Parallel L298N Driver circuit help

Discussion in 'General Electronics Chat' started by Sidco7, Sep 10, 2006.

  1. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    Hello all. New to this board.

    I am in the process of fabricating a 'parallel L298N' driver board which uses the L297 controller. I have built and tested a single L297/L298N 3-axis driver board and it works perfectly. Requiring more current capacity, I have tried to copy the single and twin the L298N and have taken care to parallel channel 1 with 4 and channel 2 with 3. For some reason the parallel board doesn't work and I have almost exhausted my trouble shooting with it. I am nearing the edge.
    I was hoping someone has tried this already and might share some insight and experiences with doing it.
    Any help would be appreciated.
    Sid
     
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    Looking at the data sheet, it appears that the L298 outputs can be paralleled. Can you post a schematic of your circuit so we can look it over? It's very hard to make anything but guesses without a schematic.
     
  3. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    What is the best way to post a diagram? JPG?
     
  4. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,274
    1,065
    jpg or gif is usually fine.
     
  5. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    It's a huge board and I don't have any kind of fancy program to simplify the circuit. All I have is an autocad drawing so I will post a PDF file. JPG or GIF will not get detailed enough.
    Let me play with that a while. Hope it works.
     
  6. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    File size is too big. 5+Meg If you send me your email address I will send the pdf file directly to you.

    Thanks,
    Sid
     
  7. Dave

    Retired Moderator

    Nov 17, 2003
    6,960
    143
    I would recommend you use the snapshot tool in Adobe Reader to select the items you wish to upload, and paste the items into paint. From there you can save them as a JPEG and upload them from the manage attachments feature when posting a reply (look at the Additional Options panel when posting a reply).

    Dave
     
  8. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    Hi, OK I've sketched out the circuit in a simplified style and hopefully it will post alright.

    thanks.
    Sid
     
  9. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    Your schematic came through very well.

    The only discrepancy I can spot is the 10K resistor in series with the sense line. The L298 example circuit has the sense line tied to ground through the .5 ohm resistor, but does not show a 10K series resistor with a capacitor to ground. The filter action is going to have a lot of effect on the signal back to the L297 driver. I'd remove the cap and short around the 10K resistor and see what happens.

    If that helps, you might also increase the 100uF cap on the Vs line. The example calls for 470 mikes, and with a .01 high-frequency bypass in parallel with it. This could also be significant.
     
  10. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    OK. Pulled the 10K resistors on the sense lines and put in a jumper. Pulled the C680p caps as well.

    Still no-go
     
  11. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    Sorry to hear that. When you have the system powered up, do the stepper motors hold in place? And when you remove the mechanical load, do they turn then? Can you drop the oscillator frequency much lower to make sure the steppers can follow the impulses - a loaded stepper usually cannot run at full speed with a big load applied. They have to be started slowly, and the oscilator ramped up to full speed in order for the motor to function properly.

    Most importantly, do you have test equipment? An oscilloscope could really be useful to visualize just what is going/not going on. A meter is also helpful to insure that Vs and Vcc are proper, and going to the right pins on the 297 & 298.

    We get to the white roosters in the next step....
     
  12. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    Hi, and by the way, thanks for your interest in this.

    No, there is no holding power to the motors. They turn freely when I power up the board. I do not have an oscilloscope but I do have a couple of meters.

    I'm wondering about my power supply. Its rated at 30V 2A DC output, but oddly enough, it seems to have an AC output value of 65V when I turn the meter to the AC side. Is this normal? Not that I think that this is the problem with the board though, because I've tried a different PS and it still doesn't work. Not unless the AC voltage zapped the L298's the first time I powered up. Is there a easy way to test the L298's with the meter?

    Also, as was noted before, the 470uF cap between pin 8 and pin 4, should I try and get that in somehow? Would I need one for each chip?

    I've built the basic circuit before (with single L298's) and it works good. I just can't understand why this one is giving me a hard time.

    Thanks,
    Sid
     
  13. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    Don't sweat the 470 mikes to pin 8. 8 is ground, so running the cap from 4 to ground anywhere will do fine.

    No hold on the steppers with power applied doesn't sound good. Neither does the 65 VAC on your 30 volt DC line. DC supplies shouldn't have more than a few tens of millivolts of AC running on them. The AC will tend to eat your electrolytics and possibly the ic's.

    If you are having to run the 298's in parallel, it is usually because you need more current to the steppers. Are you exceeding the 2 A capability of the supplies? The source doesn't need to be fancy - a rectifier and big capacitor filter is usually enough for steppers, as long as you limit current to the rating on the steppers.

    I'd want to get the 30 volt supply checked out pretty well before trying anything else.

    I've never used these ic's before, so I have no feel for how they should look to a meter. Does either one get hot with power applied? That's usually a thing to worry about. From the look of things, the 298 is doing some degree of current limiting. It may be able to run the steppers with a lower voltage - more like 24 volts if you have an alternative voltage source that is known good.

    If you clean up the source voltage and still get no result, try swapping in a spare 298.
     
  14. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    The L298's can handle a lot of heat. They have a large heatsink built in and with extra heatsink added, will throw off a lot more. There is a limit to them though and yes, that's why I am trying the parallel route. My motors are rated as 2.1V 3.42A 0.64ohm (Nema 34)

    Is there a simple power supply circuit posted here somewhere that I might be able to try? I havn't had time to browse around yet.
     
  15. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    Need some more info. The data sheet on the L298 is not real informative. It gives a 50 volt maximum on the Vs, but no minimum. It make it hard to recommend a solution. Since your steppers only need 2.1 volts, a lower Vs doesn't sound like a problem.

    Do you use the same supply for all three axes? If each motor wants 3.4 amps, then you need a supply rated at above 10.2 amps. Worse, if the controller can half step, you have two windings energized at once, which pushes the current up to 20.4 amps. That's pretty big.

    Even the transformer, rectifier and filter solution gets difficult - finding a transformer that size is somewhat hard to do. Plus the cost of copper has pushed prices through the roof.

    If the system will run on lower voltages, something like a battery eliminator might be the way to go. It's just a brute-force power supply that puts out 13.8 volts DC. Just get one that supplies enough current, which shouldn't be a problem with a battery elininator. There used to be a zillion of them on the market, and they were pretty cheap, too.
     
  16. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    The application notes really don't get into too much detail with the parallel scheme. I've run my other (single) system with a 12V power supply and it works fine. (although a little sluggish)

    I've seen lots of L297/L298 (single) circuits and some have that 10K resistor in the sense lines and some don't.

    If there doesn't seem to be any problem with my circuit dia that anyone can find, then I will have to assume it's either a chip problem or some other silly thing like a short on the board (like that's never happened before!)

    How can one test a capacitor? Is there an easy way to see if they are working? All I have is a digital multi-tester and an analog tester.

    Anyway, I think it's back to the drawing board on this one for a while.
    Thanks for your interest.
     
  17. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    I read somewhere that when I wire my 8-wire motor in series, that I should divide the rated current in two. That would put my motor currents at 1.71A.

    Now my next question, Lets say I have 3 motors to run, do I need to multiply the motor current by 3 to get the power supply Amperage. 3 x 1.71 = 5.13.

    Having said that, I then would have to always have all three motors connected on the board while the power is on to prevent overloading the other chips. Lets say I pull one motor, the chips will then have 1/3 more current to deal with. Is this correct?

    What if I have one motor with a higher or lower current rating than the other two? How would that effect the system?

    Thanks
     
  18. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    I don't quite see how to connect a stepper in series. A stepper motor will always have at least one winding energized, so the current should always be at the rating. With three motors on, the supply will have to source three times the rating of a single motor.

    Motors with different current draws will add to or take away from the total current needed. The interesting thing is if the motors can be half-stepped by the controller. This will have two windings energized, and double the current draw. Whatever the controller can do, it would be best to be able to have some headroom with respect to current - a couple of extra amps capability means your source will not get too hot.

    That said, the spec sheet says diddly about the operation of the L298. I'm familiar with Slo-Syn drivers, so I'm guessing about the L298. However, any stepper motor has to have some current through a winding in order to hold at the last position. Otherwise it freewheels, and is pretty worthless as a positioner.

    Checking caps without a purpose-biuilt meter is iffy. For electrolytics, you can always charge one up and put your meter across the terminals and observe a slow discharge. Comparing with a similar one give some idea of what shape the cap is in. If the vent is popped, or the can got taller (rose above the sleeve), then just assume the worst and pitch it.
     
  19. Sidco7

    Thread Starter Member

    Sep 10, 2006
    18
    0
    Wiring an 8 wire stepper motor in series is simply connecting the center taps of the windings (see attached page 8 fig 6)

    I have some surplus transformers that I can use, is there a way I can measure the voltage and current of them? How do I go about that? Some of them don't have markings of any kind on them.
     
  20. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Hi,

    Checking old transformer voltages is no problem - just use your meter. Current is another thing. Without some mfr's specs, it's hard to do more than guess. Anything that can give you the current you need is going to be pretty hefty, though. Probably 6 to 8 pounds.

    One thing that is interesting is how the L298 controls current through the stepper. With Slo-Syn, the current was controlled by a set of transistors. You just came up with a limiting resistor to holr the current to the stepper spec, and all was well. STmicro doesn't even speak about this.
     
Loading...