Pancake Motors

[ihavenofish]

Low Heat, Continuous Duty, Closed Loop Step Motor & Driver System - AR Series (DC Input)

Oriental Motor U.S.A. Corp. - Search Results

theres one of vextas kits. the steppers are pretty standard. 280 oz in stall in a nema23 frame. a same sized servo typically has 180 oz in continuous torque, and more than 2 times that for intermittent duty, and a flat curve accross its speed range.

for this specific application - direct drive rotary head - its the continuous low speed torque thats key, and steppers tend to offer more.

i think low cost step servo kits will be great for lower speed linear axis drives, but not direct drive rotary.

[JerryFlyGuy]

A closed loop Stepper is the lowest speed highest torque motor for its size. I've got one in the works, and Gecko does too.

Are you planning on direct driving the rotory axis's or using a wormgear or planetary ?

Larry my plan had been to direct drive. The beauty of a pancake motor is that they are relatively small in size yet produce large amounts of torque. My min torque level is 100 Nm or 1400 oz-in. I'd like even more than that [200Nm or more would be fantastic] W/ the pancake motors they don't deliver high rpm which is fine for what I need. If I can't find something that delivers this at a reasonable/finite cost then I might as well go w/ a servo and planetary reducers [all of which i've got]. I've got my hands on some nice AC servos which IIRC deliver ~700oz-in cont and that coupled w/ a 10:1 reducer will deliver everything and more that I need. These servo's spin at 4400rpm so I could go to a 100:1 reducer and it would still have too much speed [for what I need].

I've sworn of steppers and in this situation [even w/ a feedback loop] i don't see the advantage over a AC servo. They are basically the same size yet I've not seen one that can deliver the performance of a similar [physical] sized AC servo, but then again I've never looked too closely at them. I'm also not willing to be on the bleeding edge of tech [been there too many times, anyone want to buy a G100?]

Fwiw

JFG

[Larken]

Would you give an example of this with a link to a part number with specs and dimensions please?

When you say highest torque, do you distinguish between continuous stall torque or peak torque or do you mean both?

thanks.

Continuous Torque

Here is size 23 stepper with 425 in/oz of continous torque (lowspeed)
KL23H286-20-8B (1/4” shaft with a flat)
NEMA 23 BIPOLAR STEPPER MOTOR 425 oz-in

A brushed servo doesn't even come close, neither a brushless servo for a motor with the same size dimentions.
But only at very low speed does a stepper produce good torque


Jerry, a hub motor is similar to a pancake, but there is less plate flex and no side force since the stator is on an outside ring and the magnets on a narrow inside armature . High output pancake motors produce a lot of side force.
I built a big hub motor to power a Honda car, but never finished it. (gas prices dropped)

[JerryFlyGuy]

True, steppers put out high torque at low speed. But if you look at the whole picture they produce less power than a AC brushless unit of the same size.

425oz of torque at stall is only good for holding position.

400 oz at 50rpm reduced 10:1 would give you 5 rpm and 4000 oz-in of torque.
5 rpm is also well under my needed rotation speed [5rpm= 30 deg/sec] Also w/ a 12" tool tip radius 4000 oz-in is equal to 20.8 ft/lbs or 20.8 lbs to move the tool at it's tip.

600 oz at 4400 rpm reduced by 100:1 would give you 60000oz-in of torque at 440rpm. [440rpm is well higher than I need, 440rpm= 2640deg/sec]. 60000oz-in = 312.5 ft/lbs or 312.5lbs at the same tool tip to move it out of position.

There is no comparison as far as I can see.

If I was to reduce the rpm of the AC motor down to the level of the stepper the torque would be just that much higher [astonomically].

Ultimately, torque is not the end-all/be-all of specing a motor. Torque at rpm is what matters [ better known as power].

What am I missing?

Btw, I'd started looking at pancake motors and then also moved to framless motors [which are more barrel shaped, but can be aquired w/ very little length to the 'barrel' body. Regardless if it is mounted properly on a larger shaft and the rotational section is on a bearing mounted 'hub' then the side load forces should not be of concern.

Not an expert in any way, just as I see it.

JFG

[ihavenofish]

i think youve screwed up some math. 100NM = 14,000 oz in, not 1400. 4400rpm reduced 100:1 is 44 rpm, not 440.

youve never told us how big a motor you can accomodate though.

if you go for frameless, in a 89mm frame, water cooled, 1" stack you can get about 4nm continuous, and very high rpm. it will operate at 42v at 1000rpm and 4-5 amps depending on control method, so its suited to many hobby class controllers and power supplies. this motor costs about $600 with internal commutation (parker K0891009Y2).

you can get these down to 1/4" stacks, but of course the torque drops off proportionally. they also go up to 5" long if you have extra room.

the next size up from parker is 5", the one down is 2.5", which is what im using in my indexer with the brake. theres a few other makers doing these "standard" motors, but parker is easy to get and have lots of online info.

[JerryFlyGuy]

ahh... math at 1am w/out a calculator.. sorry guys I should have just gone to bed.

I'm not too concerned about frame size. I think the upper limit would be ~10" OD on the B axis and ~14" OD on the C axis. All just guess's at this point as I haven't fully designed the head, it could be made to fit pretty much anything.

Anyway, re: my previous numbers. Even w/ my brain-fart math corrected.. Steppers still don't compare, as far as I can see? The only significant change is that the AC servo, when reduced is only rotating at 264deg/sec. So, limit to limit on the C head in about 0.8 seconds.. should do the trick [excessively] fast enough.

ihavefish, do you have a site link for these parker motors? I'll have a scan, however at $600 for a small unit, I don't expect them to be in my price range, comparably i could/can/have bought AC servo's w/ a reducer for as low as $50/unit up to $250/unit. As far as cheap motion control stuff.. ebay is still king, if you know what your looking for. Dollars to dollars, it just doesn't compare. [which is why I'd pretty much written them off at this point]

JFG

[mike_Kilroy]

OK, good examples. Steppers do have more holding torque than equiv size steppers at very low speeds and yes, that is what is required for these heads, so yes, they seem like good runner for the job.

But what is being ingnored is that the servos have much higher torque available intermittently for holding the head axis in position against an external torque disturbance like machining..... Using your example of 280oz in continuous and peak Nema 23 stepper (2.3"body square x 90mm L without feedback), the equivelant size servo is for example a Kollmorgen AKD21 motor (http://www.kollmorgen.com/website/co...akm_series.php) with feedback included (2.3"sq x 95mm L) and true has 70/214in-oz continuous/peak torque - but an apples to oranges comparison since the 'motor' portion of it is much smaller than the motor portion of the stepper.....

The same series longer stack AKM24 @ 152mm L has 202/683in-oz cont/peak torque.

This more than 2x peak torque can be used to hold that head against machine forces 2x better if they are intermittent forces... So yep, it depends on the application which is best.

Then Jerry's direct drive build it in frameless motors can go a tad further yet. For this 280in-oz comparision, a Kollmorgen KMB 10x03 (http://www.kollmorgen.com/website/co..._frameless.php) is 2.3"dia x 84mm L and rated 167/522in-oz cont/peak. Again, if the peak can be used along with the continuous it may provide a stiffer axis.

Then Jerry's planetary gearhead is another option.

All of these have their advantages and disadvantages depending on what is required. At near zero speed, the high torque of the stepper is formidable but if you need to go a few thousand rpm the servos take over fast.

[ihavenofish]

ahh... math at 1am w/out a calculator.. sorry guys I should have just gone to bed.

I'm not too concerned about frame size. I think the upper limit would be ~10" OD on the B axis and ~14" OD on the C axis. All just guess's at this point as I haven't fully designed the head, it could be made to fit pretty much anything.

Anyway, re: my previous numbers. Even w/ my brain-fart math corrected.. Steppers still don't compare, as far as I can see? The only significant change is that the AC servo, when reduced is only rotating at 264deg/sec. So, limit to limit on the C head in about 0.8 seconds.. should do the trick [excessively] fast enough.

ihavefish, do you have a site link for these parker motors? I'll have a scan, however at $600 for a small unit, I don't expect them to be in my price range, comparably i could/can/have bought AC servo's w/ a reducer for as low as $50/unit up to $250/unit. As far as cheap motion control stuff.. ebay is still king, if you know what your looking for. Dollars to dollars, it just doesn't compare. [which is why I'd pretty much written them off at this point]

JFG

parkers "example" spec sheets. i say example cause they have many many other sizes and its all custom wound to your spec.

http://divapps.parker.com/divapps/em...lessMotors.pdf
Parker - K SERIES
Bayside Motion Group - Product Specifications


heres an online store with some "stock" motors and pricing.

Automation Services, Machine Shop Services, Visions System Services - Machine Vision Sytems, Pick & Place Systems | Dynamic Design Solutions

youll notice that the price doesnt go up all crazy like on the bigger motors. youll also notice that the biggest motors seem to have wrong prices, thats cause those frame sizes are inch, so they arent actually bigger. youll also see that internal commutaion is expensive, and you may not need it depending on the drive.

089 is 3.5" - 1" stack is $600 range with com
127 is 5" - 1" stack is $700 range with com
178 is 7" - 1" stack is $850 range with no com
254 is 10" - 1" stack is $1600 range with no com

-025 is 1/4" stack
-050 half inch
-100 1"
etc etc.

windings are *Y2 for wye with commutation, *Y1 with no commutation. D is delta.

the * first number/letter is thw winding, lower is less voltage per 1000rpm. since you want low speed, low currect, a higher number is better, 9 and up.


MCS Motion Control Systems, Inc. also sells these, but ive found their pricing a little higher, and youve got to call them to get any good info, their site does list all the possible frame sizes and some example toque charts though.


of course, this is all with the intention of direct driving. if you choose any sort of gearbox, you can use just about anything. ive got a box of parker nema23 motors here. with 160v they will run up to 7000rpm with 100oz-in continuous torque, 200oz in intermittent for 10 seconds in a well ventilated area. put it in a housing with a good fan and that can probabaly be extended to a minute or so intermittent at 200 oz in, and 120 continuous. geared down 100:1 youve got 70rpm, and near your 100NM torque goal. they have 1024 line renco encoders, so youll have an effective 409600 counts per revolution. of course a gear box will have some backlash to deal with. these motors are 5" long including the encoder cover.

as mentioned, xzero will be getting in some cheap servos soo. canadian made drive and power supply, with a motor from taiwan. 60mm frame and 80mm frame. the little one is 3000rpm at 1.27NM at 130V. effectively the same power as my parker, just half the rpm, so half the gear down needed. they are probabaly better suited too, since most worm gears wont take a 7000rpm input speed without a fight. dont have exact pricing yet, but they will be about the cheapest "new" ac servo system avaialble.