Are switching power supplies good for powering motors?

Thread Starter

Dritech

Joined Sep 21, 2011
901
Hi all,

I need a high power DC power supply to be used while prototyping using DC motors (ex: linear actuators, DC geared motors, stepper motors, etc etc).
To buy a bench-top variable power supply with high power will cost me an arm and a leg, so I was considering buying two Mean-well power supplies (one fixed 12V and one fixed 24V, which are the most commonly used voltages for driving these types of motors). I can then do a neat enclosure to make it a simple bench-top supply.

My question is, are these power supplies good for powering DC motors? will they withstand sudden current changes and back-EMF which may be generated by the motor while testing? Will they operate correctly when using PWM to power the motors?
I was particularly looking at the PSPA-1000-12 and the PSPA-1000-24

Any advise would be highly appreciated.
 

dendad

Joined Feb 20, 2016
4,451
A simple unregulated analog power supply may be easier and be able to supply the large starting under load current more that the switching one.
DC motors are not too fussy about the supply, and will usually run quite well on unfiltered DC.
What ever the supply, provision needs to be made to survive the starting and stall currents.
But if the motor max current is handled by the switch mode supply ok, just go with it I reckon.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
In all the systems I have installed, I always prefer linear supplies rather than SMPS, one: regulation is not really necessary for motors, and two: linear style are much more rugged and easier to fix, should the necessity arise.
Also if using motor control, PWM etc, it is curstomary to size at least 10% higher for the supply voltage.
Max.
 

Thread Starter

Dritech

Joined Sep 21, 2011
901
Thanks for your replies. I am surprised that a linear power supply is being recommended. I had the impression that to have the same power as these SMPSs, you will need one beefy transformer.

Also if using motor control, PWM etc, it is curstomary to size at least 10% higher for the supply voltage.
So lets say a 24V DC motor is being controlled using PWM, should the supply be approx 27V?
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Thanks for your replies. I am surprised that a linear power supply is being recommended. I had the impression that to have the same power as these SMPSs, you will need one beefy transformer.
So lets say a 24V DC motor is being controlled using PWM, should the supply be approx 27V?
You do need a transformer, I prefer Toroidal, if any sort of auxiliary supply is needed, a small over-wind can be added.
27vdc I would consider the minimum.
Max.
 

Glenn Holland

Joined Dec 26, 2014
703
All industrial motors are powered by some kind of switch mode power supply. For DC motors, phase control or a chopper (using SCRs or triacs) is the most common and the IGBT inverter is used for AC motors.

However, from experience with DC motors in the electric buses and street cars at my transit agency, you should not use any power source that uses IGBT switches. We tried using an IGBT chopper (to replace the old fashioned switch resistance control) for the DC motor on one of our trolley buses and the sharp edges wreaked havoc on the commutator, the armature and field windings.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
All industrial motors are powered by some kind of switch mode power supply.
All?
That's Odd I have been working in the industrial field using DC and BLDC motion control servo's etc, in the motion control area since the 80's and the majority i have implemented and spec'd in have used a linear supply.
The OP also mentions stepper motors, one of the popular manuf such as Gecko recommend linear supply also.
Max.
 

Glenn Holland

Joined Dec 26, 2014
703
All?
That's Odd I have been working in the industrial field using DC and BLDC motion control servo's etc, in the motion control area since the 80's and the majority i have implemented and spec'd in have used a linear supply.
The OP also mentions stepper motors, one of the popular manuf such as Gecko recommend linear supply also.
Max.
I was referring to the motors used in elevators and transit vehicles which are 10 Hp. or larger. A linear power supply for these applications would need some kind of intermediate control such as switched resistance or a DC chopper. There are a lot of old DC elevators that formerly used a 6 or 12 pulse rectifier bridge with switch resistance that have been upgraded to SCR phase control.
 

dendad

Joined Feb 20, 2016
4,451
The phase control power supplies with SCRs or TRIACS are not quite the same as the high frequency switch mode supplies that are in question.
They are more like "adjustable analog" rectifiers.
The phase control types are pretty good for motors. Usually, they are quite tolerant of overloads, assuming they are made well.
The switch mode supplies rectify the mains then chop the high voltage DC into a high frequency transformer and rectifier, where as the phase control circuits operate at the mains frequency.
 

Thread Starter

Dritech

Joined Sep 21, 2011
901
Hi all,

So what are the main disadvantages for using a SMPS for powering DC motors (including steppers) ?
Also, is there a high risk that a SMPS will not withstand sudden current changes etc which may be expected from a DC motor? Are these power supplies prove to being damaged by motor back-EMF ?
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
For the motion control and area I worked in, not only were SMPS (regulated) supplies not necessary, they tended to be less reliable in motor control with less failure when and if, power surges occurred etc.
And also the ease of repair of linear as I mentioned.
Paper on it here
Max.
 

Attachments

MaxHeadRoom

Joined Jul 18, 2013
28,617
The phase control power supplies with SCRs or TRIACS are not quite the same as the high frequency switch mode supplies that are in question.
They are more like "adjustable analog" rectifiers.
The phase control types are pretty good for motors. Usually, they are quite tolerant of overloads, assuming they are made well.
The only problem we used to have with 3ph Triac bridge controllers, they in most part were connected directly across the mains, in times of electrical storms and power failures, they often would blow a component, Rectifier fuses can be used to prevent the failure, but in high current values could be very pricey $$$'s.
Max.
 

MisterBill2

Joined Jan 23, 2018
18,167
I see the term "high power" but no numbers. In some realms a whole amp is "high power", while in others, "High" does not begin until at least 100 amps.
The advice depends a lot on how much power you are talking about. I have seen a medium power supply that used a 20 amp Variac to feed a 10 to one stepdown transformer with rectifiers giving zero to about 24 volts at a lot of amps. It stood next to the bench because of it's size.
 

nsaspook

Joined Aug 27, 2009
13,079
The phase control power supplies with SCRs or TRIACS are not quite the same as the high frequency switch mode supplies that are in question.
They are more like "adjustable analog" rectifiers.
The phase control types are pretty good for motors. Usually, they are quite tolerant of overloads, assuming they are made well.
The switch mode supplies rectify the mains then chop the high voltage DC into a high frequency transformer and rectifier, where as the phase control circuits operate at the mains frequency.

I much prefer the older phase control supplies for loads that can effectually be close to zero ohms at startup. You can buy switchers with motor tolerant control systems (voltage fold-back instead of shutdown) but it's not a standard specification on most cheaper models.
 
To the OP,
A lot of what is being said here is only half true as it is only half the story. Also beware statements such as "all industrial motors..." as that was a complete furphy.
Fact is an SMPS would be just fine provided the following caveats are met:
1. Beware the ripple current. If you control the motor speed with PWM from the PSU output then you are adding directly to the ripple current in the output capacitors of the SMPS. The SMPS manufacturer will not have made allowance for this hence the poor reputation with others regarding reliability. You will need to isolate the output of the PSU from the motor PWM ripple current (think inductance) and add after that sufficient capacitance with sufficient ripple current rating to handle your PWM load current.
2. Check that the PSU is good with a load that will vary as you expect the motor load to vary. If the motor is the only real load and it is turned on and off a lot (I don't mean the PWM, I mean run and stop) then the PSU will be experiencing sharp step load shifts which may over stress the PSU is one of several ways but far more complicated than I can deal with here. Suffice to say, at least check the output of the PSU with step loads and measure the overshoot and undershoot. If they are extreme then that SMPS is probably not the one for the application.
The maximum voltage you can apply to the motor windings is a function of the maximum duty cycle and the dielectric strength of the windings. Standard grade 1 polyurethane is absolutely ok with an SELV (up to 60VDC) and old fashioned varnish or grade 2 polyurethane is good for much higher voltages.
So if you wanted to run a 12V motor from 24V just keep the PWM duty cycle below 50% and you are good to go. Be careful though not to push this too hard. The higher voltage will develop much higher currents in the windings much more quickly and while magnetic saturation is unlikely given the air gaps between rotor and stator the I2R losses and the starting torque could become a problem. Once the motor is at speed it should settle down ok (all of this presumes either PM DC / mechanically commutated DC or Universal motor (field and stator windings in series with commutation in the middle of the two). One side effect of the higher voltage and lower maximum duty cycle is that the starting torque and the torque developed under load can be much higher so beware burning the motor out with overload.
The higher supply voltage is a trick sometimes used for stepper motors where the motor is initially hit with a much higher voltage with each pulse to overcome the winding inductance and so build up the magnemotive forces quickly then the voltage is reduced to the normal voltage for the remainder of the pulse to maintain the winding current and hence maintain the holding torque. This technique means higher motor speeds and can also be used for short and low duty cycle periods of higher than usual torque, ie high inertial loads.
In any event, PWM speed control is by far the most common speed control technique for PM type motors (printers for example are loaded with these motors and use this technique) and there is no good reason to exclude SMPS you just have to know what you are doing with it and how to protect it. A linear supply for a motor of any substance is a profoundly outmoded concept in the 21st century.
If you have concerns or issues with excessive arcing on the commutator as another respondent has mentioned then a snubber is the obvious solution. You can probably derive values for a standard RC snubber fairly easily by trial and error. If you'd rather design one then that gets a bit more tricky. Unless you have engineering skills I wouldn't know about, or even if you do, go the trial and error at least for a first attempt is my advice.
The high frequency you will be applying (the PWM frequency) would be a problem with the magnetics of a PM DC motor except for the facts:
1. the air gap between rotor and stator holds the vast bulk of the magnetic field energy and air has a virtually zero loss angle (tan delta)
2. the inductance of the windings is such that while the voltage applied is going rail to rail the current is not moving that much. If you could know the inductance of the windings, the resistance and back emf for the motor at steady state then you could calculate the ripple current in the windings.
I wouldn't recommend going to MHz with your PWM but a frequency high enough for your control loop (assuming closed loop control) to keep the PWM related poles beyond your care then that should be ok and be no more than 50kHz at a guess. PWM down to even 100Hz is also ok but I would be careful not to push the voltage too much higher than the motor rated voltage in that case.
If you do go with a linear supply (and what has been discussed is not really a linear supply as it has no regulation other than the phase control types mentioned) be sure to filter out your PWM current on your input (supply from the linear supply). Assuming a sizeable motor, nobody will thank you for the EMI produced that meant they missed their favorite day time soap opera ;-)
Hope that helps, and helps you sort the wheat from the chaffe :) Good luck with your project.
 
I see the term "high power" but no numbers. In some realms a whole amp is "high power", while in others, "High" does not begin until at least 100 amps.
The advice depends a lot on how much power you are talking about. I have seen a medium power supply that used a 20 amp Variac to feed a 10 to one stepdown transformer with rectifiers giving zero to about 24 volts at a lot of amps. It stood next to the bench because of it's size.
I had one just like except it was a 3phase variac at 15A per phase feeding a 6kVA 3 phase transformer followed by rectifiers on a large heatsink with a powerful fan and a motherload of smoothing capacitors. We were pulling anything up to 300A out of this rig. Given the overload protection was the 3 phase circuit breaker feeding the variac this was a scary bit of gear. You certainly knew you were alive if the MOSFET switches in the inverter it was powering died. ;-)
 
Hi all,

So what are the main disadvantages for using a SMPS for powering DC motors (including steppers) ?
Also, is there a high risk that a SMPS will not withstand sudden current changes etc which may be expected from a DC motor? Are these power supplies prove to being damaged by motor back-EMF ?
I have posted a detailed reply already but didn't answer this question about back emf: the back emf is ok generally with the following caveats:
1. if the back emf is higher than the output voltage and there is a cro bar overvoltage protection circuit then that could be an issue.
2. the back emf should not be higher than the maximum output voltage spec of the SMPS
3. If the back emf exceeds the set voltage of the SMPS then the SMPS will probably respond by reducing its own PWM duty cycle to zero which will also reduce the auxiliary power supply that runs the PWM circuits of the SMPS (assuming the circuit arrangement is as described, ie auxiliary winding to power PWM controller) and may trip the under voltage lockout of the PWM controller so your SMPS output may drop out temporarily as well. So the immediate effect may be intermittent / pulsing power supply and unless the SMPS internal bootstrap is sturdy enough that could burn out as a result. Not to mention the stress on the rest of the electronics of the SMPS.

One other consideration I forgot to mention is that higher power SMPS are required to have input power factor correction which typically means a precedent boost converter to produce the high voltage DC bus (400V for a 240V input SMPS) then the DC/DC to produce the lower output voltage with isolation from the mains. The PFC adds a level of complexity to the SMPS design and if that is what you have in your SMPS then that is the most likely circuit that will have trouble with step loads happening a lot because of the large inductance in that converter and the (typically) relatively slow closed loop control. You'd need to check with the manufacturer on this issue for this type of topology.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Hope that helps, and helps you sort the wheat from the chaffe :) Good luck with your project.
A bit more "Chaff" !
My response was from experience purely from the reliability issue, regarding SMPS. There is no disputing they will not work, just not the best choice IMO..
This was from implementing motors of all stripes and sizes.
Regulation is rarely if ever needed in motor control, even servo's. As the Electronic Design paper shows.
Many of the stepper motor manuf. also do not recommend SMPS for their products.
Max.
.
 
Also, most people coming here to ask this question, aren't going to buy the SMPS that is properly vetted to do the job, they buy for price not correctness.
Hence my suggestion for a filter to remove the ripple current of the motor before it hits the power supply output. The rest of what I said was, if nothing else, an explanation of the considerations involved and the reason so many seem to be saying an SMPS is unreliable in this application. That need not be the case for the reasons I stated.
I attempted to provide a full answer to the question and an explanation for the chaffe to give the OP a more accurate picture of the facts and not just a statement like "These don't work but these do". I assume the OP and others might like the added details, an opportunity to learn more than they knew from someone who knows. That's all. Sorry if that offends you.
 
Top