Universal Motor Controller

Discussion in 'General Electronics Chat' started by scubasteve_911, Oct 27, 2008.

  1. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    Hello everyone,

    I have been meaning to replace the induction motor on my CNC milling machine with something more powerful and that I can easily control the speed of. Currently, the motor spins at the typical ~3400RPM and I have to change speeds manually with 5 different pulley settings.

    I am planning to directly couple (helical coupling, or oldham) a 1/2HP universal motor that apparently runs from 115V AC/DC. Instead of my asymmetrical arrangement with a 1/4HP induction motor. The new motor will be able to spin up to 10500RPM.

    The CNC software I use, Mach3 Mill, has the ability to create enable and PWM signals for a spindle. Is there an easy way to convert these signals to drive the motor from 0-100%? I am not very familiar with working with AC circuitry and have never really used a triac before. Any suggestions on this? I only want open loop control, so I am not using encoders or anything.

    Thank you!

    Steve
     
    Last edited: Oct 27, 2008
  2. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    I just read through an appnote

    http://www.st.com/stonline/books/pdf/docs/2477.pdf

    which suggests that DC-mode control of universal motors lead to less heat and noise, which prolongs their lifespan.

    So, I would like to stick to this method. The circuits shown do not appear safe and I will not use them as is, but I would use an optocoupled version. I have a microcontroller that I can read in the PWM and convert it to anything I need, which I forgot to mention. This is because I would rather be using the raw PWM, than programming.

    I did find a pretty good way to do this safely, but it uses AC phase firing with an opto-triac. Can I modify the circuit to rectify the AC voltage? As shown in the above link?

    http://www.toshiba-components.com/m...SHIBA_AN37_PhaseAngleControlForUniversalM.PDF

    Steve
     
    Last edited: Oct 27, 2008
  3. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    I know I am posting a lot... I'm researching and posting my findings, eventhough I probably should have orchestrated my post better with a lot of research beforehand.. anyways

    I didn't realize that I could simply rectify the mains (through an isolation transformer of course!!!) and PWM a MOSFET. Anyone have any objections?

    Clearly, I am going to need to optocouple the fet drive. Is there a way I could hook up a fet driver past the optocoupler? How would I derive the voltage for the gate driver and still remain isolated?

    Steve

    [​IMG]
     
    Last edited: Oct 27, 2008
  4. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Well, grump. I was trying to get to a design on the EDN site, but can't get past internal server errors.

    The one I was trying to get a link to was a bidirectional H-bridge controller that used controlled conduction in the P channel FET's rather than PWM. The circuit senses armature current as the means of setting and maintaining motor speed.

    The circuit would need lots of tweaks, as it uses 16 volts and a 12 volts 4 watt motor. But there is the potential for load-compensating speed control for your spindle motor.

    The design was in the October 2 issue. I am going to mess with it, but eliminate the H-bridge and simply use a DPDT switch to select the direction.
     
  5. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,172
    395
    Here is a far left field solution with neither motor leads at ground . Fullwave rectify line,FET switch,opt-transistor isolator,FET driver,and gate ckt. as: transition detector feeding 8 bit sequential counter[4017] 8 outputs go to a 9 position DIP switch to 8 input OR gate, to coupler and FET driver. Switching takes place at low voltage, 8 step speed contro [ 8 because I had an 8 in gate]. I use a modified version to control dimming of 26 series lights used in flag [Bernard's Pictures] .
     
  6. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,684
    900
    Steve,

    I agree completely with your decision to use DC control. I have four tools that I have converted. One is universal motor with phase (triac) control; one is a lathe with DC motor (Baldor) and DC control (KBIC, KB drives); and two (drill press and shaper) are three-phase with VFD drives. The triac control is barely workable. The DC drive and VFD work well, and it is hard to say which is better . Both give good torque at low speed. The DC might be a bit better at the slower speeds. In addition, since you have the motor already, the DC drive should be cheaper.

    John
     
  7. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,172
    395
    I like the isolation xfmr. If you dont mind dropping 5 W minimum, a 2k resistor +_, would give 50 mA for driver which would be at risk on fault.
     
  8. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    Beenthere,

    Thanks for the response. I would be interested in that article, is it possible to get a link and I can see if it works for me?

    Bernard,

    Thanks! I admit, you totally lost me with your explanation of the circuit. I guess my text-to-schematic 'engine' in my head is fairly poor. I have access to a microcontroller, so I would prefer to do as much as possible through it. I prefer not to have to program though, which is why I really want a circuit that would work directly with the available enable and PWM signals.

    John,

    You've reassured the decision to go DC, I am definitely going down that road. I ended up getting this motor for 40$ as new old stock, which is a good deal. It's pretty much perfect for what I am doing. So, you bought the drives for these various motors? If you did a circuit, would you mind sharing?

    Steve
     
  9. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Go to electronicdesign.com and look for the IFD (idea for design) for October 2.
     
    Last edited: Oct 27, 2008
  10. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,684
    900
    I bought all of the controllers. There is a used KBIC for $9.99 (#130263625820) on the auction site today. A buy it now new for $75, and so forth. Just search on KBIC or any other model number. I don't know your motor specs, but check the KB Drives site first to be sure you get the right model number and features. I generally buy new, having been burned once with a "removed from working equipment" that had a roasted PCB trace and SCR (both easily repaired). Those things should not be problems with the low power units. One important note, you will need to know the DC spec for your motor (90V or 180V typically).

    I bought instead of made, because I had other things to do, and you can get one for only a little more than the minimum at DigiKey. Moreover, a commercial drive will come with nice features, shut down, etc. -- which the simplest homemade unit will lack -- and a nice heat sink. Buying the drive will also give you more time to play with the CNC.

    Drives Warehouse is another source for reasonably priced KB drives.

    John
     
  11. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,172
    395
    New drawing
     
  12. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    Bernard,

    that's a nifty circuit, you can probably use a rotary switch for control, which would make it more feasible.

    I have decided to take the safe route on this and try to find a drive on ebay. There are a lot of little safety features, etc, included. I will practice and restrain my creativity on the lower voltage circuits.

    I am basically in the middle of rebuilding my CNC machine. I just replaced the crappy original motor couplings with helicals, which required me to remove them from the machine and use the lathe. This reduces backlash significantly, now it only exists in the nut (which is the dual-type).

    I designed a servo controller last year, but it had some design issues. I am re-designing it with much more thought and I hope it works better. I will post the schematics, PCB layout, and code when completed. Basically, it's a full 3-axis servo control system that interfaces to the PC. It can be run directly from Mach3 Mill, which is a very low-cost G-code interpreter and works with hobby machines.

    Steve
     
  13. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,684
    900
    Steve,

    Glad you decided time has value. It's part of getting older. :D

    Your servo system sounds exciting. You might want to consider an upgrade to a 4-axis system. Please keep us updated.

    John
     
  14. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    Hi John,

    You are 100% correct about the relationship between age and the value of time.

    Originally, I did do a 4-axis system just in case I wanted to add on a rotary indexer. I am actually trying to fit the PCB on a 66$/each 4-layer special, which is 30"^2 of board space. Typically, I use the 33$/each 2-layer which gives me 60"^2. Four will not fit on the 4-layer. This board was a bit tricky to route on 2-layers, so I want to do it on 4-layers using the special. This way, others can get the PCB made the same way.

    I am also cheating with the servo controller IC. It's a premade solution from www.jrkerr.com called the PIC-SERVO. They're about 30$/each, so a bit price. All I have done is made a PCB around it and a simple application on the PC to program all of the tuning parameters. I will post my interface code, etc, in the projects section when I am finished.

    Steve
     
  15. scubasteve_911

    Thread Starter Senior Member

    Dec 27, 2007
    1,202
    1
    I may take back some of my words.. Looking into the design further, having a unidirectional control is very simple and I will implement it.

    The main reason I want to is for completeness of the controller. Having the option available will make the project more valuable. I am going to design for a maximum of 1HP. Secondly, if something breaks, I am able to fix the problem with available components. I will be posting the design when I am finished. I don't design much AC stuff, so the learning curve is a bit high for me. I am running an external isolator, so I am not including it in the circuit.

    Steve
     
Loading...