Using DMM 750W servo as spindle motor

[Richards]

I plan on using my DMM DYN4 750W servo motor as the spindle motor on my Taig mill. The mill comes with a 1/4th horsepower motor that is overly large. The 1 HP DMM 750W motor is smaller, lighter, and more powerful. Because I have the DYN4 drive and the servo motor on the shelf, I contacted DMM for instructions on how to use it as a spindle motor. They quickly replied and attached a schematic to show me how to hook it up to the Acorn. I'm attaching that schematic. I don't think that the encoder output from the motor will work with the Taig. As I understand it, the encoder has to send one pulse per spindle rotation. Because the Taig has a stepped pully allowing me to mix and match spindle speeds, I'll have to attach an auxilary encoder that gives one pulse per spindle shaft rotation. I have not tested my motor with the new schematic. I'll do that this week-end if possible.

[martyscncgarage]

Because you are putting on a monster of a spindle motor on the little lathe, and its an AC servo, you could belt it at 1:1 to the spindle with a timing belt. Like to see that little guy rigid tap if you get it all working.....

My .02
Marty

[Richards]

Marty,
My problem is that I'm running the mill on 120VAC which would limit the DYN4 speed to 3,000 RPM. I'd like to run the spindle at speeds at 6,000 RPM or higher when cutting aluminum. I could possibly use a dual step timing belt pulley with one set of pulleys set for 1:1 for low speed tapping and the other at 1:3 for high speed machining. The high speed would give me speeds above 6,000, similar to what I'm using now with the stock motor and pulleys.

[martyscncgarage]

Marty,
My problem is that I'm running the mill on 120VAC which would limit the DYN4 speed to 3,000 RPM. I'd like to run the spindle at speeds at 6,000 RPM or higher when cutting aluminum. I could possibly use a dual step timing belt pulley with one set of pulleys set for 1:1 for low speed tapping and the other at 1:3 for high speed machining. The high speed would give me speeds above 6,000, similar to what I'm using now with the stock motor and pulleys.

Yes you are correct. Yes you would have to mount a dedicated encoder if you intend to try rigid tapping
Marty

[cncsnw]

If the spindle turns an exact (integer) number of turns for each turn of the spindle motor, then you can still do multi-pass threading and tapping using just the motor encoder. It just means the control might have to wait an extra turn or two of the spindle before the index pulse comes around; but when it does, it will always be in the same place in the spindle rotation.

[Richards]

I started thinking about what cncsnw said about using multi-pass threading in conjunction with the DYN4/servo but then I realized that what we would normally have is the opposite problem. I may have written the ratios backwards. If the motor is turning at 3,000 RPM and the spindle is turning at 9,000 RPM, then the motor would produce just one "Z" pulse per three revolutions of the spindle shaft. I haven't looked at the Centroidcnc thread cutting routines, but I think that stepping up the speed would require an encoder on the spindle shaft to accurately synchronize the spindle to the thread cutting routine.

[cncsnw]

I may have written the ratios backwards. If the motor is turning at 3,000 RPM and the spindle is turning at 9,000 RPM, then the motor would produce just one "Z" pulse per three revolutions of the spindle shaft.

If it were set up that way, it would work. All that is required for synchronized threading is that, when the index pulse does come around, it comes around at the same spindle/tool angle every time. If that is once every 1080 degrees, instead of once every 360 degrees, that is still okay.

The opposite ratio would not work: if you had three turns of the motor for one turn of the spindle, then the index pulse would come around once every 120 degrees, and threading cuts would begin at any of three different angles, more or less at random.

Going back to your 1:3 example: in order to get the correct advance rate (lead), you would need to set Parameter 34 to 1/3 of the encoder counts per motor revolution (so it is the number of encoder counts per spindle revolution).

[Richards]

Cncsnw,
Thanks. It's all becoming clear.

[Richards]

I finally found some time to wire the DMM DYN4 750W motor as a spindle motor on the test bench. Everything seems to work expect the speed is about 30% higher than expected. I have the maximum spindle speed in high range set to 3,000 in Wizard. Spindle encoder counts = 2,000 with Line number = 500 in DMM setup. Gear number in DMM setup = 4,096 (default). I've tried Spindle encoder counts of 8,000 / Line = 2,000 and Encode = 8,196 / Line = 2048. The encoder readout on Centroid is about 30% higher than expected. Unfortunately, I don't have a Tachometer to cross check the actual speed. However, when I command 750 RPM, the analog output from the Acorn is 2.56V. When I command 1,500 RPM, the analog output from the Acorn is 5.05V.

I'm not using shielded cable to the encoder. I'll order some as soon as I can verify that I've setup the software parameters correctly. DMM OnPosRange=111 and MaxSpeed=13 as directed by DMM. MDI shows speed set to 1,500RPM with actual at 2,014 RPM.

Any advice?

[martyscncgarage]

Give DMM a call. I vaguely remember dinking with a value in DMM DRV. I think it was Gear_NUM raising or lowering it until it mached the expected speed. I personally abandoned using the DYN4 and AC motor for a spindle motor. I understand some got it working. I also understand you can not use AUTOTUNE in DMMDRV to tune a DYN4 connected to a spindle and you have to use gain values provided by DMM.

Watch your Acorn Analog output, make sure it is linear with your commanded speed from 0-10VDC based on what you programmed in MIN to MAX Spindle RPM in the Wizard.

I'm following your progress to see if you are successful.

Marty