South Western Industries TRAK DPM All in One DC Conversion

[martyscncgarage]

I have a client that would like his new to him TRAK DPM converted to Centroid. The machine has a late 90's vintage A.G.E. 3 Control
Google Album here:

A few questions

1. The machine has Acu-Rite Non-Absolute Scales on X/Y/Zquill. He would like to reuse them with Centroid. Does anyone know if these scales output the correct differential signals? Pin outs?

2. Since he won't be using a spindle encoder, can I assign the 3 encoder (4/5/6) inputs for the scales?

3. The Z axis is of course the column along with the scale on the quill. Currently, you can adjust the Z column wherever you like and when you manually operate the quill, the A.G.E. 3 control will deduct/add the quill value from the Z axis DRO. Does CNC 12 do the same?

4. I would assume the DC servo motor encoders are still needed with glass scales. Are the original SWI TRAK encoders usable or do they need to be replaced with differential encoders? (I have not gotten into the end of the servo motor yet)

5. Interestingly, the A.G.E. control servo driver is attached to the side of each DC Brush servo. Says 115V. I assume that voltage is being rectified at the servo motor and I will need a cap board low for the DC Buss voltage, which will rectify the 115VAC.

Any other gotchas I should be aware of before I start ordering parts and building the All in One DC Control?

Thanks in advace.

[centroid467]

I have a client that would like his new to him TRAK DPM converted to Centroid. The machine has a late 90's vintage A.G.E. 3 Control

A few questions

1. The machine has Acu-Rite Non-Absolute Scales on X/Y/Zquill. He would like to reuse them with Centroid. Does anyone know if these scales output the correct differential signals? Pin outs?

2. Since he won't be using a spindle encoder, can I assign the 3 encoder (4/5/6) inputs for the scales?

3. The Z axis is of course the column along with the scale on the quill. Currently, you can adjust the Z column wherever you like and when you manually operate the quill, the A.G.E. 3 control will deduct/add the quill value from the Z axis DRO. Does CNC 12 do the same?

4. I would assume the DC servo motor encoders are still needed with glass scales. Are the original SWI TRAK encoders usable or do they need to be replaced with differential encoders? (I have not gotten into the end of the servo motor yet)

5. Interestingly, the A.G.E. control servo driver is attached to the side of each DC Brush servo. Says 115V. I assume that voltage is being rectified at the servo motor and I will need a cap board low for the DC Buss voltage, which will rectify the 115VAC.

Any other gotchas I should be aware of before I start ordering parts and building the All in One DC Control?

Thanks in advace.

Hello Marty,

1. Those look like Acu-Rite Mini-Scale or Mate encoders (see manual here: https://acu-ritesolutions.com/wp-conten ... Manual.pdf). You'd have to verify by checking the part numbers. That manual says that the Mini-Scale models use single ended 5V TTL quadrature signals while the Mate models do differential sine/cosine. It is not likely that they will directly work with the ALLIN1DC.

2. Yes, there's nothing stopping you from using those encoder ports for the scales.

3. We have had axis summing like that available since CNC7. There's a TB about it here: https://www.centroidcnc.com/centroid_di ... term=TB131. I would check the CNC12 manual for parameter differences.

4. No idea. Hopefully someone who has retrofitted one of these will chime in.

5. I would be interested to find out the rectifier arrangement. There may be a transformer in the electrical cabinet.

[martyscncgarage]

I have a client that would like his new to him TRAK DPM converted to Centroid. The machine has a late 90's vintage A.G.E. 3 Control

A few questions

1. The machine has Acu-Rite Non-Absolute Scales on X/Y/Zquill. He would like to reuse them with Centroid. Does anyone know if these scales output the correct differential signals? Pin outs?

2. Since he won't be using a spindle encoder, can I assign the 3 encoder (4/5/6) inputs for the scales?

3. The Z axis is of course the column along with the scale on the quill. Currently, you can adjust the Z column wherever you like and when you manually operate the quill, the A.G.E. 3 control will deduct/add the quill value from the Z axis DRO. Does CNC 12 do the same?

4. I would assume the DC servo motor encoders are still needed with glass scales. Are the original SWI TRAK encoders usable or do they need to be replaced with differential encoders? (I have not gotten into the end of the servo motor yet)

5. Interestingly, the A.G.E. control servo driver is attached to the side of each DC Brush servo. Says 115V. I assume that voltage is being rectified at the servo motor and I will need a cap board low for the DC Buss voltage, which will rectify the 115VAC.

Any other gotchas I should be aware of before I start ordering parts and building the All in One DC Control?

Thanks in advace.

Hello Marty,

1. Those look like Acu-Rite Mini-Scale or Mate encoders (see manual here: https://acu-ritesolutions.com/wp-conten ... Manual.pdf). You'd have to verify by checking the part numbers. That manual says that the Mini-Scale models use single ended 5V TTL quadrature signals while the Mate models do differential sine/cosine. It is not likely that they will directly work with the ALLIN1DC.

2. Yes, there's nothing stopping you from using those encoder ports for the scales.

3. We have had axis summing like that available since CNC7. There's a TB about it here: https://www.centroidcnc.com/centroid_di ... term=TB131. I would check the CNC12 manual for parameter differences.

4. No idea. Hopefully someone who has retrofitted one of these will chime in.

5. I would be interested to find out the rectifier arrangement. There may be a transformer in the electrical cabinet.

Thanks for the reply "Centroid467".
SWI has a habit of keeping everything about their machines and parts near and dear to them. No part numbers, no detailed schematics or pin outs
I can't find a part number on the scales. I sent a picture to Acu-Rite but of course no response. I was hoping Marc L. has had experience with them and had thoughts. He's been around the WORLD with Centroid and its conversions more than just a "block"
As I dig into the control, I can see what I can find out about them. I can't imagine SWI would have a proprietary scale made for them, but wouldn't put it past them.

Marty

[xr4x4ti]

I don't have to much to add except that I converted a similar 97 vintage machine and all of the scales worked fine with the Allin1DC after changing the connector.

The mill has been awesome after the conversion.

Tim

[martyscncgarage]

I don't have to much to add except that I converted a similar 97 vintage machine and all of the scales worked fine with the Allin1DC after changing the connector.

The mill has been awesome after the conversion.

Tim

Hey thanks Tim. I appreciate the input. Were you scales the same as the ones I have? Acu-Rite Non Absolute?
I'd be curious if you took the time to note the wiring color code or pin out of the Acu-Rite connector.

Did you reuse the servos? Replace encoders? I haven't had time to get into the servos yet. Been hoping I would hear back from some one that did a similar conversion.

Thanks again!
Marty

[cncsnw]

I have not done a machine like that.

With luck, Tim (xr4x4ti) has some notes about the pin-out.

If you have to start from nothing, you can make some educated guesses based on the existing wiring. If it is connected with 8 wires -- and especially if those wires are in 4 twisted pairs -- then it is likely differential. If it is connected with 5 or fewer wires, it is not differential.

If you can guess by context which wires are the 5VDC supply (maybe red and black; maybe brown and white; maybe brown and blue?), then you could connect 5VDC to it and measure the output on the other wire pairs.

You can sometimes identify the 5VDC pair by measuring resistance across the pairs. If you have three pairs with approximately the same resistance measurement across the pair, and one that is different, then it is likely the one that is different is the 5VDC supply.


On a separate note, you will want to remove those aluminum handwheels. If you leave the aluminum handwheels in place, you should refrain from running Centroid's torque-mode Autotune. The flywheel effect of the handwheels can cause a high enough regenerative voltage, during Autotune's high-speed moves, to damage the Allin1DC drive section.

[martyscncgarage]

I have not done a machine like that.

With luck, Tim (xr4x4ti) has some notes about the pin-out.

If you have to start from nothing, you can make some educated guesses based on the existing wiring. If it is connected with 8 wires -- and especially if those wires are in 4 twisted pairs -- then it is likely differential. If it is connected with 5 or fewer wires, it is not differential.

If you can guess by context which wires are the 5VDC supply (maybe red and black; maybe brown and white; maybe brown and blue?), then you could connect 5VDC to it and measure the output on the other wire pairs.

You can sometimes identify the 5VDC pair by measuring resistance across the pairs. If you have three pairs with approximately the same resistance measurement across the pair, and one that is different, then it is likely the one that is different is the 5VDC supply.


On a separate note, you will want to remove those aluminum handwheels. If you leave the aluminum handwheels in place, you should refrain from running Centroid's torque-mode Autotune. The flywheel effect of the handwheels can cause a high enough regenerative voltage, during Autotune's high-speed moves, to damage the Allin1DC drive section.

Thanks Marc.
My hunch is that they are probably differential. The control is still operational and I can check for 5VDC so that is a help.

Thanks for the reminder on the handwheels. I don't like them any way, the handles DON'T fold. Customer does want to use it manually until he is back up to speed with Centroid. I will pull them for the Auto tune and then put them back on.

As always, thank you for your input on the forums. Your suggestions have always been invaluable.
Hopefully Tim did document things on the encoder.

Thanks Marc
Marty