-
Measurement of gear DC motor characteristics
Posted by
fishandchips
on 31 Mar, 2017 23:06
-
Hello, I have a DC motor with a pre-installed gearhead. I would appreciate some suggestions.
I am trying to obtain some parameters for modelling the motor using the widely used Equation 4 shown in the following link:
http://ctms.engin.umich.edu/CTMS/index.php?example=MotorSpeed§ion=SystemModelingWhen measuring parameters such as resistance, inductance, turning speed, back EMF constant, torque constant, voltage caused by back EMG, should I have the gear head removed before doing the experiment?
For example, to measure the motor resistance by Ohm's Law, one connects an ammeter in serial with the motor and record the current when the motor is spinning. Then, use V=IR to obtain the resistance. In this case, should I measure the resistance of the motor without the gear head, with the gear head installed, stall current (with gear head) or stall current (without gear head)?
-
#1 Reply
Posted by
EPTech
on 01 Apr, 2017 19:19
-
Hi there,
If you want to measure the resistance I would do it the other way round. I would put a constant current through the winding and measure the voltage across the winding as close a possible to that winding using a mV meter. That will give you the most accurate result. If you are planning on using a lab supply it will go into constant current mode anyway.
You can measure the inductance with an LRC meter.
The hardest will be to determine the torque constant. If you have another motor with advanced drive you can put the drive in constant torque mode, connect the two motors and measure the amature current under known load.
You could also connect a winch to the motor shaft and put a weight on a string and have the motor winch that up. The amount a torque involved is an easy calculation and the current can be measured. You need to take care though that the string cannot coil up on itself. The diameter needs to remain constant. Measure the current when the motor is more or less at constant speed and use a descent enough weight to overcompensate the friction of the string. Use string with as little friction as possible. Remember you need the torque to be as constant as possible.
Just measuring the static torque with a rigid bar on the shaft and a dynamometer is not deterministic because the static torque will be dependent of the angular position of the rotor within the magnetic field at constant current. You can observe this effect best with 2 pole motors with permanent magnet or externally exited field, a bit like the motor in you mathematical example.
If you want to characterize the motor I would remove the gear head for all dynamic measurements. For static measurements like winding resistance, locked rotor current and winding inductance, you can leave the gearbox on.
-
-
Thanks Pascal.
Since I am going to remove the gearbox for dynamic measurements, will the static measurements be the same if I remove the gearbox before doing all the measurements including the static one?
What suggestions do you have in terms of measurement the back EMF constant and voltage caused by back EMG?
If I want to do the kind of Simulink modelling of DC motor shown in the video, should I just use a meter to measure the current by putting it in serial to the motor without gear or measure the stall current when I stop the motor from spinning by force? Not so sure about what lock current is.
(see 11:08)
-
#3 Reply
Posted by
EPTech
on 02 Apr, 2017 01:13
-
Ah, back EMF will be one of the easiest things to measure.
Back drive the motor. That means using another motor to drive the motor under test at a known speed and measure the back EMF with a voltmeter.
Locked rotor current is a useful parameter to know in induction motors or squirrel cage motors. I guess it is not that important for your DC motor, besides, if it is a brushed motor it is a bad idea to lock the rotor. If the supply current is unlimited it will overload much faster and potentially weld the collector and burn out.
What type of motor is it actually.
All static measurements can be done just fine with the gearbox attached.
-
-
It is a DC motor with gear head. It has only two terminals (+ and -). Not sure what Poles: 3P denotes. From the description, is it brushed? Can't get the parameters from the company. So, I need to measure by myself. I shall try to get the resistance, current, inductance and the speed first. Could you please advice?
1. I suppose that with the gear head, the measured current will be between the one measured without the gear head and the one with the gear head and my hand preventing the motor from turning (i.e. stall current). Based on the youtube video, which one should I get to model the motor?
2. If I measure the resistance using the method you mentioned, does that mean I have to take apart the motor? Not sure what you meant by putting a constant current through the windings. I have a regulated power supply that I can set the input voltage.
3. If I do not have a LRC meter, is there another way to get the inductance?
4. Based on the video, is the motor speed the one with the gear head on but no other extra load?
Thank you
-
#5 Reply
Posted by
EPTech
on 02 Apr, 2017 02:35
-
If you post a picture of the motor, I can probably teel what type it is. But I am guessing it is a brushed motor with a permanent magnet stator.
1. You need to ask yourself what you want to quantify. The motor or the motor with the gear box. There will be a difference in current since the gearb box also presents a load due to mechanical friction.
2. No, it means you need to put a suitable current through the motor, lock the rotor and measure the voltage across the terminals. Start of with a modest current and work your way up until you get a decent reading in the mV range. Be careful not to burn the motor.
3. Yes, put a sine wave on the terminals and measure voltage and current (over a low inductance series resistor) on a scope and calculate the inductance. Or put a known capacitor in parallel and determine the resonance frequency. You may want to try several capacitors to get in the right frequency range of you function generator.
4. It looks to me like they are simulating just a motor without gearbox and a static load, as in a flywheel on the shaft.
-
-
Whether to leave the gearbox in place or not depends on how you intend to use the equation in your model of the system. If you are modelling the gearbox output shaft you will either have to measure the motor constants at the gearbox output, or model the gearbox outside of your motor model. For the simplest models, which are often enough, this is merely a matter of scaling by the gear ratio. If you don't know the gear ratio you might as well measure and model at the gearbox output.
-
-
Thanks Pascal and CatalinaWOW,
My initial thought would be to model the motor in Simulink using the method mentioned in the video. Then, add a gearbox outside the motor model in Simulink. What do you think of this approach? I do know the gear ratio. I will take a photo of the motor.
-
#8 Reply
Posted by
EPTech
on 02 Apr, 2017 03:15
-
In all honesty, you are starting of with too many unknowns.
Are you stuck to this motor? If you are not, I would advice you to start of with a motor where the parameters you summed up in your first post are known. If you think you can determine these parameters by doing the simulation, you are mistaking. In the video all motor parameters are know, probably from a datasheet of fictive. Determining the parameters will be a greater challenge than making the simulation IMHO.
-
-
Your mention of Simulink implies that you are creating a dynamic model. Step response, bandwidth, stability and all that jazz. Which opens up another whole list of parameters to measure or model. At the very least moments or other measure of rotational inertia. If you are not doing a dynamic model, just want steady state speed, maximum supported load and the like there is no need to bring Simulink into the picture.
-
-
Unfortunately I am stuck with this motor and I need to study the dynamic of the system. Please help.
-
-
Double checked with the company, it is a brushed motor.
-
-
There aren't any one line answers to your question. I would recommend hitting the library and doing some intense reading. I doubt that you can find a copy, but my favorite recipe book on the subject is "DC Motors, Speed Controls, Servo Systems" published by Electro-Craft Corp.
You may also find yourself understanding better if you realize that these mechanical systems are modeled by differential equations very similar to those used for electrical circuits. Figuring out the analogous properties to inductance, capacitance, charge and current may lead you to a better understanding of what you need to measure and model.
-
-
I just tried to measure the current. When I connected the multi-meter in serial with the motor, the motor did not turn. I read that I need a better meter with 20A or get a current clamp. Checked the website of some big local stores. Their meters can handle only 10A and only AC current meters are available. I guess I need a DC current meter. Am I right? Is there another way to solve this problem?
-
#14 Reply
Posted by
EPTech
on 02 Apr, 2017 19:08
-
Well,
You probably blew up the fuse in the low amp range of your DMM as well. The armature current can be very high in a DC motor, even in a small one. That is quite basic. I cannot help but wonder what your background is. But I concur with Catalina. You need to read up on the subject. You can always get back to us for further ideas and suggestions.
-
-
Hello, my major is not in EE unfortunately.
Will the following product get the job done in terms of measuring the current? It can take DC current up to 400A.
https://www.amazon.com/Auto-ranging-Multimeter-Resistance-Capacitance-Frequency/dp/B01N014USEIn case I get a LCR meter for measuring the inductance, which ones do you recommend? Ideally a low cost one that could get the job done. How about the uxcell DMiotech Digital Multimeter Resistance Capacitance Inductance LCR Multi Meter Test UA6243L?
-
-
Hello, my major is not in EE unfortunately.
Will the following product get the job done in terms of measuring the current? It can take DC current up to 400A.
https://www.amazon.com/Auto-ranging-Multimeter-Resistance-Capacitance-Frequency/dp/B01N014USE
In case I get a LCR meter for measuring the inductance, which ones do you recommend? Ideally a low cost one that could get the job done. How about the uxcell DMiotech Digital Multimeter Resistance Capacitance Inductance LCR Multi Meter Test UA6243L?
OK, since you are not a EE you are a step ahead. What field are you studying? Mechanical? You should be able to write the differential equations for rotary motion. Again, Simulink is a nice way to represent and solve systems of differential equations, but you have to understand what equations you want to solve, what the parameters and relationships are.
The meter should be fine, but as with all instruments the operator is the limit. Look in some catalogues (on line is fine) for motors and particularly motors with attached gearheads. See the range of motor parameters that are available. Your motor should be somewhere in that range. Plug extremes into your motor equations and see what currents and voltages would arise from those extremes of current. You will notice associations between some of the constants and physical size. This will let you narrow the extremes somewhat as you compare the physical attributes of your motor to the catalog motors. After this kind of preparation you will know how to set up your experiment. The range of possible values. And you will recognize if you get a number you haven't seen in your modeling that something is wrong and you need to do some more thinking. What is wrong may be something in your test setup, in your thinking or even a defect in your motor. How would you figure out which of them it might be?
-
-
Ah, back EMF will be one of the easiest things to measure.
Back drive the motor. That means using another motor to drive the motor under test at a known speed and measure the back EMF with a voltmeter.
Hello, thanks. I have measured the no load current and the internal resistance of the motor. I am working on measuring the speed of the motor.
For measuring the back emf of the motor, do you have any other good method? I don't have anything to couple one motor to the other. Not sure if the method presented in:
http://electronics.stackexchange.com/questions/54997/how-can-i-measure-back-emf-to-infer-the-speed-of-a-dc-motoris valid. While the motor was spinning, I measured the voltage (Vrm) across the motor. Since the current was fairly constant, I could ignore L. So, no voltage across the inductor. In my case, I connected the DC source directly to the motor. Thus, the duty cycle is 100%. Voltage effectively applied to the motor is 9V (the source input voltage). Voltage drop over motor resistance is the Vrm. Back EMF is then 9V-Vrm. Is this acceptable?
-
-
3. Yes, put a sine wave on the terminals and measure voltage and current (over a low inductance series resistor) on a scope and calculate the inductance. Or put a known capacitor in parallel and determine the resonance frequency. You may want to try several capacitors to get in the right frequency range of you function generator.
Thanks. Do you mean put a sine wave on the two terminals of the DC motor and measure the voltage across the motor? Without the power supply, maybe the motor won't turn?
As for the current, do you mean just put the multimeter in series with the motor and also a low inductance series resistor? By "low inductance series resistor", how many ohm do you recommend? Does the order of the resistor and the motor matters?
-
-
Ah, back EMF will be one of the easiest things to measure.
Back drive the motor. That means using another motor to drive the motor under test at a known speed and measure the back EMF with a voltmeter.
Hello, thanks. I have measured the no load current and the internal resistance of the motor. I am working on measuring the speed of the motor.
For measuring the back emf of the motor, do you have any other good method? I don't have anything to couple one motor to the other. Not sure if the method presented in:
http://electronics.stackexchange.com/questions/54997/how-can-i-measure-back-emf-to-infer-the-speed-of-a-dc-motor
is valid. While the motor was spinning, I measured the voltage (Vrm) across the motor. Since the current was fairly constant, I could ignore L. So, no voltage across the inductor. In my case, I connected the DC source directly to the motor. Thus, the duty cycle is 100%. Voltage effectively applied to the motor is 9V (the source input voltage). Voltage drop over motor resistance is the Vrm. Back EMF is then 9V-Vrm. Is this acceptable?
Could anybody please check if this way of measuring the Back EMF is valid? Thanks
-
#20 Reply
Posted by
rstofer
on 16 Apr, 2017 15:26
-
For measuring the back emf of the motor, do you have any other good method? I don't have anything to couple one motor to the other. Not sure if the method presented in:
http://electronics.stackexchange.com/questions/54997/how-can-i-measure-back-emf-to-infer-the-speed-of-a-dc-motor
is valid. While the motor was spinning, I measured the voltage (Vrm) across the motor. Since the current was fairly constant, I could ignore L. So, no voltage across the inductor. In my case, I connected the DC source directly to the motor. Thus, the duty cycle is 100%. Voltage effectively applied to the motor is 9V (the source input voltage). Voltage drop over motor resistance is the Vrm. Back EMF is then 9V-Vrm. Is this acceptable?
Back EMF will obviously vary with load but the internal resistance of the motor probably doesn't change a lot. So, Ohm's Law still applies as does Kirchoff's. You know the internal resistance and, absent back EMF, the current would just be the applied voltage divided by the internal resistance. What you will actually read is the applied voltage MINUS the back EMF (voltage) divided by the internal resistance. As you load the motor, you expect the current to increase and, when it does, the only possible interpretation is that the back EMF is reduced at that load.
Obviously, the motor needs to be running to measure back EMF and the source needs to be sufficient to drive the motor at load. So, it is necessary to measure both the motor terminal voltage and current at the same time just in case the battery internal resistance starts to be a factor (and it will). Once you get voltage and current versus some load (how to measure?), you will have a graph for back EMF.
Applying a load is non-trivial. Something like a Prony Brake is usually used. Something like a slipping clutch with a lever arm attached to a scale.
Unloaded back EMF is interesting but since the value changes dramatically from no-load to full-load, it would seem to me that multiple measurements are desireable. Or, you can straight line the graph from unloaded back EMF at some RPM (to be measured) down to 0V back EMF at 0 RPM. I doubt that the straight line graph is true anywhere except the end points.
Your DMM will add inductance (and a small bit of resistance) to the circuit. I would ignore it. Assuming the motor is sufficiently small, just use the meter 10A current scale for a very short period of time (RTFM).
All of the above gets you back EMF. To get inductance, use the other procedure of applying a sine wave and looking at the phase shift between voltage and current. I'm less comfortable with this measurement. Inductance is a function of frequency and this is typically measured with a nice clean sine wave. The motor is almost never driven this way. Square waves (from PWM) have harmonics so which frequency are we talking about?
-
#21 Reply
Posted by
rstofer
on 16 Apr, 2017 15:30
-
You can also determine the internal resistance by stalling the motor with full applied voltage and measuring the current. Again, you need to take both measurements simultaneously and be quick about it. The ammeter is getting hot! (probably)
-
-
Thanks. I have an external power supply. I could use a digital multi-meter to measure the applied voltage before connecting it to the motor. Why do I need to take the current in addition to the applied voltage at the same time? Isn't the applied voltage the same before and after I connect the power supply to the motor?
-
-
All of the above gets you back EMF. To get inductance, use the other procedure of applying a sine wave and looking at the phase shift between voltage and current. I'm less comfortable with this measurement. Inductance is a function of frequency and this is typically measured with a nice clean sine wave. The motor is almost never driven this way. Square waves (from PWM) have harmonics so which frequency are we talking about?
Is it much better to just buy a LCR meter?
-
#24 Reply
Posted by
retrolefty
on 16 Apr, 2017 21:40
-
Thanks. I have an external power supply. I could use a digital multi-meter to measure the applied voltage before connecting it to the motor. Why do I need to take the current in addition to the applied voltage at the same time? Isn't the applied voltage the same before and after I connect the power supply to the motor?
It depends on the current capacity of your power supply. It's usually the case that the voltage will sag depending on actual motor load. Voltage under load X current will give you true power wattage consumed.