Shop OBEX P1 Docs P2 Docs Learn Events
Motor question — Parallax Forums

Motor question

UghaUgha Posts: 543
edited 2008-12-21 03:18 in General Discussion
I have a·few questions about normal DC motors (especially the cheap ones).

First, which is more likely to reduce the life of a motor, continous usage with low or no load for a long duration,
or short usage sessions under heavy load?

How bad is stall (load to high for it to turn) for a motor?

The motor I have will have its life cut to 1/4th normal if 5 volts are used instead of 3.
Can PWM be used to "make up" for having too high of a supply for a motor and extend its life?

Sorry for the newbie questions, I've never worked much with normal motors before.
«1

Comments

  • LeonLeon Posts: 7,620
    edited 2008-12-13 15:53
    Most of the above should be covered in the spec. for the motor. With PWM you could check the expected duty cycle, and estimate the average voltage. For instance, if the duty cycle is 50% most of the time you could probably use twice the voltage.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle

    Post Edited (Leon) : 12/13/2008 4:01:03 PM GMT
  • Carl HayesCarl Hayes Posts: 841
    edited 2008-12-14 06:06
    There are many types of "normal" DC motors. Except for brushless ones, which are really AC motors anyway, they are series-wound, shunt-wound, and compound. Fortunately, your answers won't vary much for these types, except as I note below.

    Two things wear in a motor: The bearings and the commutator/brush system. Each of these wears when the motor turns, but the wear on either doesn't particularly depend on the load. So running it continuously, loaded or unloaded, will wear it out faster than running it intermittently, loaded or unloaded.

    Series motors, and (generally) only series, tend to overspeed when run unloaded. This will cause accelerated wear. But if you control the speed somehow, then series motors will wear no faster than shunt or compound motors. If your motor would have its life reduced by a modest overvoltage, it is probably a series motor, which would run hot, and overspeed, with overvoltage.

    Shunt motors will conduct large currents through the armatures when stalled, or nearly stalled -- so the brushes and commutator will wear a little (not much) faster when very heavily loaded but still turning. At stall, these motors will run hot -- perhaps hot enough to melt the armature connections. This would be undesirable. But some motors are designed for service at or near stall.

    Bottom line: Run them intermittently, run them only with load, and don't let them stall or overspeed. They'll last a long time.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • UghaUgha Posts: 543
    edited 2008-12-14 12:59
    Great info... thanks!

    One more stall-related question...

    If the duration of stall is very brief (a half second to two seconds) would that be a big enough impact to run hot and increase
    wear?

    After the stall, no operation will take place for several seconds to several minutes, then followed by normal operation.
  • LeonLeon Posts: 7,620
    edited 2008-12-14 16:30
    You could detect the stall condition and do something about it before any damage occurs.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • UghaUgha Posts: 543
    edited 2008-12-14 16:58
    Detect it how?
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2008-12-14 17:59
    Cheap motors are really not intended to last. They usually have cheaper bearings, cheaper brushes, and cheaper cases. And so they don't often come with any specs to help you out.

    If you have something that is running 100% of the time 24/7, it requires a 'full duty cycle motor' which just costs much more. If you are reversing direction on a DC motor, the cheap motors may not have good enough brushes for it. They tend to adapt themselves to one direction and that is all. If you are burning out motors due to stall, you may just be using too small of a motor in the first case.

    Regarding series wound, shut wound, and so forth...
    With the advent of neodium magnets, such configurations are more common in larger, more powerful motors. They are trivial to most minature applications as permanent magnets are used. Most of the wear is in the brushes. The shaft and bearings are usually quite durible.

    At 5 volts, I suspect that all the 'overvoltage wear' is in the brushes. You may need to look for a better, more rugged motor in terms of brushes. The copper enamel wire that is used for either a 3 volt or 5 volt motor more than likely has insulation that can easily withstand 50 volts or more. It is just easier to product more than adequate insulation on wire than it is to taylor insulation to small differences.

    And a good motor will have cooling. If the motor offers no cooling, it is meant for only occasional use or a toy.

    One alternative if you want good motors for a cheap price is to get a 12volt or higher motor. With the exception of drive motors removed from CD/DVD players, it is hard to locate quality 5 volt motors these days. You might find replacement tape cassette motors in some instances, but they are getting rare.

    If you really need to overcome stall, a geared motor is a much better option. No mention was made of what RPM you are using, but the faster an electric motor turns, the cooler [noparse][[/noparse]due to internal cooling] and smoother it usually can run. The brushes are happier too.

    There is an exception. Stepper motors can really provide a lot of torque at a snail's pace. If you look inside a 3.5" floppy disk drive, there is a little stepper that I suspect you could run at 5 volts and at a crawl. Sorry no specs.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    It's sunny and warm here. It is always sunny and warm here.... (unless a typhoon blows through).

    Tropically, G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
  • LeonLeon Posts: 7,620
    edited 2008-12-14 18:16
    To detect a stall condition simply measure the speed, either by measuring the back EMF or with a suitable sensor, and cut the supply when the speed starts to drop. Or, measure the current taken by the motor.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2008-12-14 18:48
    I've always been confused about how to measure back EMF from a motor.
    Does it require using diodes to recognize when the motor is producing an opposing voltage?

    Current sensing is well documented on the Web. Google is your friend.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    It's sunny and warm here. It is always sunny and warm here.... (unless a typhoon blows through).

    Tropically, G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
  • LeonLeon Posts: 7,620
    edited 2008-12-14 19:29
    The supply is removed periodically, and the back EMF is measured with an ADC. It's a standard technique.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • stamptrolstamptrol Posts: 1,731
    edited 2008-12-15 13:26
    Leon,

    Can you show us a practical method of measuring the back EMF by your method?

    Just curious....it would seem removing the supply periodically would really upset the smoothness of mechanical power delivery.

    Cheers,

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tom Sisk

    http://www.siskconsult.com
    ·
  • LeonLeon Posts: 7,620
    edited 2008-12-15 16:30
    The back-EMF can be measured between the PWM pulses. There are lots of examples on the Web, it's a standard technique.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • SRLMSRLM Posts: 5,045
    edited 2008-12-15 19:09
    Take a look at this link, it seems to be a good description:

    http://www.acroname.com/robotics/info/articles/back-emf/back-emf.html
  • Carl HayesCarl Hayes Posts: 841
    edited 2008-12-15 19:37
    I wouldn't worry about it, since you say the stall condition will be less than two seconds. Two seconds, no problem. Two minutes, maybe. Two hours, goodbye motor unless it is designed for that kind of service. Most really small motors, especially small DC motors, will withstand a stalled condition for a long time without damage. So will many types of AC motors, but don't try it with a typical induction motor.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2008-12-16 00:54
    As Leon said, the back EMF can be measured between the PWM pulses. First think of the motor as an inductor hit with a square wave. On the transition edges the voltage will "ring" proportional to the inductance. The bigger the inductance the bigger the ring voltage or EMF. Based on this, there are two basic ways to measure the EMF... One is to measure the voltage amplitude of the ring, another is to measure the duration of the "ring" or the Width of the "ring"... since the width will be directly proportional to the amplitude you can use one of these methods or both to determine the relative amount of energy over time.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • SRLMSRLM Posts: 5,045
    edited 2008-12-16 01:24
    Is it possible to do this with any of Parallax's motor controllers? Specifically, I'm thinking of the HB-25...
  • stamptrolstamptrol Posts: 1,731
    edited 2008-12-16 02:05
    Thanks for the link to the article.

    It is a crafty method of measuring motor speed with some limitations as to getting full power out of the motor and taking a reading with a forward/reverse capability.

    It reinforces the idea that at stall, there is no rotation and therefore no back-EMF being generated, so current is only limited by the resistance of the armature winding. If all you want to do is protect against stall, measure the current with a linear Hall effect sensor or shunt resistor setup.

    Cheers,

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tom Sisk

    http://www.siskconsult.com
    ·
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2008-12-16 02:20
    There have been a couple of statements made that I'm going to leave unchallenged.

    Notwithstanding, I guess a fuse (for "stall detection") is out of the question?
  • T ChapT Chap Posts: 4,223
    edited 2008-12-16 07:06
    stamptrol said...


    If all you want to do is protect against stall, measure the current with a linear Hall effect sensor or shunt resistor setup.


    I once tried using a hall sensor on a motor driver and wasn't happy with the results, too small a variance in the sensors output. Instead I used an Anologue Devices AD628 differential amp tha can be used with a shunt resistor to read the voltage across the resistor and amplify the difference, it has programmable gain and adjustable output levels to suit your needs. With adjustable gain, you can set up the output to swing a great deal (several volts) with moderate load on the motor, and with a stall you'd see a significant output swing. For stalls I suggest using this chip feeding an A/D or comparator with a trim pot to set your trip point.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2008-12-16 17:12
    So, measuring the current DOES NOT requie a PWM supply to the motor, while measuring Back EMF does.

    Does back EMF measurement require amplification, like current measurement? If not, a comparator feeding directly to an I/O pin that is appropriately calibrated seems a much easier wiring job. And the software seems less complicated as well.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    How do you like my name change?

    aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
  • LeonLeon Posts: 7,620
    edited 2008-12-16 17:21
    One generally needs speed control with a motor and many MCUs have an ADC, so it makes sense to use it for measuring the back EMF. Amplification isn't needed, the signal might even need to be attenuated.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2008-12-17 05:32
    I haven't got much time so I'll try to cover the issues briefly and succinctly.

    1. Back EMF tells you how fast the motor is running. A DC motor with permanent magnets can be used in reverse as a dynamo. By spinning the motor mechanically it will generate a DC voltage. If you use PWM it will run the motor during it's active phase and then while it is not driving the motor there is a voltage that the motor is generating that is related to it's speed as if it were a dynamo.

    2. Current measurement tells you how much current the motor is drawing or more correctly how much load it is handling. If the motor is stalled by jamming it then this is an overload and without any back EMF the motor looks just like a dead inductive load. This simple method is used to determine if the motor has stopped as in the case of a door opener, when the door as fully opened or closed then the motor will be stalled, the current draw increases sharply and the circuitry will turn off power to the motor or reverse it if necessary.

    Nobody wants fuse (other than permanent fault style thermal fuses) because it necessitates maintenance. A polyfuse is a possibility though.

    Leon is correct when he says that amplification isn't needed because several volts will be generated and you will probably need to divide the voltage and filter it with a small capacitor.

    In saying all this it is way easier to actually hook up a motor and try it yourself, why miss out on the fun.

    *Peter*
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2008-12-17 12:24
    So, a hopelessly archaic device, a·fuse, is the answer --·unless you see that as a horrible, vexing·maintenance issue.· Guess they really missed the boat with the fuse on the HB-25.· Oh, well.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2008-12-17 13:58
    Fuses are supposed to be the absolute final safety device in case the protection circuits have failed and smoke and flames are a real possibility. Maintenance issues are major, horrible, and vexing as complete production lines could be shut down until somebody can jump on a plane to get over there and eventually find out that the designer had skimped on fault protection circuitry relying simply on a 10 cent fuse that now cost $$$$$ + &^!@#*%!! literally.

    If you are hovering over it on the bench then yes, you could just use a fuse.

    The HB-25 supposedly has protection circuits and should not rely upon a fuse for normal operation which would include motor stalls. If it does rely upon the fuse then it wouldn't go any further afield than my bench with me hovering over it.

    *Peter*
  • 66_114_97_10066_114_97_100 Posts: 4
    edited 2008-12-17 21:28
    ·
    Ugha said...
    After the stall, no operation will take place for several seconds to several minutes, then followed by normal operation.
    Unless I am misunderstanding Ugha's intentions, a fuse will not be practicable. Sounds to me like he wants to detect the stall, pause, continue normal operation.


    This link http://www.acroname.com/robotics/info/articles/back-emf/back-emf.html·and this comment
    Somebody said...
    1. Back EMF tells you how fast the motor is running. A DC motor with permanent magnets can be used in reverse as a dynamo. By spinning the motor mechanically it will generate a DC voltage. If you use PWM it will run the motor during it's active phase and then while it is not driving the motor there is a voltage that the motor is generating that is related to it's speed as if it were a dynamo.
    [font=Arial, sans-serif]It has been a while since I learned about this kind of stuff but isn't this method measuring EMF and not Back EMF?[/font]
    Somebody said...
    It reinforces the idea that at stall, there is no rotation and therefore no back-EMF being generated
    Is not the opposite true? At stall would not the current be at maximum? Therefore would not·the Back EMF be at the maximum? (as long as the supply was alternating or pulsating)
  • Carl HayesCarl Hayes Posts: 841
    edited 2008-12-17 22:35
    66_114_97_100 said...
    Somebody said...
    It reinforces the idea that at stall, there is no rotation and therefore no back-EMF being generated
    Is not the opposite true? At stall would not the current be at maximum? Therefore would not·the Back EMF be at the maximum? (as long as the supply was alternating or pulsating)
    No, the original statement is correct. Back EMF is generated in the motor and opposes the EMF from whatever source is driving the motor.· Back EMF increases as speed increases.·

    Current is maximum at stall because there is no back EMF opposing the applied voltage.

    Incidentally, the idea of measuring the back EMF by momentarily disconnecting the supply ought to work OK with permanent-magnet motors, but not with other types.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2008-12-17 23:42
    66,

    When we measure back EMF (voltage) we are measuring the voltage being generated BACK from the load (motor) rather than from the source (power-supply). This voltage can only be generated if the motor is spinning so that it acts like a DC generator. I remember doing a +12V design once where I would operate a grinder motor with a heavy wheel and the design called for an emergency stop switch that cut the power to the control board. Well, surprise surprise, guess what happened if the grinder was spinning when the emergency switch was hit and cut the power? Yes, the grinder motor would continue to generate enough power to keep the board and other motors running for a few seconds. Solution was simply to put a heavy-duty blocking diode in series with the motor and driver MOSFET. BTW, there was no way to shunt the motor as that would cause it to decelerate rapidly and the grinder wheel would unscrew and fly across the room with such speed and force that really curled your hair, thinking "Gee, that WAS a close one".

    *Peter*
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2008-12-17 23:50
    Good Grief.
    This is about a crummy hobby motor.
    Ugha said... normal DC motors (especially the cheap ones).
    This·is·not about production lines in plants without "facilities technicians" and plane trips around the world on a mission from·God·owing to a poorly designed control circuit.· If the simple stuff can be so poorly designed that the motor fuse goes flop-pot without good cause then·the stall-detecting, EMF-sensing, gold-plated horse-apple that's aeons past a rotten old fuse doesn't stand a chance anyway.
    ·
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2008-12-18 00:22
    LOL, but many others read these posts and try to apply the suggestions to their requirements so the answers need to be both specific and general. The OP could always use a driver that is rated heavier than the power-supply so then it would be up to the PSU to worry about overloads.

    *Peter*
  • 66_114_97_10066_114_97_100 Posts: 4
    edited 2008-12-18 00:40
    Looks like I will have to dig up my old text books because my recollection of Back EMF is that it is produced by any coil, (solenoids, transformers, etc.) not just permanent magnet motors. It is dependent on inductance and a varying current.

    I apologize Ugha for taking your post off on a tangent

    Thanks guys
  • LeonLeon Posts: 7,620
    edited 2008-12-18 17:35
    If the motor is stationary due to a stall condition, no EMF will be generated.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
Sign In or Register to comment.