Motor speed control by PWM

I am planning to use this 12V DC brushless motor. It includes a 2 pulses/rev tacho output (at least that's what I think the datasheet page 14 means).

I want to control speed of the motor at 2000rpm, its nominal operating speed. I thought an FET with PWM could achieve this, using pulsin on the tacho output to regulate the mark/space. What frequency would anyone suggest for the PWM? Does it matter?

Or should I approach the problem differently? I think I could manage by just measuring the RPM and altering other things to match but I think constant speed would be simpler; the application is a POV display.

Derek

I am no expert on brushless motors but one has to be aware that they usually contain active electronics so may not respond well to having their supply PWM controlled.

It may be necessary to convert PWM to a voltage and buffer that so the motor is actually DC-voltage controlled.

As to PWM frequency, the impression I get, and from my limited experience with DC motors, finding the right frequency seems a matter of experimenting to see what suits, finding what allows the motor to spin smoothly without creating any whine or overheating.

The higher the period, the lower the frequency, the wider range of control one has over the duty, so one is probably looking to achieve that.

Besides the internal electronics to make the motor turn (as Hippy reminds us), my first question is, "Why do you want to vary the speed?"

I would use the tach output to synchronize the stream of data for the display, and run the motor with a well regulated power source to maintain a constant speed.

Reading further in the datasheet "7. Do not use power or ground PWM to control the fan speed. If the fan speed needs to be adjusted, please contact CUI to customize the product design for your application." so it seems like Plan B is needed.

Note to self - read all the datasheet first.

cpedw said:

Note to self - read all the datasheet first

Click to expand...

Yep; I missed that too.

Despite that 'do not' it might be worth a try anyway. I experienced the problem in a commercial product ( not Rev-Ed ) so it had to be fixed right but it may be good enough, and perhaps with just a reservoir cap across the supply.

techElder said:

I would use the tach output to synchronize the stream of data for the display, and run the motor with a well regulated power source to maintain a constant speed

Click to expand...

I did think about that and thought a mains synchronous motor might fit the bill. But I expect the problem predicted is that it's not easy to synchronise the output stream to the motor rotation with the fine degree of control one might like.

I must admit I hadn't done the maths or even thought about how it could be done, so it may be more feasible than it at first seems. SPI or synchronous UART output would allow a fairly steady stream of output fired from a synchronisation pulse so, using control of the baud rate, that may be one route.

And on synchronous mains motors; how do record turntables adjust their speed ? Maybe that is PICAXE controllable ?

Does it have to be that actual fan?
There are many 80mm fans available with PWM control and tacho output at a reasonable price. The Picaxe as a simple closed-loop controller would give you full speed control.

Perhaps see if you can buy (borrow?) the Elektor articles that describe the POV Clock printed around six years ago. That uses a diode sensor (IR?) to detect the start of a rotation and uses that to sync the updating of the display. The article(s) could give you some ideas for your design?

Thanks for those suggestions. I''ve already got that motor and I think it can be made to work. If it doesn't, I will investigate alternatives as suggested.