Stepper / Driver advice - Accuracy/Torque > Speed

[axefxpwner]

Hello!

I am in the middle of a small/med mill build (14x7.5), and am looking for advice on my axis motors.

Here is the mindset - I want to get set up nicely with steppers as cheaply as possible, and then upgrade to DMM servo's on each axis when possible (Any advice as to motor size in this regard would be appreciated as well). But I'd like them to be decent as I plan to repurpose them on a lathe build after they are swapped out.

I don't need super fast rapids, as it's a smallish work envelope. I more value accuracy and torque.

I have been researching Drivers, and have steered away from the Gecko G201x (7A 80V, $110), towards either the Leadshine AM882 (8.2A 80VDC - $66 Ebay) or the Stepperonline DM80i (7.2A 80V - $43.73). The stepperonline is very tempting because of its specs, price, and it's apparently made by Leadshine. Anyone have any input on these?

As far as steppers go, I want to keep in mind that I'm upgrading to DMM's later, and buy steppers that will make sense on the same ballscrew that will be used later with DMM's.

It will be used for machining aluminum and steel, with either a 750W DYN 4 (with 1.25:1 or 1.5:1 reduction), or a 1.8kw (2;1 ish reduction) DYN 4. Should i go with a Nema 34 right out of the gate on the Z axis?, and will a Nema 23 option be large enough for XY?

Finally, and most importantly,,, does anyone have specific recommendations or links towards the most ideal steppers for my application? I am of the impression I want as little inductance, and for it to be as high of voltage/amp draw possible right?

Thank you in advance.

Similar Threads:

• Stepper speed with current limiting driver
• stepper motor speed and torque calculation
• Problem- STEPPER MOTOR AND DRIVER CONNECTION ADVICE
• Newbie- Stepper/Driver Advice (I'm new, but have done some research.)
• Newbie needs stepper driver advice

[Jim Dawson]

Hello!

I am in the middle of a small/med mill build (14x7.5), and am looking for advice on my axis motors.

A model of the mill or pictures would be good.

Here is the mindset - I want to get set up nicely with steppers as cheaply as possible, and then upgrade to DMM servo's on each axis when possible (Any advice as to motor size in this regard would be appreciated as well). But I'd like them to be decent as I plan to repurpose them on a lathe build after they are swapped out.

I don't need super fast rapids, as it's a smallish work envelope. I more value accuracy and torque.

I have been researching Drivers, and have steered away from the Gecko G201x (7A 80V, $110), towards either the Leadshine AM882 (8.2A 80VDC - $66 Ebay) or the Stepperonline DM80i (7.2A 80V - $43.73). The stepperonline is very tempting because of its specs, price, and it's apparently made by Leadshine. Anyone have any input on these?

As far as steppers go, I want to keep in mind that I'm upgrading to DMM's later, and buy steppers that will make sense on the same ballscrew that will be used later with DMM's.

Just go with Stepperonline NEMA 34, 1200 oz/in and the DM80i drivers. Note the shaft sizes on the steppers, not all follow the NEMA 34 1/2'' shaft spec, this will save you from having to change couplers when you install the DMM servos, and everything will just bolt in.

It will be used for machining aluminum and steel, with either a 750W DYN 4 (with 1.25:1 or 1.5:1 reduction), or a 1.8kw (2;1 ish reduction) DYN 4. Should i go with a Nema 34 right out of the gate on the Z axis?, and will a Nema 23 option be large enough for XY?

I'm not sure why you would want to gear down the 1.8KW DMM servo, they have enough torque to twist your machine into a pretzel when direct coupled. I'm using a 1.8KW DMM servo, direct coupled, on the X and Z axis on my lathe. It has about 12 inches of Z carriage travel and it will hit >400IPM in a 4 inch rapid move. Oh, and the carriage assembly weighs over 2000 lbs. The DMM 750W NEMA 34 servos and the DYN4 drives would be more than large enough for your mill, on all axes. If I were going to replace the servos on my 10x52 (12.5x39 work envelope) mill, that's what I would use.

Finally, and most importantly,,, does anyone have specific recommendations or links towards the most ideal steppers for my application? I am of the impression I want as little inductance, and for it to be as high of voltage/amp draw possible right?

Thank you in advance.

See above

[axefxpwner]

Thank you for the input! That’s exactly the kind of input I was looking for. I will go ahead and grab the DM80i drivers and Stepperonline steppers.

I meant that I was thinking of doing the 1.8kw at a 1:2 ratio, as I feel it would have more than enough torque for my needs, and it would be nice to have the added RPM’s.

Is there any reason to go into nema23 sized steppers/servos on a smallish work envelope like this? Or just go nema34 1200oz in on each axis?

Your machine sounds awesome and terrifying

[Jim Dawson]

My pleasure

1:2 would make more sense, but do you really need 1000 IPM rapids on a 14x7.5 machine? Direct coupled, the DMM servos will give you up to 600 IPM rapids with a 0.200 (5mm) lead ball screw, the steppers would max out at about 300 IPM or a bit less.

I have seen mild steel cut at 1000 IPM, but that is on a ~15,000 lb machine with a 15,000 RPM, 30 HP spindle. On our knee mill with a 3 HP spindle I limit the rapids to 100 IPM, and normally cut in the 5 to 20 IPM range. This is due to the limited spindle speed. On out Haas with a 7.5 HP, 6000 RPM spindle we normally cut aluminum at 120 to 180 IPM, and stainless steel at ~30 IPM, with 400 IPM rapids, I don't think we have ever cut mild steel on it.

When I retrofitted our lathe, I upped the servo power just a little bit from the original factory servos. Accelerating and controlling that much mass does require some HP, and it will do 600 IPM rapids, but I limit it to 400 IPM and it's still scary. This is a 10,200 lb machine with a 10 HP spindle.

I would stick with the NEMA 34 vs. the NEMA 23 motors, that way you won't have to rebuild motor mounts or anything else to install the DMM servos, they will be a direct bolt in. The difference in cost would be more than compensated for from the avoided labor just in the motor mounts, not to mention new couplings.

The 1.8KW DMM servo would make a pretty good spindle motor for your mill. And in that case a 1:2 step up might be very useful.

[axefxpwner]

Thanks again! I should have clarified earlier, I intend to have the servo's/steppers on the axis' be direct coupled, and intend to use either a 750W or 1.8kw DMM for the spindle motor, which would be 1:2 step up.

Gonna go ahead and jump straight into the Nema 34's, which i plan to repurpose on my lathe build later this year after upgrading the mill to DMM's

[Jim Dawson]

OK, that makes sense now. At 1:2, you would need the 1.8KW servo.

Regarding the NEMA 34 vs. NEMA 23, you can always turn the power down, but it's pretty hard to get more power out of an under powered system. I would much rather have the power available if needed.

[QuinnSjoblom]

The 1.8kw for spindle drive has been on my mind a lot lately too. I was gonna gear up even further, like 1 to 4 for 12k rpm. That still gives about 6 ft/lb of torque all The way down to zero. Down at 1000rpm, that's still 1hp. Best horsepower (about 3 continuous, up to 6hp for short periods) up between 6k and 12k. I work mostly in aluminum so these numbers would be great for me. Only time id be down in the low rpm range would be for decking off small amounts with maybe a 2 inch facemill. With 6ft/lb all the way down to zero, I could still rigid tap pretty large threads in aluminum. In steel it looks like maybe a 1/4-20, but getting close to the limit.

Anyway, the dmm 1.8 is looking pretty awesome for a spindle, especially when you consider the weight (about 20 pounds) vs the huge motor you would need for equal power. On top of that, it doesn't need a separate encoder, also has ability to easily reference at encoder zero position for tool changes. Just so many advantages and all for 700 bucks with servo, drive, wiring, etc. Why are guys still running these big primitive motors? Lol

[Jim Dawson]

The 1.8kw for spindle drive has been on my mind a lot lately too. I was gonna gear up even further, like 1 to 4 for 12k rpm. That still gives about 6 ft/lb of torque all The way down to zero. Down at 1000rpm, that's still 1hp. Best horsepower (about 3 continuous, up to 6hp for short periods) up between 6k and 12k. I work mostly in aluminum so these numbers would be great for me. Only time id be down in the low rpm range would be for decking off small amounts with maybe a 2 inch facemill. With 6ft/lb all the way down to zero, I could still rigid tap pretty large threads in aluminum. In steel it looks like maybe a 1/4-20, but getting close to the limit.

Anyway, the dmm 1.8 is looking pretty awesome for a spindle, especially when you consider the weight (about 20 pounds) vs the huge motor you would need for equal power. On top of that, it doesn't need a separate encoder, also has ability to easily reference at encoder zero position for tool changes. Just so many advantages and all for 700 bucks with servo, drive, wiring, etc. Why are guys still running these big primitive motors? Lol

The big ClearPath motor is something I have considered for a spindle motor on my mill (to replace my 3HP Baldor). https://www.teknic.com/model-info/CP...433A-ELNB_Fan/ The downsides for me are that there is no direct encoder feedback to the controller possible, it will only accept step and direction input, and the maximum RPM is 2760 at which point it basically has no torque, useful range is to 2500 RPM. My preference is to use analog control and have feedback to the controller. ClearPath does have an analog to step & direction converter device and using an external encoder would be possible.

[QuinnSjoblom]

The big ClearPath motor is something I have considered for a spindle motor on my mill (to replace my 3HP Baldor). https://www.teknic.com/model-info/CP...433A-ELNB_Fan/ The downsides for me are that there is no direct encoder feedback to the controller possible, it will only accept step and direction input, and the maximum RPM is 2760 at which point it basically has no torque, useful range is to 2500 RPM. My preference is to use analog control and have feedback to the controller. ClearPath does have an analog to step & direction converter device and using an external encoder would be possible.

Looks like the clearpath is definitely a step up in power over the dmm, about 50%. Probably the next best choice if you need the extra juice. For me I think the dmm is gonna provide more than enough power for what I want to do, also huge flexibility for driving it however you want.

Biggest puzzle for me is finding a spindle to drive since I'm building from scratch. Was considering tormach bt30 but I'm not sure I can push it as hard as the dmm would allow. I wish there was something out there with just a tad more capability than the tormach for just a bit more money, but there doesn't seem to be much available for bt30 cartridges in similar price range. The tormach is 800 buck and rated for 10k rpm. Not sure what hp it could handle. Tormach motor puts out 1.5hp but I don't know if that's the limit for the spindle. I asked in the tormach forum but no responses. Id really like to have 12 to 15k rpm and 3hp.

Sorry, not trying to hijack, I think this is all useful info for the op

[Jim Dawson]

Looks like the clearpath is definitely a step up in power over the dmm, about 50%. Probably the next best choice if you need the extra juice. For me I think the dmm is gonna provide more than enough power for what I want to do, also huge flexibility for driving it however you want.

Biggest puzzle for me is finding a spindle to drive since I'm building from scratch. Was considering tormach bt30 but I'm not sure I can push it as hard as the dmm would allow. I wish there was something out there with just a tad more capability than the tormach for just a bit more money, but there doesn't seem to be much available for bt30 cartridges in similar price range. The tormach is 800 buck and rated for 10k rpm. Not sure what hp it could handle. Tormach motor puts out 1.5hp but I don't know if that's the limit for the spindle. I asked in the tormach forum but no responses. Id really like to have 12 to 15k rpm and 3hp.

Sorry, not trying to hijack, I think this is all useful info for the op

A wild guess tells me that the limitation might be in the bearings. A chat with a spindle rebuilder might be worthwhile. I have these guys do my spindle work. https://www.cm-spindle.com/

[QuinnSjoblom]

A wild guess tells me that the limitation might be in the bearings. A chat with a spindle rebuilder might be worthwhile. I have these guys do my spindle work. https://www.cm-spindle.com/

Yep, my assumption too, which mostly makes me wonder how capable the tormach bt30 could be with really nice bearings in it. Balance is gonna be really important also if I'm trying to push it to 12 to 15k. Not sure what's involved or what equipment is needed to dynamically balance a spindle. After upgrading bearings and balancing, I'm not sure what other limitations it could really have. Maybe have the taper reground after rebuilding it if runout isn't great, but keeping in mind that I'll be able to clock toolholders, I should be able to get tools configured with low and consistent runout. That's the main limitation in machining accuracy for me now. runout variance based on how the tool and collet is held. The taper in my chinese 2.2kw is actually really good, maybe 1 or 2 tenths on the taper, but runout in the tool tip varies anywhere from 0 to about 5 tenths with random positioning of collet and tool. Being able to clock tools and have consistency will be huge.

[Jim Dawson]

Yep, my assumption too, which mostly makes me wonder how capable the tormach bt30 could be with really nice bearings in it. Balance is gonna be really important also if I'm trying to push it to 12 to 15k. Not sure what's involved or what equipment is needed to dynamically balance a spindle. After upgrading bearings and balancing, I'm not sure what other limitations it could really have. Maybe have the taper reground after rebuilding it if runout isn't great, but keeping in mind that I'll be able to clock toolholders, I should be able to get tools configured with low and consistent runout. That's the main limitation in machining accuracy for me now. runout variance based on how the tool and collet is held. The taper in my chinese 2.2kw is actually really good, maybe 1 or 2 tenths on the taper, but runout in the tool tip varies anywhere from 0 to about 5 tenths with random positioning of collet and tool. Being able to clock tools and have consistency will be huge.

Balancing and taper regrinds is part of a good spindle rebuild. That's what those guys do, they have the equipment. When my spindle was rebuilt they reground the taper, TIR was 0 as best I could measure. I don't think they balanced it, but it's only a 5000 RPM spindle.

Balancing is done on a balancing machine. There's one on my local Craigslist for $6500 if you want one Good (expensive) tool holders are important when you get up into the higher RPM ranges.

[QuinnSjoblom]

Balancing and taper regrinds is part of a good spindle rebuild. That's what those guys do, they have the equipment. When my spindle was rebuilt they reground the taper, TIR was 0 as best I could measure. I don't think they balanced it, but it's only a 5000 RPM spindle.

Balancing is done on a balancing machine. There's one on my local Craigslist for $6500 if you want one Good (expensive) tool holders are important when you get up into the higher RPM ranges.

Yeah, I figured balancing was probably something I wouldn't be doing myself. So just rough ballpark, what do you think it would cost to have someone regrind and balance a tormach bt30? And what about the rebuilding? Is it difficult to swap out bearings in a spindle? Not sure what kind of tolerances are involved. Is everything gonna need to be press fit? Or just heat/cool different parts and bearings and simply slide it together? I should probably do some research on spindle building. I'm also very curious about actual machining and grinding of spindles.

[Jim Dawson]

Yeah, I figured balancing was probably something I wouldn't be doing myself. So just rough ballpark, what do you think it would cost to have someone regrind and balance a tormach bt30? And what about the rebuilding? Is it difficult to swap out bearings in a spindle? Not sure what kind of tolerances are involved. Is everything gonna need to be press fit? Or just heat/cool different parts and bearings and simply slide it together? I should probably do some research on spindle building. I'm also very curious about actual machining and grinding of spindles.

I had my BP clone spindle done for $900, but that included doing the 5 bearing upgrade (normal is 3 bearings). I think the standard rebuild is about $600. I'm guessing a Tormach spindle would be in that ballpark also.

Many people stuff new bearings in their spindles and call it a rebuild, I would call it stuffing in new bearings. I have never actually done it, but yes it involves some pressing. If I had to, I would do it in my shop, but I would rather pay the properly equipped experts to do it for me. Setting proper preloads, proper grease, amount of grease, and test running are just some of the things that you are paying for.

[QuinnSjoblom]

I had my BP clone spindle done for $900, but that included doing the 5 bearing upgrade (normal is 3 bearings). I think the standard rebuild is about $600. I'm guessing a Tormach spindle would be in that ballpark also.

Many people stuff new bearings in their spindles and call it a rebuild, I would call it stuffing in new bearings. I have never actually done it, but yes it involves some pressing. If I had to, I would do it in my shop, but I would rather pay the properly equipped experts to do it for me. Setting proper preloads, proper grease, amount of grease, and test running are just some of the things that you are paying for.

Very true, better off leaving it to the professionals. Another thought, if I was planning on a rebuild with new bearings right off the bat, I could probably pick up a used tormach bt30 for 500 bucks and send it off. A used spindle with new bearing, regrind, and balance would probably be not a lot different than starting off with a fresh one. Potentially have a very capable spindle for 1000 to 1500. For that matter, I could probably keep my eyes open and find something better than the tormach bt30 if I buy used. Since I'm rebuilding, the mystery of what condition the bearings are in doesn't matter.

[Jim Dawson]

Very true, better off leaving it to the professionals. Another thought, if I was planning on a rebuild with new bearings right off the bat, I could probably pick up a used tormach bt30 for 500 bucks and send it off. A used spindle with new bearing, regrind, and balance would probably be not a lot different than starting off with a fresh one. Potentially have a very capable spindle for 1000 to 1500. For that matter, I could probably keep my eyes open and find something better than the tormach bt30 if I buy used. Since I'm rebuilding, the mystery of what condition the bearings are in doesn't matter.

That makes sense.

[axefxpwner]

Great conversation going on here, I think I've decided to go with the 1.8kw DMM as well, and plan to run it around 3hp 12k-15k rpm as well Quinn. I am currently on the hunt for a spindle and entertaining the possibility of building my own and having it ground somehow? I am a member of a facebook group "The taig mill and lathe group" and a member on there constructs ATC R8 spindles at a decent price.

[Jim Dawson]

Great conversation going on here, I think I've decided to go with the 1.8kw DMM as well, and plan to run it around 3hp 12k-15k rpm as well Quinn. I am currently on the hunt for a spindle and entertaining the possibility of building my own and having it ground somehow? I am a member of a facebook group "The taig mill and lathe group" and a member on there constructs ATC R8 spindles at a decent price.

You might have a bit of problem trying to run a 1.8KW (2.4 HP) motor at 3 HP (2.2 KW) But the DMM servo will work well in its normal operating range.