• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need help controlling 110vDC motor

Status
Not open for further replies.

Belphgor

New Member
Hello everyone,

I've searched the archives for an answer for this, and although I didn't find any, I did see a lot of interesting future projects :)

I have a Bench top Mill. It's small, but I can use it to mill out printed circuit boards, metal parts for RC models, and plastics.My problem is with plastics. At full speed the plastic melts and gums up my tool. At slower speeds the motor stalls and the tool breaks.

I have a Ubuntu program that controls the mill, EMC2. I put an encoder on the motor to help control the speed, but the speed really fluctuates. I put an oscilloscope to watch the control pulses. They were definitely trying to do something. Then I put the scope on the voltage to the motor. I was surprised at what I saw. It's a DC motor, but it's powered through a bridge rectifier and controlled by a triac. Since the computer program can't reference the zero crossing, well, that's the problem I believe.

What I want to do is replace the bridge and triac with a DC power supply and use a IGBT to control the motor. I ordered the parts and wired it together. Just a bridge and capacitor, I shouldn't need anything more elaborate should I? I've got a little more ripple then I expected, but more capacitance should take care of that.

My question/problem has to do with the transistor. It's not turning on. I have a opto-coupler with the collector tied high, and the emitter bringing up a voltage divider tied to the base of the IGBT. The data sheet says it needs 20 volts (emitter to base) to conduct. Do I need a current limiting resistor in the base? I just don't have any experience working with them.

Thanks and my best regards
 

bobledoux

Member
Please share your results with us.

I've been working on the same problem. I have a 130 volt permanent magnet motor to power a metal lathe. It originally had a treadmill controller which burned out.

My idea was to take a low cost controller designed for routers or electric drills. I was going to run the output through a full wave bridge and smoothing capacitor.

Web searches turn up little on this project.
 
Last edited:

tcmtech

Banned
Most Helpful Member
Post a schematic of what you have so far.
There are dozens of ways of controlling the speed of a DC motor being run from an AC source.

If all your after is slowing down the speed I would use a variac transformer and the bridge rectifier. It will give you full range speed control while maintaining the constant torque of the motor.
However your motors actual output power is drectly proportional to its speed.
If you cut the speed in half your total power output is also cut in half.
 

bobledoux

Member
Are you sure a Variac gives as much torque as PWM?

I'm currently using a variac and bridge rectifier. How is that different than just running DC through a series voltage drop resistor? I thought the reason for PWM was to retain torque at low rpm's.
 

Ubergeek63

Well-Known Member
first, the spec is not how much it takes to turn on the IGBT, it is how much is safe to apply.

second, IGBTs do not have bases they have gates that love to see over an amp for under a micro second for each transition.

third, triac drives are difficult but not impossible to get to work at low speeds.

start with a PWM IC. filter the motor voltage and apply it to the feedback pin instead of an output voltage. apply a speed command voltage to the feedback pin as well.

that will work much better but will still bog down a bit at low speed due to the voltage drop across the armature resistance. this can be sensed and fed into the feedback junction as an error term as well.

This is EXCACTLY how simple treadmill motor controls do it.

Dan
 

Belphgor

New Member
Thanks for the responses everybody.

I don't want to just slow down the motor, I want to control the speed. I have feedback through a 1024 line encoder to a PC, and the PC outputs the pwm. What I want to do , and what I'm having trouble with, is controlling the motor. I've attached a schematic of what I'm trying to do. The opto-coupler in the upper right is driven by a transister with 24 volt supply. This transister is controlled by the PC.

Yes, I said base, I meant gate. Call it a typo. :)

Maybe the IGBT isn't the best device for this. If the gate needs an amp to turn on, it's going to be drawing as much power as the motor. I've not been able to find anything that mentions anything about the IGE

Maybe a Hexfet or something.

Regards,

Bill
 

Attachments

Ubergeek63

Well-Known Member
Thanks for the responses everybody.

I don't want to just slow down the motor, I want to control the speed. I have feedback through a 1024 line encoder to a PC, and the PC outputs the pwm. What I want to do , and what I'm having trouble with, is controlling the motor. I've attached a schematic of what I'm trying to do. The opto-coupler in the upper right is driven by a transister with 24 volt supply. This transister is controlled by the PC.

Yes, I said base, I meant gate. Call it a typo. :)

Maybe the IGBT isn't the best device for this. If the gate needs an amp to turn on, it's going to be drawing as much power as the motor. I've not been able to find anything that mentions anything about the IGE

Maybe a Hexfet or something.

Regards,

Bill
You have an opto that will stand up to 200V?

At what frequency are you attempting to PWM it? Neither the PC or opto is likely to be able to do it. At normal PWM frequencies "standard" speed IGBTs would not either since they are rated for only 1KHz hard switching.

FETs and IGBTs REQUIRE over an amp instantaneous to switch properly. Gate drivers are typically very small parts since they are only required to supply the rated current while the device capacitance is being charged or discharged. Average current required can be determined by multiplying the "total gate charge" spec by the operating frequency.

Dan
 

tcmtech

Banned
Most Helpful Member
Maybe the IGBT isn't the best device for this. If the gate needs an amp to turn on, it's going to be drawing as much power as the motor. I've not been able to find anything that mentions anything about the IGE
An IGBT is a voltage controled device! Unless your running a 1200 volt 600 amp IGBT at 40khz consider it a voltage controled device ZERO amps Gate current.
Also IGBTS typicaly are at full on by under ten volts. The 20 volt number is the internal zener diode protection voltage. ;)

Still for a very basic motor control like this just use a Variac and set it at a lower voltage. I do this alot when I need my dremmel to run slow. PLus I do it to my angle grinders all the time when I dont need the full 10000 RPM.

Trust me it works! The PWM control circuits work great if you are using a feedback loop in order to control the motor speed for a acurate RPM or apsolutely need to be able to control the motor speed from a controller of some type.:)
 

Belphgor

New Member
OK, the IGBT I'm trying to use is IRGB4B60KD1. I'm not married to it, just kind of picked it out. I'm pretty sure that I've got it connected right, at least according the the drawing I attached to the earlier post.

I do want to stick with the PWM. The IGBT/MOSFET driver Transgenesis mentioned sounds interesting. I'm going to look into those too.

I've also got some formulas that I can plug some numbers into to find gate resistance and stuff.

Thanks again for the ideas everybody.

Bill K.
 

Ubergeek63

Well-Known Member
at what current?

check the circuit... that IGBT is only good for 10-30KHz and the opto is likely to only be good for 50V and 3KHz
 

tcmtech

Banned
Most Helpful Member
Can your CNC mill controller output be reprogramed for just a variable frequency output? That is if the frequency goes up it would be telling a motor to speed up?

If so, I have an idea! If not, then never mind!
 

Belphgor

New Member
According to my calculations, the motor draws about 1.6 amps at full speed.

Right now I'm investigating a mosfet driver. From what I've been reading, IGBT's are recommended for 600 volts and over, mosfet's for less then 250 volts.

The program I'll be using to control the drive / motor has a number of ways to do it, with either PWM where I can set the frequency. or PDM what they call pulse density mode where the computer sends out a pulse of set width, less then a millisecond, but then increase the frequency to control the speed. The computer is pretty busy, counting pulses from the encoder, and sending out pulses for this motor, and also two axis. The third axis, "Z" doesn't move that often. I think it's pretty amazing just to watch it "do stuff". Sometimes it really gets busy.

Again my thanks for everyone who has suggested ideas. It's nice to talk to people who understand what I'm talking about. Even if I don't always :)
 

Ubergeek63

Well-Known Member
smaller pulse widths, either the ON state or OFF state, load down a CPU. as I said before, you need to compare the frequencies with the speeds of your parts.

you also need to supply a separate gate supply since neither the opto or gate can handle line voltage, which I implied earlier.

that can all be done readilly on an 8051 with power to spare. on a $5 ARM you could tell it to draw a circle and it could control all three motors itself. Heck on the ARM I could interpret the whole gerger photoplot file.
 

Belphgor

New Member
Yes, I understand about the frequency. If I go the PDM route I'll measure the length of the pulse. If, for example it's 1ms, the frequency would be 1kHz. Same for PWM, the minimum time, either high or low, would be the frequency. I don't think higher frequencies would control a motor any better then lower, just generate more EMF.

As far as the voltage, I've got a small variac that I can use to lower the voltage to 100 volts DC or so. That should make it a little easier to find a driver and mosfet combo.

One thing I liked about the opto was it allowed me to isolate the higher DC from the rest of the circuit. Maybe it doesn't make a difference. I don't know.

I don't know what you mean by the ARM. Is the 8051 a chip made by Intel from the early 80's? I was involved more with Motorola µPU back then.
 

smanches

New Member
Here are the optoisolators I'm using in my projects. They work great at 160khz, the fastest I've tried to drive them so far.

http://www.electro-tech-online.com/custompdfs/2009/04/6n137vo2.pdf

Which in turn drive these...

http://www.electro-tech-online.com/custompdfs/2009/04/21424d.pdf

Which in turn drive these...

http://www.st.com/stonline/products/literature/ds/13154/stp20nf20.pdf

The entire project is a h-bridge control for an ACIM blower. It's working so far on the scope, but I have yet to actually attach the blower to it. Hopefully this weekend. :)
 
Last edited:

Ubergeek63

Well-Known Member
I don't know what you mean by the ARM. Is the 8051 a chip made by Intel from the early 80's? I was involved more with Motorola µPU back then.
The ARM architecture is being produced by numerous manufacturers. I mostly use NXP LPC2134s.

The same chip with more on it is made by a large number of companies. I am most familiar with the current generation from NXP, the P89V51RD2 that has all the 51's stuff plus timer T2 and a programmable counter array (PCA) timer that can do captures, compares, and PWMs as well with a single counter and 5 channels.

Dan
 

causalitist

New Member
Please share your results with us.

I've been working on the same problem. I have a 130 volt permanent magnet motor to power a metal lathe. It originally had a treadmill controller which burned out.

My idea was to take a low cost controller designed for routers or electric drills. I was going to run the output through a full wave bridge and smoothing capacitor.

Web searches turn up little on this project.
i got a 130vdc 3.5hp treadmill motor (must have been a big treadmill!)
i run it at 148vdc from lithium polymers. but a full wave bridge and a few big caps is the same.. this powers a bicycle.

i used a sg3524 ... almost half the pins are grounded out or connected together, so its really easy circuit.. i use a murata dc-dc 1watt converter to get my 12v (few bucks) for the chip... but u guys got AC so whatever... :)

anyway... all u need is a few caps , a few R's and a trimpot for throttle,.. then output to a tc4422 driver... then output that to whatever.

igbt u may want negative gate volts at turn off, so go with a driver like td351 that has active miller clamp.... or actually use negative volts.

i bet the schematic is on this site lol..
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top