PDA

View Full Version : Stepper Motor Torque



herbet999
11-29-2012, 12:47 PM
How does one figure out the proper torque needed to motorize the cross slide of a lathe?

I have a Maximat Standard lathe (about a 9 inch throw) and the best I've come up with based upon some web researching is about a 1 newton meter stepper motor... but that is just a wild guess.

Forestgnome
11-29-2012, 01:13 PM
An accurate method would be to mount a wheel to the cross slide handle, wrap with a string, and hang weights on the string until you get a consistent movement in the slide. Multiply th radius of the wheel times the weight and you get the minimum force in inch ounces, or whatever units you want.

MaxHeadRoom
11-29-2012, 01:30 PM
That measures the break-away torque which is OK for low rpm applications, in high rpm apps where a high acceleration is required, the required torque will be higher.
Max.

herbet999
11-29-2012, 01:54 PM
An accurate method would be to mount a wheel to the cross slide handle, wrap with a string, and hang weights on the string until you get a consistent movement in the slide. Multiply th radius of the wheel times the weight and you get the minimum force in inch ounces, or whatever units you want.

I never considered a straight empirical method. It's about time I put all of those physics classes to good use. I would think it would be wise to choose a motor with greater torque than what is measured. Do you think 40% greater would be sufficient?

Evan
11-29-2012, 02:15 PM
There are also very significant cutting forces involved. These feed back even with an acme screw. The stepper will need at least ten times the torque required to simply move the cross slide. Steppers generally produce the most torque vs velocity in half step mode with torque falling off more or less linearly as the finer step modes are used at the same velocity.

Acceleration depends both on the driver algorithm and the angular inertia of the rotating components driven by the stepper.

I have a spreadsheet written by Alexander Slocum, an MIT professor and top authority on the design of mechanical systems. This sheet is a motor size estimator for driving leadscrews. You may download it here: http://ixian.ca/server/motor_leadscrew_size_estimator.xls

Note: This link will become invalid in 24 hours from Nov 29 2012 11:30 am PST

While this material is copyright it was also provided freely to the general public and does not contain any restrictions on reproduction. However, it is no longer available from source.

herbet999
11-29-2012, 02:57 PM
Thanks Evan.

herbet999
11-29-2012, 03:02 PM
Steppers generally produce the most torque vs velocity in half step mode with torque falling off more or less linearly as the finer step modes are used at the same velocity.


so does that mean that it's more important to use half steps at faster speeds and less important at slower speeds or the other way around?

BigJohnT
11-29-2012, 03:32 PM
If you take a look at the Gecko 203v stepper driver (the most cost effective bullet proof drive I've found) and it starts at 10 micro steps and in the mid range it morphs to full steps. The morphing point is adjustable for the smoothest transition. Any fixed step rate will suffer at one end or the other of the speed range. Acceleration is mainly how much voltage you use and I'm not a MIT (Mississippi Institute of Technology) graduate... all other things equal. Some good common sense info is here (http://wiki.linuxcnc.org/cgi-bin/wiki.pl). Also on the Gecko web site is some very good information on steppers.

The torque rating of a stepper is while it is doing nothing IE it is stationary and the rating is how much it resists moving, not how much work it can do. The torque falls off with RPM. Look at the stepper torque vs speed charts at Automationdirect.com.

John

Barrington
11-29-2012, 03:49 PM
I have a spreadsheet written by Alexander Slocum, an MIT professor and top authority on the design of mechanical systems. This sheet is a motor size estimator for driving leadscrews. You may download it here: http://ixian.ca/server/motor_leadscr..._estimator.xls

Note: This link will become invalid in 24 hours from Nov 29 2012 11:30 am PST

While this material is copyright it was also provided freely to the general public and does not contain any restrictions on reproduction. However, it is no longer available from source.
For those reading this later - don't panic !!

It is still available from MIT:-

http://web.mit.edu/2.75/resources/FUNdaMENTALs Book pdf/FUNdaMENTALS Design Spreadsheets/Topic 6 Screws and Gears Spreadsheets/motor_leadscrew_size_estimator.xls (http://web.mit.edu/2.75/resources/FUNdaMENTALs Book pdf/FUNdaMENTALS Design Spreadsheets/Topic 6 Screws and Gears Spreadsheets/motor_leadscrew_size_estimator.xls)


edit: May be of interest - Slocum's "FUNdaMENTALS of Design" is here:-

http://web.mit.edu/2.75/resources/FUNdaMENTALs%20Book%20pdf.zip

from:

http://web.mit.edu/2.75/resources/FUNdaMENTALS.html

Cheers

.

PaulT
11-29-2012, 04:20 PM
I think the follow quote applies to this question-

"Speed costs money, how fast do you want to go?"

Almost any reasonable size stepper motor will give you enough torque for cutting operations on a small lathe.

Its when you want the machine to accelerate fast and have fast rapids that you need a big motor.

I used a large size Nema 23 stepper motor for the cross slide on my 12" lathe, its still in the build process and yet to be fired up but the motor I picked puts out a fair amount of torque for its size and also RPM's pretty well so it should work well for a cross slide app. Typically on a cross-slide you'll see stepper motors used with torque in the range of 400 to 700 oz-in depending on the size of the lathe (and on how fast you want to go).

The motor is the larger Nema 23 one sold on the Geckodrive site, their model G723-400-4, it puts out 400 oz-in. I liked it because it has a lot of pop for its size and also has low enough inductance to develop close to full power on a 60 volt power supply. Picking a stepper with the correct voltage range for your power supply/driver combo is very important, otherwise you won't get the rated power, especially at higher RPM's.

The formula to calculate the optimal power supply for a given stepper motor is:

Voltage = 32 x sqrt(L) where L is inductance in mH.

The advantage of using the 23 size motor on the cross slide instead of a 34 size is that its a lot smaller, in my case it fits nicely on a bracket that clamps right on the cross-slide dial flange, a 34 size motor wouldn't fit there and would be too heavy for a mounting like that.

However for the Z drive (carriage) I used a 1200 oz-in Nema 34 motor as for drilling and movement of the heavier carriage you need more than a 23 motor can put out. Surprisingly the large 34 size motor they sell on the Gecko site has a pretty high inductance and won't make close to full power with a Gecko driver, which has an 80 volt maximum supply level.

I found a good 34 size motor from automationdirect.com that will work well on a 60 volt supply with a Gecko driver, its model number is:

STP-MTRH-34127 1292 oz-in, 6.3A, .49 ohm, 4.14 mH (optimal voltage is 65 volts)

Good luck-

Paul T.

BigJohnT
11-29-2012, 04:25 PM
Paul T.

Do you have a photo of your 23 mount on your lathe cross slide?

John

Evan
11-29-2012, 05:27 PM
For those reading this later - don't panic !!

Thanks for finding that. Slocum's class notes have been off line for years. I guess there has been enough demand or something to repost them.

edit: I see what happened. Slocum took all the material off line while he was writing the book. I didn't expect him to put it back on line for free. Nice guy. I depended on that information to design my milling machine.

Evan
11-29-2012, 05:36 PM
so does that mean that it's more important to use half steps at faster speeds and less important at slower speeds or the other way around?

A stepper will generally produce the same torque or close to it when half stepping as it can full stepping at the same rpm. When stepping at finer divisions such as 1/4 or 1/8 the torque declines for the same rpm. 1/2 stepping also produces fewer resonances, 1/4 even less etc. When it gets down to 1/16 steps the drive signal fairly closely approximates a sine wave and the motor is running as a synchronous AC motor. Steppers are a type of synchronous motor. Some are sold as synchronous motors as well as steppers.

Evan
11-29-2012, 05:42 PM
However for the Z drive (carriage) I used a 1200 oz-in Nema 34 motor as for drilling and movement of the heavier carriage you need more than a 23 motor can put out. Surprisingly the large 34 size motor they sell on the Gecko site has a pretty high inductance and won't make close to full power with a Gecko driver, which has an 80 volt maximum supply level.

You can get around that somewhat by putting in a low series resistance to the motor coils. That acts to reduce the effective inductance.

Black_Moons
11-29-2012, 07:02 PM
You can get around that somewhat by putting in a low series resistance to the motor coils. That acts to reduce the effective inductance.

Close. The idea is to put a series resistance *and* increase the voltage, such that it acts as a very cheap type of constant current supply by overvolting the motor untill the current rises to the standard level, without overcurrenting the motor at lower RPM where the inductance is less a factor. the basic math is if you have a 12V 1A stepper, you can assume its DC resistance is 12ohms. If you wanted to run it at 48v, you would use 36ohms in series such that its limited to 1A current (12+36 = 48ohms = 1A at 48v), the motor will only see more then 12v when its current is lower then 1A. (Voltage drop across the 36ohms is V = Current * 36ohms, and hence voltage across the motor is 48v - (current * 36ohms)

A much better method is a 'chopper' stepper driver that can PWM the indivual coils to maintain a constant current level. (Better as in: less wasteful of electricity)

Evan
11-29-2012, 07:32 PM
That works. But, you can also use a much lower resistance to simply slow the rise time a bit. That will allow for better efficiency in the motor at the expense of some loss in the resistance. It does depend a lot on the type of driver used and how it deals with short pulses at high rpms.

PaulT
11-30-2012, 12:18 PM
JohnT, the motor mounts not currently on the machine as I took it off for a few more tweaks and then got swamped at my real job.

It will be going back together soon and I'll post some pics of it. I "borrowed" the basic design from a retrofit I saw on cnczon.com that I liked.

Basically the bracket is a 4" x 18" x 3/4" 6061 bar with one end bored to clamp around the cross slide dial flange.

The other end has the motor slots and some control switches and there is an additional support block that bolts to an existing hole on the carriage to support that end of the bracket, so no permanent mods to the lathe were required to mount the motor (its a nice Clausing 5914 and I didn't want to start drilling holes it it).

The cross slide dial was replaced with the drive pulley for the motor but the motor I mentioned has shafts on both ends so I placed a dial on the "extra" shaft that faces the user so I still have a dial readout on the machine. The cross-slide is using a zero backlash acetal nut using the slick method Evan documented on this forum (search "acetal nut" and you'll hit this thread).

The other advantage of this mounting method is I can slack the motor belt in about 30 secs so its very quick to put the machine back in full manual mode. This is also easily done on the Z drive so I have my cake and can eat it too, the machine can be quickly put back in full manual mode or can run full CNC with about a 30 second change over.

Paul T.

Evan
11-30-2012, 12:39 PM
This is how I mounted the cross feed drive on my SB9. It isn't a stepper, merely an electric drive which I would not be without. It drives a magnetic clutch operated with a simple toggle switch. The power is from a reversible pulse width modulated supply I designed that will vary the speed from a creep to about an inch per minute. It is bolted to the back end using existing holes for the taper attachment and connected via a spring bellows coupling to an acme threaded connector. It consumes about .25" of travel which has very rarely been an issue but if it is the entire unit can be quickly removed. If the drive reaches a hard stop the magnetic clutch just slips with no damage. It can also be overridden by hand at the front.

http://ixian.ca/pics10/crossfd1.jpg

http://ixian.ca/pics10/crossfd2.jpg

Forestgnome
12-01-2012, 09:49 AM
Nice setup Evan! That's thinking out of the box. I've been pondering CNC'ing my lathe, and never considered putting a drive on the backside of the carriage. It would be nice to keep the manual controls uncluttered.