[rcheli]

The most common formula for calculating the voltage to run a stepper is 32 times the square root of the inductance. So using 2mH that would be about 62 volts but you are limited by the drive that is rated at 50. I used 381oz steppers with an inductance of 2.8 with is 53.5 volts on my X & Y of a mill using a 48 volt power supply. I have plenty of power and enough speed for a mill.

[cr88192]

The most common formula for calculating the voltage to run a stepper is 32 times the square root of the inductance. So using 2mH that would be about 62 volts but you are limited by the drive that is rated at 50. I used 381oz steppers with an inductance of 2.8 with is 53.5 volts on my X & Y of a mill using a 48 volt power supply. I have plenty of power and enough speed for a mill.

Interesting.

Looking into it, the NEMA23's on my G0704 appear to be 2.2mH and rated for 6.0A.
I am running them with a 48V supply (more or less, this particular supply seems to be open-loop and its voltage varies a bit depending on load, ...).
This metric would give ~47V.

I am generally running them at 20 IPM, with.
IIRC, ~ 10 revs/inch; or ~200 RPM for X/Y; and ~ 32 revs/inch or ~ 512 RPM for Z (IIRC, it was 5 TPI for X/Y with 2:1 reduction, 8 TPI for Z with 4:1); could be wrong here, I am going off memory.

But, as noted, 20 IPM was chosen as this is the speed I could run it at without getting any lost steps during a sequence of rapid moves (would issue a sequence of rapid moves and verify that the mill was still in the expected place).

The 570 oz-in NEMA23's on my lathe are 2.5mH, rated for 5.0A. I ended up running them at a similar speed.
I still haven't extensively tested this, but it appears to work.
There is a 2:1 reduction (due to the lathe's gearbox); working backwards, it seems to be 28 revs/inch, or 14 TPI on the leadscrew (may need to verify), implying ~ 560 RPM.
At 1:1, the torque requirements (to move the carriage) tend to overload the motor (could require closer to ~ 1000 oz-in?).

In both cases, it seems I am running them a bit under their rated current, but then again this limits the motors getting overly warm.


The NEMA34 I am using as a spindle is 5mH (as wired), working out to around 73V.
As noted, it is rated 8.8A, but I am currently running 5.5A (and it runs decidedly warm), but is apparently probably OK (after a few minutes, reaches a temperature of ~ 40-50C or so, then mostly levels off).

The way that the Automation Technologies page said to wire it would make it have 20mH, but whatever voltage the KL-11080 is putting out doesn't really appear to be enough to get particularly high RPMs (this metric gives 143V, and bridge-rectified 110VAC could theoretically give 170V, so dunno...). Extrapolating from other patterns, the driver seems to perform more like it is operating at 70V than would be expected if it were operating at 140-170V.

As-is, running the spindle at 250 RPM, is ~500 RPM on the motor.
If it were 140-170V, I would expect a wider RPM range than this (with a motor at 5mH).


I am not sure the voltage the driver is putting out (whatever it is putting out is seemingly something the multimeters I have are unable to deal with, granted all I have on hand are the Harbor Freight / Cen-Tech DMMs).

[ger21]

FWIW: I have been off experimenting with using a NEMA34 stepper as an alternate lathe spindle motor ( https://www.automationtechnologiesin...34h2160-62-8a/ ), with the KL-11080 driver.

The NEMA34 I am using as a spindle is 5mH (as wired), working out to around 73V.
As noted, it is rated 8.8A, but I am currently running 5.5A (and it runs decidedly warm), but is apparently probably OK (after a few minutes, reaches a temperature of ~ 40-50C or so, then mostly levels off).

The way that the Automation Technologies page said to wire it would make it have 20mH, but whatever voltage the KL-11080 is putting out doesn't really appear to be enough to get particularly high RPMs (this metric gives 143V, and bridge-rectified 110VAC could theoretically give 170V, so dunno...). Extrapolating from other patterns, the driver seems to perform more like it is operating at 70V than would be expected if it were operating at 140-170V.

As-is, running the spindle at 250 RPM, is ~500 RPM on the motor.
If it were 140-170V, I would expect a wider RPM range than this (with a motor at 5mH).

What it basically comes down to is that the bigger the stepper, the slower it'll spin. No matter how good the drive is. 500 rpm is the max you're likely to get out of anything over 1400oz.

[rs4race]

The 640ozin Nema 34's are a pretty good combo on the g0704, way better than the 570 nema 23.

[bogiestl]

Well, right now, it is looking like I'll just go for the 2x23, 1x34 kit with the 570 motors from Automation Technologies. I'd like to do 34s all the way around, but frankly, it is probably right to go with the 570s... The control will likely be the Centroid.

I may go with larger mounts in the future if I upgrade to the quieter Clearpaths. I don't know.

[rs4race]

Thats ok. With 5mm pitch ball screws the 570's loose a lot of torque at about 100ipm. This may limit your rapid speeds. Its why I changed from the 570's to the 640's.

Well, right now, it is looking like I'll just go for the 2x23, 1x34 kit with the 570 motors from Automation Technologies. I'd like to do 34s all the way around, but frankly, it is probably right to go with the 570s... The control will likely be the Centroid.

I may go with larger mounts in the future if I upgrade to the quieter Clearpaths. I don't know.

[ger21]

A 6 amp 460oz Nema 34 would probably be a decent upgrade over the 570.

[arizonavideo]

The 640 look good with 2,2mH of inductance.

You generally get with a 5 TPI screws about .8 pounds of peak force from every once inch of torque.

So a 570 will give about 456 pounds of force and the 640oz 512 pounds for a mill that weighs 250 pounds.

[rs4race]

The 570 ozin nema 23 drops to ~100 ozin at 550 rpm which is about 110 ipm with 5 mm ball screw. In contrast the 640 ozin nema 34 (KL34H280-55-4AT) has ~250 ozin at 550 rpm. More than double the torque. I had problems stalling at 100ipm rapids with the nema 23 this matches the torque / speed curve. I haven't had a problem with the nema 34 so far. Theres more to consider than the holding torque.

[arizonavideo]

Well the 6.3A 640 looks to be the fastest stepper of all the steppers he has. The driver looks like a standard one. I don't know if the digital drivers have copied Gecko drives "drive Morphing" function which would give even better high speed performance.

Something is wrong with the 570 oz torque curve. All the stepper should behave close to the same. Even power in the chopper current limit range then a 24DB roll off after that. I don't understand why the 570 falls faster. I think the driver was having a hard time controlling the resonance the stepper.

Either way the 640 looks like a winner. I thought it was 3MH I must have been looking at the wrong one.

Still for the smaller mills the 570 is the best Nema 23 mount. I don't know if the closed look will out perform at higher speeds but I do have both on hand so it night be a fun test.

[dazp1976]

Bit pricey for me especially with the con-men in the import V.A.T. offices. looks a nice unit though.
I think I've found something I can live with for my new one as a starter then look at servo's later depending on how it goes:
https://www.ebay.co.uk/itm/German-Sh....c100290.m3507
Shows 4.1mh inductance and under the earlier formula (square root MH x 32) comes out at 64.8V which seems to be a close enough match to the psu's (60V). At 1090ozin if I get half out of it it will do me. 2000mm/min (78in/min) looks to be about 400rpm on a 5mm screw which is adequate. Even 1500mm would do.
Took me a while to find something reasonable, a heck of a lot of motors are above 6mh inductance.
Some of the 8 wire units can be wired in such a way to get them down to 3.5 but I can't be bothered with them. I'm bound to wire them wrong

[bogiestl]

Well, speed isn't the overwhelming goal here - Being able to push the button, and then come back later is preferred - so if I have to run rapids slower, that's fine.

[arizonavideo]

You should run then at 80V.