On Friday 02 May 2008, John Kasunich wrote: >Gene Heskett wrote: >> On Thursday 01 November 2007, Kirk Wallace wrote: >>>> Kirk Wallace wrote: >>>>> What is the best way to drive a 4hp, 90VDC, 40A spindle motor on a >>>>> Hardinge CHNC? Some of the large DC motors I have seen, look like >>>>> universal motors which would operate on AC or DC. Does anyone know if >>>>> the CHNC spindle motor could work on AC? I'm thinking, a few dimmer >>>>> switches in parallel and a stepper to turn the dials ;). I wonder if it >>>>> would be cheaper to go with a 3-phase motor and VFD. Thanks. >>>> >>>> No, these motors are NOT universal motors. You could rectify >>>> the mains to drive the motor, but then it would draw 40 A from >>>> the line. You should be able to find an SCR drive for this. >>>> You can't parallel dimmers, they can't be trusted to give the >>>> same pulse width. >>> >>> I was joking about the dimmers. So far I have found this: >>> >>> http://www.baldor.com/products/detail.asp?1=1&page=1&catalogonly=1&catalo >>>g=B >>> C155&product=DC+Controls&family=One+Way%7Cvw%5FDCControls%5FOneWay&voltag >>>e=1 80 >>> >>> but this puts out 180V and I need 90V. Plus it seems that this kind of >>> controller does not interface with a computer well at all. It's >>> beginning to look like DC lathe spindle drives are a specialty item, >>> which means, hard to find and big bucks to buy or repair. >>> >>>> If it is a standard-frame motor, then you >>>> could go the 3-phase and VFD route. That may not give you the >>>> range of speeds and torque required. >>> >>> The newer VFD's are much better with this aren't they? >>> >>>> If you want smooth, >>>> controlled reversing, then you need a DC servo drive. If you >>>> don't need a complete 4-quadrant servo drive, then the SCR DC >>>> motor drives may be the way to go. Baldor and plenty of others >>>> make these, they show up on eBay all the time. >>>> >>>> Jon >> >> Hey guys, even the lowly board out of a Harbor Freight 47xx8 micromill, >> except for the current rating, makes a truly excellent vsc when combined >> with a PMDX-106. >> >> This board, with a much larger bridge rectifier replacing its puny 4 amp >> device, and about 6 of the mosfets it uses in parallel with smallish >> current sharing R's in series with the src's of each of the devices could >> probably do that just fine. The speed control is very stiff with this >> unit, so stiff that if I'm doing something heavy, I have to rig an ammeter >> in series with the motor in order to see how close I am to the 'red line'. >> Otherwise it just blows the fuse with no detectable motor slowdown first. >> >> The only problem that I could foresee might be related to this boards >> ability to drive that much gate capacitance of the paralleled mosfets and >> still achieve good switching speeds. It is something I have not played >> with personally, so one might approach this by adding one device at a time >> & watching the heating. Running correctly, the mosfets shouldn't heat >> more than 10F if screwed to a good heat sink and delivering 50% of their >> rated current. In my own case, that would be at least 10 amps, but I >> didn't yet replace the puny bridge rectifier either. Mine is currently in >> a closed plastic box with the PMDX-106 and has run that way for half a day >> at a time without the box getting noticeably warm. The direction relays >> make as much heat as the rest of the circuitry. > >I would NOT attempt to beef up a MOSFET H-bridge to run a 4HP 40A DC >motor, unless you are fairly experienced with power electronics. There >are lots of things that you can ignore at 4A but will bite you at 40A. > >If this was 180V DC instead of 90, I would go farther and say flat-out >that an SCR based drive would make more sense than an H-bridge, but the >low voltage changes things just a little. > >Some of the issues that need to be considered to decide between an SCR >based drive and a MOSFET H-bridge: > >Torque ripple - H-bridge will be best, a three phase SCR drive will be >good, a single phase SCR drive will produce some ripple. Whether that >will affect your part finish or cause vibration will depend on the motor >and the machine. > >Motor insulation life: The H-bridge produces fast-rising voltage edges >which may or may not stress the motor winding insulation. Older motors >are more likely to have problems. On the other hand, 90V is so low (I'm >used to 480V stuff) that it is probably a non-issue. > >Input issues: Kirk didn't say what his available line voltage is. Four >HP is a LOT for 120V single phase, regardless of the drive technology. >On 240V single phase its not so bad. Three phase is ideal of course. >SCR drives do not have large internal capacitors that need to be charged >on startup - H-bridge drives do. In fact the dealing with the caps and >their inrush current is one of more difficult challenges if you scale up >a small H-bridge drive. > >Ken Lerman wrote: > > Someone on some list suggested that DC motors have much better > > torque/horsepower curves than a corresponding 3ph motor with a > > VFD. That's another reason to stick with the DC motor. > >Agreed 110%. If the original motor was designed for variable speed >spindle duty (which seems the case), it will outperform just about any >standard AC motor plus VFD combination. > >A lot depends on how much time and effort you want to put into it, and >how much performance you want to get out of it. That original motor is >probably a gem, and with the _right_ drive can deliver excellent speed >torque performance as well as low vibration at lots of other nice stuff. > But driving the DC motor will be a bit of a project. SCR drives are >robust from a power electronics perspective but require complex control. > H-bridges are simple to control but require better power electronics >design. Either way, you need to understand the motor and how Hardinge >used it - I bet they are doing field weakening to extend the >constant-power region of the speed-torque curve. > >The AC motor plus VFD approach is easier, as long as the AC motor is >mechanically compatible with the lathe. But you won't get the same >performance, especially at low speeds. > >Regards, > >John Kasunich > Sorry about the late reply here John, but I've been out of pocket for the weekend. 440 miles out of pocket TBE. I'd had a spell of very slow pulse last Wed afternoon, & they are going to do a stress test on me this Wed, the 7nth.
The controller in question in this case isn't an h-bridge, its simply a full wave bridge rectifier that has as a DC load, a mosfet in series with the motor, and the motor has a free-wheeling diode for when the mosfet is turned off. This is considerably simpler than an h-bridge. Scaling such as that up to a 4hp motor is probably not possible on the normal 127 volt input line as that would scale the current up to what 60 amps? Correct me there as I have done the lookups & math. But its current 4 amp bridge, 7 amp or so flywheel diode and 15 amp mosfet could surely be scaled up to a 20 amp bridge, with several of those free-wheeling diodes with some .02 ohm sharing r's in series with each one, and a pair of the mosfets if there is not one large enough, could be done to realize a 15 to 20 amp controller that could handle maybe a 2 horse motor? When I was referring to paralleling the mosfets, there are 2 considerations that I'm aware of, first being the current sharing under the on conditions, enforceable by very small resistances in series with the srcs of each device, and the total capacitance of the combined gates causing a slew rate limit of the rise & fall times of the currents, which in turn jack up the 'switching' losses. And its switching losses of that type, where the device is operating in its linear region as opposed to being full on or fully off, that are murder on mosfets. I once tried to use 2 very heavy rated such devices in an analog, linear circuit, and found their life was measured in months, this for devices rated at 250 watts, 20 amps & 1KV, actually handling an average of 125 milliamps at about 75 volts across them & generating no discernable heat while bolted to a pound of copper heat sink each. Back here at the ranch though, going up to a 254 volt single phase (commonly called 220 power) doubles the power available, and probably doesn't need exceptionally higher rated mosfet voltages because most of the ones I've looked up are available in 800 to 1000 volt ratings also. You mentioned capacitance, and I'm puzzled as to where, other than un-avoidable circuit strays, these might have to be figured into the overall picture. Can you point out something I've missed? >------------------------------------------------------------------------- >This SF.net email is sponsored by the 2008 JavaOne(SM) Conference >Don't miss this year's exciting event. There's still time to save $100. >Use priority code J8TL2D2. >http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javao >ne _______________________________________________ >Emc-users mailing list >[email protected] >https://lists.sourceforge.net/lists/listinfo/emc-users -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Support your right to bare arms! -- A message from the National Short-Sleeved Shirt Association ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
