Using a DC Motor as a Controllable Brake

Thread Starter

qrb14143

Joined Mar 6, 2017
112
Hi all,

I have a brushed DC motor rated at 225W @ 3000rpm 24V. Datasheet available here if needed: https://www.parvalux.com/media/wysiwyg/parvalux/datasheets/pm90-data-sheet-10-10-2016.pdf

I have mechanically attached the output shaft of this motor to an off-the-shelf AC induction machine running at 3000rpm. The purpose of this experiment is to use the DC machine as a brake to apply a known load torque to the AC machine to evaluate the performance of the AC machine controller.

As far as I am aware, the torque required by a DC machine acting in the generator mode is proportional to the armature current. In other words, if I can control the current coming out of the machine, I can control the torque. My plan is to connect the DC motor output into a programmable four-quadrant amplifier and set it to sink a precise amount of current, therefore allowing me to set the load torque. I am not sure where the terminal voltage falls into this though, if the motor speed reduces as a result of the applied torque, the terminal voltage will reduce so do I just let the four quadrant amplifier deal with this on its own and try to maintain a constant current?

My other piece of thinking was to feed the DC machine into a fixed output buck converter and convert to say 12V. I would then sink the output of the converter into the four-quadrant amplifier. If voltage is fixed and I am controlling current, I am therefore controlling power. If I can control power and I am measuring rotor speed, it follows that I can deduce the applied torque, so in other words, this would be another way of controlling torque. Is this correct?

I would be interested to hear if anyone has any experience of a similar setup as I am not sure which approach to go with!

Thanks
 

tcmtech

Joined Nov 4, 2013
2,867
An other option for a simple but pretty effective electric dynamometer is to use a common induction motor and feed DC to the windings.

It doesn't take much DC power input to get a huge amount of dynamic braking power that way. ;)
 

LesJones

Joined Jan 8, 2017
4,174
Many years ago we used that method to measure output power. The shunt wound DC motor (Being used as a generator.) was mounted on an extra set of bearings. The motor casing was then prevented from rotating with an arm attached to a spring balance. This enabled us to measure the torque. The brushes were connected to a resistive load and the field curent controlled (From a separate power source.) to enable us to adjust the breaking torque. Using the torque measured with the spring balance and the speed measured with a tachometer we could calculate the mechanical power.

Les.
 

Thread Starter

qrb14143

Joined Mar 6, 2017
112
The device you need is called a Dynomometer if you Google you should find a good deal of information on constructing one.
Max.
I am familiar with a dynomometer but most of the designs available use a beam attached to the rotor spindle with a measured mass applied at one end allowing torque to be deduced directly by T= Fr.

I am already committed to using the DC motor as it is available to us and can be easily connected to the AC machine. I have tested using the DC machine as a brake by using it to brake another motor. This was done as I said above by sinking the current into an amplifier. This produced a braking effect in proportion to the current flowing into the amplifier.

This very simple test comprised of measuring the current with a current clamp and visibly and audibly verifying the reduction in motor speed in response to the applied load torque. What I would like to do though is use that measurement of current to deduce exactly how much torque I am applying with a view to using this knowledge to apply a torque profile over time.
 

Thread Starter

qrb14143

Joined Mar 6, 2017
112
Many years ago we used that method to measure output power. The shunt wound DC motor (Being used as a generator.) was mounted on an extra set of bearings. The motor casing was then prevented from rotating with an arm attached to a spring balance. This enabled us to measure the torque. The brushes were connected to a resistive load and the field curent controlled (From a separate power source.) to enable us to adjust the breaking torque. Using the torque measured with the spring balance and the speed measured with a tachometer we could calculate the mechanical power.

Les.
This sounds a lot like what I want to do. I'm not all that interested power though, only torque.
What I want to do is take a spreadsheet of torque vs time values for a typical pump or suchlike and feed them into the four quadrant amplifier. The DC machine will then apply this torque profile to the AC machine, allowing me to verify whether the AC machine controller responds correctly to the torque profiles it is likely to experience in service.

You mention that you controlled field current to alter the torque and then measured that torque using a spring balance. If the motor is operating at or near its rated speed won't the torque be proportional to current, allowing the torque to be approximated by measuring the current as opposed to a dedicated measuring device?
 

LesJones

Joined Jan 8, 2017
4,174
I assume in your last paragraph you are talking about the shunt motor that is acting as the brake. I think the torque is proportional to the product of the field current and armature current. So assuming a fixed value resistive load if the speed is increased with the field current constant the ouput voltage will increase causing the armature current to increase. So I think to adjust the torque you would need a servo system controlled by the reading of the spring balance (Or probably in practice a load cell.). Another aproach would be to have a constant current load (In place of a simple resistive load.) so that the armature current did not change with speed.

Edit. I have just noticed that the motor you plan to use is a permanent magnet motor so you would need a variable constant current load so you could change the torque. This is what you are saying in the last sentence of post #7

Les.
 
Last edited:

Thread Starter

qrb14143

Joined Mar 6, 2017
112
I assume in your last paragraph you are talking about the shunt motor that is acting as the brake. I think the torque is proportional to the product of the field current and armature current. So assuming a fixed value resistive load if the speed is increased with the field current constant the ouput voltage will increase causing the armature current to increase. So I think to adjust the torque you would need a servo system controlled by the reading of the spring balance (Or probably in practice a load cell.). Another aproach would be to have a constant current load (In place of a simple resistive load.) so that the armature current did not change with speed.

Edit. I have just noticed that the motor you plan to use is a permanent magnet motor so you would need a variable constant current load so you could change the torque. This is what you are saying in the last sentence of post #7

Les.
What I am proposing to do is use the four quadrant amplifier to provide that constant current load. I just wanted to verify that controlling the current would provide a means of controlling torque which, by the sounds of things, it will.
 

LesJones

Joined Jan 8, 2017
4,174
I think the current through the motor should be a good aproximation of the torque but if I was using the system to measure the torque II would measure it directly by the arm and spring balance on the motor casing.

Les.
 
Top