[doanwannapickle]

On the contrary, the motor is excellent. Lots of torque, rpm range right where I want it and light and compact.
It's the controller. It's optimized for a sewing machine. Original control was from a foot pedal connected to a hall sensor. The voltage range is in the order of 2.4v to 4.8v. Pretty tight.
It's easy to replace that with a 5 turn pot and get good smooth transition thru its rpm range.
But, I'd like to control it with the 0 to 10v pwm signal from Mach3. That's where the fun begins. Converting one voltage range to another.
There's a thread on a similar motor here: https://www.cnczone.com/forums/gecko...ine-motor.html
I made the circuit he came up with and had moderate success. I'll probably go back to that and settle for whatever I can get.

Out of curiosity, what motor are you using that you control with that new controller?

[CitizenOfDreams]

Out of curiosity, what motor are you using that you control with that new controller?

The "400W BLDC spindle" from eBay. Works pretty good on a small CNC machine.

[CitizenOfDreams]

But, I'd like to control it with the 0 to 10v pwm signal from Mach3. That's where the fun begins. Converting one voltage range to another.

I believe this is what your problem is: converting a low frequency PWM signal to voltage. Here is the circuit I made for this purpose; it gives me an acceptable trade-off between ripple and response time. A better way to do it would be a microcontroller.
The simulation waveform shows the conversion of a 50Hz 50% square wave signal to DC.

[doanwannapickle]

[QUOTE=CitizenOfDreams;2238642]The "400W BLDC spindle" from eBay. Works pretty good on a small CNC machine.

Those are really nice for high speed stuff. My motor maxes out at 4500 and has a ton of torque. Just right for me.

[doanwannapickle]

A better way to do it would be a microcontroller.

That, I believe, is exactly the best solution. Unfortunately, I have no clue how to make that happen.

[CitizenOfDreams]

There are some BLDC drivers similar to ours that take PWM input. I had one of those (until it blew up), it worked fine with Mach3 without any additional voltage conversion. It had a DIP switch that changed the input mode from "Voltage" to "PWM".

[CitizenOfDreams]

That, I believe, is exactly the best solution. Unfortunately, I have no clue how to make that happen.

On the firmware side, you measure the input signal duty cycle, average it across a few periods, then generate a high frequency PWM signal with the same duty cycle.
On the hardware side, you filter the microcontroller output and buffer it. Same exact circuit I attached to post #23, but the time constant can be much lower (faster response to duty cycle changes).

[doanwannapickle]

The controller I bought took 0-10v pwm input. That's why I bought it and wanted to make it work. Plugs right in to my G540. It's just not compatible with this motor.

[Al_The_Man]

I'm a bit confused, Are trying to control with a PWM signal or the 0-10v analogue signal such as G540 etc?
For 0-10v analogue there are plenty of these kind of controllers on ebay etc such as the A-M-C versions, also Aerotech and Copley etc. for BLDC motors.
Al..

[doanwannapickle]

I'm a bit confused, Are trying to control with a PWM signal or the 0-10v analogue signal such as G540 etc?
For 0-10v analogue there are plenty of these kind of controllers on ebay etc such as the A-M-C versions, also Aerotech and Copley etc. for BLDC motors.
Al..

I'm connecting to the VFD +10v/VFD Gnd/VFD Output on the G540. I was under the impression that that voltage varied in response to the Mach3 pwm signal.
As far as the controller is concerned, the one I started this thread with uses that 10 volt input. It just won't drive this motor.

[Al_The_Man]

Personally I would not use any Chinese origin motor controllers of any type.
The A-M-C drivers are plentiful usually on ebay and are quite capable to drive a BLDC motor with halls and supply the 5v for them.
You just need the max motor voltage and operating current for selecting one.
Al.

[doanwannapickle]

You just need the max motor voltage and operating current for selecting one.

Of which I have neither.

[doanwannapickle]

On the firmware side, you measure the input signal duty cycle, average it across a few periods, then generate a high frequency PWM signal with the same duty cycle.
On the hardware side, you filter the microcontroller output and buffer it. Same exact circuit I attached to post #23, but the time constant can be much lower (faster response to duty cycle changes).

That's mostly a bit beyond my electronics capability. Do you have that schematic in a little more readable format? PDF? DOC? DXF?
I wouldn't mind studying on it a bit and try to figure it out.

[CitizenOfDreams]

Do you have that schematic in a little more readable format?

Attached is a better quality file, although you don't need this schematic if your controller already has PWM to 1..10V converter.

U2 amplifies the input signal to 10 volts peak-to-peak.

R1-C1 and R2-C2 is a second order RC filter.

U1 is the unity gain output buffer.

[doanwannapickle]

Good Morning Citizen,
Let me back up and clarify where I'm at on this. Humor me.

The original controller that came with the motor does a good job driving the motor. It spins it to 4500 rpm with good torque through its range.
BUT, its input is in the range of 2.5v to 4.8v.

The Gecko G540 that I'm using has three terminals specifically geared toward variable frequency drives. They are +10v, Gnd and Output. You provide the the 10 volt source.

This 10 volt output from the Gecko obviously isn't very compatible with that controller's input.

So, this is where the new controller that I started the thread with comes in. It's laid out like a VFD in that it has the +10v source for the Gecko and its input likes the output from the Gecko.

Sadly, that controller simply doesn't have the right stuff for that motor.

So, I'm back to trying to make the original controller work. That means taking that 0-10v output from the G540 and turning it into the 2.5-4.8v for the controller.

BTY, thanks for all the help you've been.

[CitizenOfDreams]

So, I'm back to trying to make the original controller work. That means taking that 0-10v output from the G540 and turning it into the 2.5-4.8v for the controller.

There are two ways you could approach this.

The software way: edit Mach3's Linearity.dat file (takes a special editor).
The hardware way: make an opamp circuit that divides the input signal by 4.35 and adds 2.5 volts to it.

[doanwannapickle]

The software way: edit Mach3's Linearity.dat file (takes a special editor).

Sounds interesting. Where do I get this editor?

The hardware way: make an opamp circuit that divides the input signal by 4.35 and adds 2.5 volts to it.

Even more interesting but beyond my electrical skills. If you want to sketch something up, I'll be happy to put it together.

[CitizenOfDreams]

I found the editor somewhere on the Web. Don't know who made it, can't guarantee anything, but it worked for me. Use at your own risk.

As for the circuit, my mouse died yesterday, so I can't draw much. Will get a new one tomorrow. In a word, it takes a summing amplifier, one of the basic opamp circuits.

[doanwannapickle]

Cool, thanks.

[CitizenOfDreams]

Here is a circuit that converts 0..10V to 2.5..4.8V. It is not (too) sensitive to the supply voltage: anything from 6 volts up will do.

R1-R2 is a voltage divider that divides the input signal by 4.35.
U1 is a buffer.
R4-U3 is a 2.5V voltage reference.
R3-R5-R6-R7-U2 is a summing amplifier that adds 2.5V to the signal.
Probably would not hurt to add an output buffer, but it should work as it is.