Controlling high power AC or DC motors.

[jakeypoo]

Hello all,

I have a question for those of you who deal with larger motors.

In terms of simplest control, what is the best choice for large (1-3 hp/ 0.7-2.2kW) electric motors?

I deal with low power motor control most of the time (<200W) where some power MOSFETs and PWM will do the trick. But looking at motors in this larger power class, I've come up with 3 options.

          - Single phase AC motors
          - Three phase AC motors
          - Three phase BLDC motors

It seems like in this power realm (>750W) DC motors scale up very steeply in cost, so AC is preferable.

By control I mean speed control and ideally braking.

I have a rough idea of control for all three options. My question is, for those of you who have been down this path before, what is the least headache inducing path to take (specifically power supply and drivers) to control this with low voltage logic?

Thanks in advance.

[IanB]

The traction industry has been down this path. Historically traction motors were all DC. Speed was controlled by varying a resistance in series with the motor and by varying excitation in the field coils. Lately everything has moved over to AC motors with electronic variable frequency drives. Early drives used thyristors but later drives use IGBTs.

I remember an article somewhere that give a typical application map: lower power drives might use MOSFETs, medium power might use SCRs and high power might use IGBTs. I think 1-3 hp would be considered small in an industrial drive context.

Can you buy an off the shelf drive rather than designing one from scratch? I imagine there are plenty of traps for the unwary in designing such a thing.

[Gall]

Small IGBTs are usable at low and medium power too. Older high power units might use SCR/triac.

[Zero999]

Hello all,

I have a question for those of you who deal with larger motors.

In terms of simplest control, what is the best choice for large (1-3 hp/ 0.7-2.2kW) electric motors?

I deal with low power motor control most of the time (<200W) where some power MOSFETs and PWM will do the trick. But looking at motors in this larger power class, I've come up with 3 options.

The best option is to buy a ready made variable frequency drive  (VFD).

          - Single phase AC motors

Don't use them, use a three phase motor running off a VFD powered by single phase.

          - Three phase AC motors

Cab be powered from either single or three phase with the appropriate VFD.

          - Three phase BLDC motors

Basically the same as a three phase motor except they may have other features such as permanent magnets (synchronous), designed to be run off a square wave rather than a sine and have feedback.

It seems like in this power realm (>750W) DC motors scale up very steeply in cost, so AC is preferable.

Yes DC motors are rarely used in industry these days.

By control I mean speed control and ideally braking.

A VFD can do that and if you can buy another inverter to inject power lost through breaking back into the grid, saving you both power and room for a large breaking resistor

I have a rough idea of control for all three options. My question is, for those of you who have been down this path before, what is the least headache inducing path to take (specifically power supply and drivers) to control this with low voltage logic?

A ready made VFD will have all of that built in, will be programmable and can be controlled by standard analogue or digital signals.

[jakeypoo]

Thank you,  everyone!

Variable frequency drive using one of the available single chip power modules or IGBTs was where I was heading with this one.
Really needed the reassurance since these kind of powers are uncharted waters for me.   

Can you buy an off the shelf drive rather than designing one from scratch? I imagine there are plenty of traps for the unwary in designing such a thing.

A ready made VFD will have all of that built in, will be programmable and can be controlled by standard analogue or digital signals.

This is a one-off rig for some mechanical fatigue testing, and low cost is an issue. Industrial controls are pricey.  Still hoping to find a suitable surplus motor.

[IanB]

This is a one-off rig for some mechanical fatigue testing, and low cost is an issue. Industrial controls are pricey.  Still hoping to find a suitable surplus motor.

This might be a blind alley for you, but check out what is available in the remote control hobby area. There are all sorts of motors and variable speed drives used there. Often the power levels are only in the 100's of watts, but it's possible you may find something in the kilowatt range. It's not unheard of in the bigger size models.

[Kiriakos-GR]

Hello all,

I have a question for those of you who deal with larger motors.

In terms of simplest control, what is the best choice for large (1-3 hp/ 0.7-2.2kW) electric motors?

The question is very generic, for about every specific application there is specialized solutions.
Gives to us more info of what you have in mind.

[Zero999]

A ready made VFD will have all of that built in, will be programmable and can be controlled by standard analogue or digital signals.

This is a one-off rig for some mechanical fatigue testing, and low cost is an issue. Industrial controls are pricey.  Still hoping to find a suitable surplus motor.

Think of the time spent designing your own VFD (this id what you'll need to do) and the amount of times you'll need to replace components or even motors you killed during its development. Look at the surplus market for VFDs or you may be able to get one from an old washing machine, tumble dryer or air conditioning unit but controlling it may be more difficult.

[Gall]

If you really, really have a reason not to use a off-the-shelf VFD, think about cloning one. It is easier to do some little changes (such as using a different layout and enclosure) than to design everything from scratch.

(Also applies to switch-mode PSUs).

[Hypernova]

I'm currently in my first job out of uni and it just happens to be writing firmware for BLDC drivers with dsPIC and Renesas MCU's. My best advice is to go off the shelf unless you are prepared to deal with dead MOSFET's and IPM's on a daily basis when you first start out. Another biggie is that high powered industrial motor's tend to run on 300V+ so you need to take isolation between the HV(inverter) and LV(MCU, interface) sides into account.

[jakeypoo]

If you really, really have a reason not to use a off-the-shelf VFD, think about cloning one. It is easier to do some little changes (such as using a different layout and enclosure) than to design everything from scratch.

My thought's exactly.

I like the looks of some ST products for this sort of application.
http://www.stmicroelectronics.com.cn/internet/evalboard/product/251251.jsp?WT.ac=fp_sep11_steval-ihm028v1
http://parts.digikey.com/1/parts/1900435-igbt-ipm-module-17a-600v-25sdip-stgips20k60.html

The problem with these sort of one-off jobs is that there are a handful of other sensors and other actuators that need to be integrated into the system, so spinning my own board and cloning the required parts of a VFD makes sense. 

Maybe I'm just too much of a control freak to use everything off the shelf. I like to call it being detail oriented
Besides doing new things and learning is more fun.

Again, thanks for the input from everyone.

[Zero999]

What's your budget and power requirements?

There are lots of Chinese VFDs on ebay.

Here's one which is much cheaper than the ones I use at work but I don't know the brand.
http://www.ebay.co.uk/itm/VARIABLE-FREQUENCY-DRIVE-INVERTER-VFD-NEW-3HP-2-2KW-10A-/110718279173?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item19c752de05