[Jim Dawson]

Ahhhh, Look at the top of page 10. J3, Pins 1 & 2 that may be why the speed is slower.

[drhanger]

Interesting find. I'll have to look at that this afternoon. So if those two pins are indeed connected, either I need to disconnect them to see if that brings the speed to parity, or they need to be connected and I need to make an adjustment somewhere, but can't really tell where, maybe one of the pots mentioned in the adjustments section. Is that how you would interpret that?

[Jim Dawson]

That is the same conclusion that I came to. There is a reference to the AUX pot when using those terminals, maybe that's the answer. Page 13

[Marty_Escarcega]

Interesting find. I'll have to look at that this afternoon. So if those two pins are indeed connected, either I need to disconnect them to see if that brings the speed to parity, or they need to be connected and I need to make an adjustment somewhere, but can't really tell where, maybe one of the pots mentioned in the adjustments section. Is that how you would interpret that?

Curious to know what you find out.

Marty

[drhanger]

Looks like a dead end. The two pins on J3 were not connected, when I did connect them there was a noticeable slow down of the spindle, but the AUX pot did nothing for it, pins connected or not. I don't think this is gonna go where I was hoping to.

[Marty_Escarcega]

Your only option is to build the circuit I did that will invert the 0-10VDC signal with a 5VDC input signal. Its not complex. I'll try and dig up the schematic and parts list for you.

[drhanger]

Marty, that would be great. I'm no electronics expert but I've done a fair amount of small board making and a lot of soldering, so it shouldn't be a big challenge for me. I've used Eagle board layout software a number of times and built small signal distribution boards for my cnc router and also for this mill. It's a fun project for me. Thanks, look forward to seeing that.

[drhanger]

Marty, any luck scaring up the info for that reversing circuit?

[Marty_Escarcega]

Marty, any luck scaring up the info for that reversing circuit?

Sorry Dan, I have been swamped.....
See attached. Review thoroughly, use at your own discretion. This is similar to what I used on my DM4400, however, I added a relay and fixed voltage for a consistent spindle orient for my ATC.

Do note, you need a +12V and -12V power supply. Meanwell makes on that will supply these voltages, +5VDC and the +24VDC



Marty

[drhanger]

Marty-

Thanks a bunch, I really appreciate it. So let me see if I understand how this goes:

1. The two outer connections are the +/- 12V inputs from the PSU. I am still using the original PSU's that came with the machine, one of them has 5V and +/- 12V terminals, so I should be set there.

2. 0-10V IN -- This is where I would connect the PWM signal I'm already using from my LinuxCNC controller. This would be the actual speed control signal. Currently I am feeding this signal into the MX3660 BOB, which outputs the correct voltage to pin 2 of the spindle driver, so your circuit replaces that part.

3. DIR -- this would control the spindle direction depending on whether the signal is low or high. I assume a high signal would be for forward (+ output to pin 2)? I have to think about how I would implement this in Linux....

4. 10V OUT -- This is the final output signal to pin 2 of the spindle driver, modulated by the microprocessor to be either positive or negative with a varying voltage for speed.

Have I got it figured correctly? I just realized that I will need an additional output signal from LinuxCNC for the directional signal, which I don't have. LinuxCNC supports an additional add-in parallel port card just for adding i/o's, I've attempted this in the past but had some problems with Linux recognizing the add-in card, so I'll have to work my way through that before I can proceed with this part of the project.

I'll let you know how it turns out once I can get to it. Thanks again.

Dan

[Marty_Escarcega]

Marty-

Thanks a bunch, I really appreciate it. So let me see if I understand how this goes:

1. The two outer connections are the +/- 12V inputs from the PSU. I am still using the original PSU's that came with the machine, one of them has 5V and +/- 12V terminals, so I should be set there.
YES

2. 0-10V IN -- This is where I would connect the PWM signal I'm already using from my LinuxCNC controller. This would be the actual speed control signal. Currently I am feeding this signal into the MX3660 BOB, which outputs the correct voltage to pin 2 of the spindle driver, so your circuit replaces that part.
YES, your analog spindle control signal

3. DIR -- this would control the spindle direction depending on whether the signal is low or high. I assume a high signal would be for forward (+ output to pin 2)? I have to think about how I would implement this in Linux....
YOU need an output to turn on when M4 is called, this will supply +5VDC to the input and invert the 0-10VDC signal

4. 10V OUT -- This is the final output signal to pin 2 of the spindle driver, modulated by the microprocessor to be either positive or negative with a varying voltage for speed.
This goes to the SPINDLE DRIVE input requiring -/+ 0-10VDC control input (The SD3060)

Have I got it figured correctly? I just realized that I will need an additional output signal from LinuxCNC for the directional signal, which I don't have. LinuxCNC supports an additional add-in parallel port card just for adding i/o's, I've attempted this in the past but had some problems with Linux recognizing the add-in card, so I'll have to work my way through that before I can proceed with this part of the project.

I'll let you know how it turns out once I can get to it. Thanks again.

Dan

My responses below yours above. There are likely more elegant ways to do this but its down and dirty and works for me. Jim Dawson might have other suggestions....he's a sharp cookie.
Centroid Acorn supports 8 inputs and 8 outputs and is a step and direction motion controller. Currently will not support your ATC until they add more I/O. I'm very happy with it and will never go back to other control software as long as Centroid CNC Acorn is around and they continue to improve on it.
Hardware AND software from same manufacturer.

Good luck Dan.

[Jim Dawson]

Thanks for the vote of confidence Marty. I really don't have anything to add, I have never tried to convert PWM/DIR to analog. But I am watching this thread and learning.

[drhanger]

Jim, I'm glad I'm not the only one that might benefit from this.

I've ordered the transistor and microprocessor from Amazon, I should have resistors on hand. Prolly have to stop at Fry's Electronics one day for the caps. I hope to be able to breadboard this together for testing this weekend. I plan to just manually apply the 5V signal to the DIR pin for now, if all works well then I'll have to work on the additional output I'll need. Also design and cut a circuit board of course.

[drhanger]

Marty-

Your BOM lists the transistor as 'NP', apparently there is a P missing. Do you remember which, PNP or NPN?

Dan

[Marty_Escarcega]

It also lists the part number 2N2222 which is an NPN transistor. Dirt cheap.

[drhanger]

Hey Marty-

I finally got that breadboard circuit put together and tested it out just now. It works as expected generally, however in reverse the spindle is somewhat erratic. The driver makes a 'loping' sound and the tach was kinda unsteady. I checked the output voltage at the +10V out of your circuit--in forward the output was rock solid, but in reverse the output was erratic. I checked the voltage at the PSU (+/- 12VDC) and they were both steady under power, so I think the PSU is fine. There is one difference I made in your circuit based on availability of parts at Fry's Electronics. I was able to find the 100 uF electrolytic caps, but they had none of the 0.1uF in electrolytic, only in ceramic. Do you think this might be the reason? Doesn't make sense since forward is solid, but I really don't know anything about it.

[Marty_Escarcega]

Hey Marty-

I finally got that breadboard circuit put together and tested it out just now. It works as expected generally, however in reverse the spindle is somewhat erratic. The driver makes a 'loping' sound and the tach was kinda unsteady. I checked the output voltage at the +10V out of your circuit--in forward the output was rock solid, but in reverse the output was erratic. I checked the voltage at the PSU (+/- 12VDC) and they were both steady under power, so I think the PSU is fine. There is one difference I made in your circuit based on availability of parts at Fry's Electronics. I was able to find the 100 uF electrolytic caps, but they had none of the 0.1uF in electrolytic, only in ceramic. Do you think this might be the reason? Doesn't make sense since forward is solid, but I really don't know anything about it.

Dan,
I built several of these and never had the problem you described.
Did you substitute the caps or didn't use them at all? They are small ceramic...

[drhanger]

I substituted the caps with the ceramic ones. As a test to eliminate the spindle driver as the problem I bypassed the input to the driver with a battery and changed polarity back and forth. I wasn't able to get the same erratic behavior that way so I think maybe I just don't have the circuit working quite properly, maybe the breadboard I'm using is not quite right. Well, even if I don't get the problem cleared up it will probably still work for what I want, which is just being able to tap with a tapping head. I have a floating tap head, and the change in spindle speed is so minor that I'm sure the floating head will make up for it for the few seconds of reversing the tap. I'm going to go ahead with making a proper circuit board for it. I still have to add another output to my control using the additional parallel port card I mentioned before, so once those two things are completed I should be good to go. Thanks for your help.

[Marty_Escarcega]

I substituted the caps with the ceramic ones. As a test to eliminate the spindle driver as the problem I bypassed the input to the driver with a battery and changed polarity back and forth. I wasn't able to get the same erratic behavior that way so I think maybe I just don't have the circuit working quite properly, maybe the breadboard I'm using is not quite right. Well, even if I don't get the problem cleared up it will probably still work for what I want, which is just being able to tap with a tapping head. I have a floating tap head, and the change in spindle speed is so minor that I'm sure the floating head will make up for it for the few seconds of reversing the tap. I'm going to go ahead with making a proper circuit board for it. I still have to add another output to my control using the additional parallel port card I mentioned before, so once those two things are completed I should be good to go. Thanks for your help.

Suggest you fix the circuit.
I thought you were going to route a PCB?

[Meri]

Just got an old DM4400 but it does not have original power supply under base, actually has no base. Machine sat outside local machine shop for years. I jumped the internal power supply from 240v to 120v and hooked up 120v and the x,y,z axis move fine. The servo amplifier is also missing as is the door. Think it fell off a truck as beat up cabinet.

Plan on replacing the servo amp with a Bardac 3200i board I have.
There is a set of wires that ran from the original base power supply #3 on page 7-1 of the original owners manual to the bus bar #5 on page 7-2.
The white wire on pin 8 is 220v, pin 9 black wire is 110v. Can anyone please tell me what the voltages should be on the other 7 wires?

Thanks guys.
Don