euros is, anymore. One thing I note, is it does not use the HAll signals for commutation. The problem there is it has no idea how to operate the motor until it has seen the index pulse. I guess maybe it tries random application of voltage until the motor moves in the desired direction until it sees the index pulse, then it knows from then on. I might ask them for additional info on that. Jon ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
