DC Motor Control Project

Discussion in 'The Projects Forum' started by mmazzuc1, Oct 2, 2012.

  1. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    Hi Guys,

    I am a mechanical engineering student and am trying to set up a DC motor to use for my capstone project. Unfortunately, we do not get any hands on experience at my school, so I am trying to get some guidance...

    I need to open and close a valve to about 30lb-in (480 oz-in) of torque. In order to fully open the valve, it will need about 10-15 revolutions on whatever geared shaft that is attached to my valve. I am not really concerned with any position or speed control other then the valve needs to be completely open or completely closed (480oz-in).

    Requirements:

    Closes to 480oz-in
    10-15 revolutions = Open
    Low Power Consumption
    Low Cost
    Open/Closes once a week so there is very little run time.


    I was thinking that since my requirements of 10-15 revs = open is not tight, I should be able to keep my design open loop, and just have the motor run for a certain time that will produce those resulting number of revolutions. Then I would like my motor to close and be able to shut off my motor at a certain level of torque (480lb-in). I would think that I could just have some sort of current controller that is defined by Current=Torque/Kt.

    Since the motor will not be running very much, motor degragation isnt really an issue. I would think this type of control would be good for a cheap brushed /gear head motor. However if anyone has any recommendations as to why something would work better, Im all ears.

    Does this sound like I am taking the right approach here?


    Thanks!!!
     
  2. cork_ie

    Member

    Oct 8, 2011
    348
    58
    Your idea is great but you have one slight problem - the current drawn by the motor starting up will be much higher than the running current.

    This in turn means that you will need a reasonably high current threshold to initially get the motor started, this in turn may over-tighten your valve with even higher current requirement to open it in the next cycle.

    Limit switches are of course the ideal answer, even for rotary applications. You can just attach a nut to a piece of threaded bar which will convert your rotary motion to linear.

    It may also be possible to surmount your problem with a slip clutch and a timer (which it seems you will need anyway) or to take a more sophisticated electronic route.

    Have a look at this application note and see if it would meet your needs.http://www.irf.com/technical-info/appnotes/an-1118.pdf
     
  3. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    I guess perhaps there may be away to only limit the motor current when its tightening, and some how remove the limit when the motor is running in reverse to untighten it?
     
  4. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    Couldnt I just put a resister in line with the motor to lower the max current supplied to the motor that way? Essentially I could customize my peak torque/current that way?
     
  5. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    3,788
    945
    use a positive PHYSICAL stop. The motor can tolerate some slight heating caused by a momentary stall when it reaches the stop. This assures it can open the valve again since it cannot tighten it past the point of the stop.
     
  6. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    The current to start my motor once its tighten should be less current then was used to torque the valve to 27lb-in correct? Simply because it takes less force to untighten the valve then the force to tighten it. I noticed it took about 15-18 lb-in to untighten my valve.
     
  7. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    3,788
    945
    You stated the valve would be operated approximately once a week. This increases the chances of it 'sticking' when closed. You've been given very good advice here, but ultimately YOU must decide what you want to do. If you think it will work as you have designed it, then go 'head and do it.
     
  8. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,010
    1,530
    It takes the same force to break a bolt loose as it does to tighten it, theoretically. In the real world maybe a little more force to overcome static friction. I know you have a valve, but it is still basically a bolt, the stem threads into the valve body.

    Can the valve be changed to a different type? A "ball valve" does take the same force both ways, open and closed. And with a ball valve the driving mechanism could then be a crank arm and rod on a gear motor.
     
  9. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
  10. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,010
    1,530
    But there is static friction between the thread of shaft and thread that shaft moves in.
    Haven't seen that type of valve. What does the body of the valve look like?

    Does this have to be an electronic solution? Another way to do it would be to use an adjustable type slipper clutch. By that I mean the type that is a positive drive in one direction(opening) and slips when it reaches the correct torque in the tightening (closing) direction. Here is a link to the type of clutch - http://en.wikipedia.org/wiki/Slipper_clutch
     
    Last edited: Oct 6, 2012
  11. Sensacell

    Well-Known Member

    Jun 19, 2012
    1,130
    266
    This is a very tricky problem, more difficult than it would seem at first
    Here is why you never see electric valves designed this way.

    The big problem you will encounter is that the mechanical mass of the motor armature stores significant energy when running at speed, when the motor runs the valve closed, into a hard stall, all that stored rotational energy will tighten the valve to a torque level that will be larger than the motors stall output torque- it will never open again, without external assistance, that is.

    Friction clutches would only work if they had substantially different torque levels in different directions. (difficult/complex)

    If you installed some position sensors you might be able to mitigate the problem by slowing the motor way down (velocity servo) just before the valve seats, this way you could diminish the role that inertia plays and arrive at a torque level based primarily on the armature current.
     
  12. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    I understand what you are saying about the rotational inertia. However this valve can be opened very slowly to accomodate any of those effects. I only need to turn the valve 10-15 times. my torqued values can very slightly as well. 27lb-in was what I am using as a minimum. Realistically, I could tighten it to 27-37lb-in (perhaps more) and be fine. This open and shut valve isn't used for any high precision dependent applications.

    The slip clutch's seem like they might be able to work for me. I could implement one on a sprocket or a shaft to shaft. Cost is more of an issue then any size limitations. I was looking at some shaft/shaft applications with adjustability up to 30-50 in-lb. I was getting in the price range of 200$ and up. Seems like there has got to be a cheaper alternative.
     
  13. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    Motors draw significant torque at startup, and this not purely because whatever load is on them. Even with zero load, there will be a significant current surge, because at rest, the current path through the motor is basically a direct short. The larger the load, the more this problem is exaggerrated
    What way are you referring to? Are you referring broadly to valves controlled by motors or specifically to one of the previous described control schemes?
    Motor operated valves are very common.
     
  14. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    Im not too knowledgeable with the reliability of servo motors in an application like this... But if I closed the lid, and set that as my closed servo position. would I be able to open the lid and close it using a servo motor? Mainly I am not sure how the position control works under loaded conditions.
     
  15. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    So, you're going to be building some apparatus on top of this valve to spin it, correct?

    The unmodified valve (catering to operation by humans) uses a screw to achieve linear motion via rotary motion.

    Trying to design a system that mimics the human hand and accommodates the rotary motion is reinventing the wheel IMO.

    There is a prepackaged solution available, the linear actuator. The linear actuator gives the valve what it really wants - linear motion. It has all of this designed inside; the rotary motion tansferred to linear motion, the adjustable end stops, the slip clutch, everything.

    All you have to do is build some kind of apparatus on top of the valve to hold it, which you were going to have to do any way. Oh, and remove that screw that came with the valve - maybe replace it with a smooth guide rod.
     
  16. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    There are some good reasons to use linear and rotational actuation as we have already considered. However, for what we are trying to accomplish, we are planning to do a rotational style.

    Any thoughts on using a servo motor though? Would there be any limitations?
     
  17. Sensacell

    Well-Known Member

    Jun 19, 2012
    1,130
    266
    Valves where a DC motor drives a valve into a hard seat, dealing with this stick/slip breakaway situation.

    I like the idea of running the motor really slowly, then you could make the current control method work.

    Use a tach signal or armature feedback to create a velocity servo to control the speed, run it really slowly with two different current level thresholds for opening / closing.
     
  18. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0

    Sensacell,

    If this method should work. What would be the tradeoffs using a motor controlled as meantioned above vs. low RPM servomotor or low RPM stepper motor?
     
  19. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    A servo is essentially just a (ac or dc) motor with some kind of position feedback, usually with a high gear reduction. hence, the term servo is applied to a lot of things, but I think what most people would associate the word servo with, are the servos in R/C cars and trucks - is that what you're talking about? those type of servos don't make multiple revolutions that I'm aware of.

    I think a larger, multiple turn servo would definitely do the job, but will probably require you to make or buy a circuit to count the pulses and stop the motor once the correct amount of revolutions is reached. The R/C servos have this circuitry built in, but others do not.
     
  20. mmazzuc1

    Thread Starter New Member

    Oct 2, 2012
    11
    0
    Yeah, I was referring to the mult-irotation servomotors.. I would think a stepper motor might work just fine too. The CNC ones have a very high torque like 425 oz-in.

    If I was to do this via current / armature voltage, would I be able to do this with some off the shelf parts? I am not really looking to get involved with any circuit design, as I am a Mech Engineering student. However, I do have an arduino uno if I need one.
     
Loading...