I want to control the speed of a DC motor which runs between 12v and 48v. Its a 300W motor, and I'm assuming that means 300W at 48v so 6.25 amps. My power supply is 48v 1200w, the only way i can think of is to put together a dc to ac converter, and get a variac, then put bridge rectifier on the end. But is there an easier way?
That's a pretty complicated and very expensive way to go about it. What about using a 10A adjustable switching regulator?
Edit: What type of supply is your 48V, 1200W PSU? Is it regulated at 48V? ie Is it likely to go higher than 48V at any time?
Is that 300W mechanical output power? If so the electrical input will be higher, 300 / motor efficiency.
Assuming efficiency of 85% that would be 300 / 0.85 = 353 Watts / 48 volts = 7.35 Amps.
Is your motor going to be reversible?
Normally, you adjust the speed of a DC motor with [u]PWM[/u] (the same way you dim an LED with the Arduino). With PWM you are switching the DC voltage on & off rapidly to vary the average voltage. In practice, this turns-out to be a LOT easier than linearly varying a DC voltage.
With the Arduino, you'd just need to add a MOSFET (and a resistor) to handle the voltage & current. And, you'd need a pot if you want a manual control.
You don't need a programmable microcontroller to generate a variable duty-cycle pulse. You might be able to use a 555 timer. But, it's probably easier to do it in software.
OldSteve:
That's a pretty complicated and very expensive way to go about it. What about using a 10A adjustable switching regulator?Edit: What type of supply is your 48V, 1200W PSU? Is it regulated at 48V? ie Is it likely to go higher than 48V at any time?
It's totally regulated, made for powering sever computers.
DVDdoug:
Normally, you adjust the speed of a DC motor with [u]PWM[/u] (the same way you dim an LED with the Arduino). With PWM you are switching the DC voltage on & off rapidly to vary the average voltage. In practice, this turns-out to be a LOT easier than linearly varying a DC voltage.With the Arduino, you'd just need to add a MOSFET (and a resistor) to handle the voltage & current. And, you'd need a pot if you want a manual control.
You don't need a programmable microcontroller to generate a variable duty-cycle pulse. You might be able to use a 555 timer. But, it's probably easier to do it in software.
PWM, I forgot about that. But don't dc motors have a bounce back when they first turn off? Like an inductor, the fly back voltage spikes. Would this high frequency on/off cause any wear or damage?
That's generally why you use protection circuitry around the motor and why many simple DC IC motor controllers are 75% made up of that type of circuitry.
I agree with those suggesting a conventional motor controller (aka h-bridge).
Parallax sells a nice (though pricy) 25A h-bridge which is controlled using servo control pulses. (This makes it very easy to control with a microcontroller.)
https://www.parallax.com/product/29144
I'm using a pair of these to control two 12V motors in one of my robots.
Pololu sells a wide variety of h-bridges. I've always been positively impressed with my dealing with Pololu.
I think you're asking for trouble attempting to control the motor by using a voltage regulator.
Check out the P-channel logic level mosfets on Sparkfun as well if you plan to build an H-bridge.
V = 60V
I = 32A
Do the math
More than adequate for a 300W motor.
I built a DIY h-bridge with those parts plus a TC4427 chip.
See Reply#6 of this post.
DuaneDegn:
I think you're asking for trouble attempting to control the motor by using a voltage regulator.
A variable-speed motor controller is a variable switching regulator. I was merely suggesting that the long route of DC to AC to DC was not needed. And when I said 10A switching regulator, I didn't mean an off-the-shelf one, but a DIY circuit using PWM.
I object to calling a PWM motor controller "a regulator" on the grounds that "a regulator" , to me , means a black box that has an output that looks like a steadfy state voltage , and not a PWM signal. The LM7805 is the single device that defines the meaning of the word "regulator" because that word has been used more times to desribe the LM7805 than any other device. When I look at the output of the LM7805 with a scope I don't see a PWM signal.
When I think of a motor controller I think of a VFD. Like a Danfoss FC302-37 kW)
raschemmel:
I object to calling a PWM motor controller "a regulator" on the grounds that "a regulator" , to me , means a black box that has an output that looks like a steadfy state voltage , and not a PWM signal. The LM7805 is the single device that defines the meaning of the word "regulator" because that word has been used more times to desribe the LM7805 than any other device. When I look at the output of the LM7805 with a scope I don't see a PWM signal.
When I think of a motor controller I think of a VFD.
Objection noted. Bad choice of words on my part.
I was thinking more in terms of the fact that at the heart, they're the same thing. Both are switching DC-DC converters, although it is true that a switching 'regulator' has better output filtering.
More often than not, PWM is also filtered to make it useful.
And comparing only a linear regulator was stretching it. Switching DC-DC, (PWM), buck converters are also regulators.
Switching Regulator
Thank you for the help everyone. I will be using a simple PWM controller with arduino controlled pulse width, and big caps to protect from the fly back.
Last time I looked at the output of an ebay DC-DC converter , the output was a steady state voltage.
Like I said , a "black box" can have anything inside, including a switching regulator. And just for the record, I use an OKI 5V switching regulator, not an LM7805. (I used LM7805 as an example because the OKI was not available when I started learning electronics in 1979)
and big caps to protect from the fly back.
"FLYBACK" is typically handled by DIODES
raschemmel:
Last time I looked at the output of an ebay DC-DC converter , the output was a steady state voltage.
Like I said , a "black box" can have anything inside, including a switching regulator. And just for the record, I use an OKI 5V switching regulator, not an LM7805.
"FLYBACK" is typically handled by DIODES
I did mention the main difference was in the filtering. Actually, the other big difference is that a regulator uses voltage-sensing feedback, whereas a decent motor controller uses speed-sensing feedback, (often measuring the motor's back-EMF voltage).
Anyway, I'm not interested in getting into long, drawn-out arguments about precise definitions.
My whole point from the beginning, was that griffin175 should use direct DC to DC conversion rather than DC to AC to DC, and he's doing that now, so all's well that ends well.
He's doing that now because besides being "drawn out" , what we do is also educational.
raschemmel:
He's doing that now because besides being "drawn out" , what we do is also educational.
True.
And now back to my project. As much as I like talking about electronics, I'd much prefer to be doing it. 
And now back to my project
Which is ?
raschemmel:
Which is ?
Just a little robot car. (Nothing exciting)
Do you have onboard realtime video with a 5.8 Ghz video transmitter ? ( I used a GOPRO) . It has composite video out which is compatible with the video transmitter. The only way to see the HD is to plug in the component cable that comes with it so that's no good for transmitting but
the composite is excellent quality. I had a GPS module with OSD (On screen display) but of course the GPS only works outside.
raschemmel:
Do you have onboard realtime video with a 5.8 Ghz video transmitter ? ( I used a GOPRO) . It has composite video out which is compatible with the video transmitter. The only way to see the HD is to plug in the component cable that comes with it so that's no good for transmitting but
the composite is excellent quality. I had a GPS module with OSD (On screen display) but of course the GPS only works outside.
No, I don't need video. It only cruises around the room, and I can see everything in front of it anyway.
I only just got the basics together, and am working out bugs. ie The US ping sensor is fixed, so the car sometimes runs into walls when it approaches on an angle and the 'ping' reflects off in another direction. Also not so hot at detecting thin chair legs and objects really close to the ground. At the moment, I follow it with a long stick and wave it in front of the robot if it looks like hitting something.
The next step is to mount the ping sensor on a servo and swing it back and forth, then have the robot travel in the direction with the longest clear run. I can see myself adding another one or two ping sensors, too, but for now I'll just get the front one swinging.
It's all good practice for using the Arduino. I've only had my UNOs for about 5 weeks, so I need all the practice I can get.