Questions -- best practice for DC gearmotor use

I am looking at an application for driving a linear slide (ballscrew + round rails)at a very slow speed, for use in planarizing a horizontal rotating surface made of pitch for optical polishing purposes. I think a DC gearmotor is the right drive for this, but I want to confirm whether I know what I think I know:

My assumption is that a DC gearmotor will have constant torque throughout its operating voltage range. Is this true? For instance, if I get a 90V motor with shaft output of 10 RPM as built, can I run it at 9V and achieve controlled 1 RPM output with no loss of torque (assuming that current is available)?

Is there any advantage in terms of control in using a 180V vs. a 90V motor? Related to the whole question of the motor running at very low speed.

Are there any associated thermal issues with low-voltage operation of the gearmotor?

Does an AC gearmotor work just as well? (I think not, but I may be wrong here).

Any inputs welcome.

specfab said:

I am looking at an application for driving a linear slide (ballscrew + round rails)at a very slow speed, for use in planarizing a horizontal rotating surface made of pitch for optical polishing purposes. I think a DC gearmotor is the right drive for this, but I want to confirm whether I know what I think I know:

My assumption is that a DC gearmotor will have constant torque throughout its operating voltage range. Is this true? For instance, if I get a 90V motor with shaft output of 10 RPM as built, can I run it at 9V and achieve controlled 1 RPM output with no loss of torque (assuming that current is available)?

Is there any advantage in terms of control in using a 180V vs. a 90V motor? Related to the whole question of the motor running at very low speed.

Are there any associated thermal issues with low-voltage operation of the gearmotor?

Does an AC gearmotor work just as well? (I think not, but I may be wrong here).

Any inputs welcome.

Click to expand...

If i'm reading your thought process correctly, i suspect it may be flawed. This may help you: AB-032 : DC motors - Voltage Vs. Output speed Vs. Torque - Precision Microdrives

It the lower speeds the tork will drop off, just oversize the motor and all will be fine....Phil

I could be wrong, but as I see it, a DC motor that uses a PERMANENT MAGNET for the field, will have a direct relationship between the current and the torque. The more current, the more torque. And the less current, the less torque.

My assumption is that a DC gearmotor will have constant torque throughout its operating voltage range. Is this true? For instance, if I get a 90V motor with shaft output of 10 RPM as built, can I run it at 9V and achieve controlled 1 RPM output with no loss of torque (assuming that current is available)?

Click to expand...

The problem that I see here is that you are talking about two different things, the supplied Voltage and the current in the motor and ASSUMING that they are independent of each other. That is simply not true. It is not even possible. At the most basic level, Ohm's Law states that the current (I) is directly proportional to the Voltage (V) and inversely proportional to the Resistance (R).

I = V / R

The current and Voltage are locked together and can not be considered as independently variable unless there is a way of altering the resistance to make it so. Now a DC motor does generate a back Voltage (EMF) as it rotates. And this back Voltage (EMF) IS a way of altering the Voltage/current relationship: it effectively alters the R or resistance of the motor. Unfortunately this change in R is not in the direction which would maintain a given current as the Voltage is reduced in order to slow the motor down. When the Voltage is decreased, the PM, DC motor will go slower and that means that the back EMF will also be smaller. So, the effective resistance of the motor will be greatest at the highest speed and will be less when the motor is running slower. However, the net result, assuming the same load, is that as you decrease the Voltage to the motor it's speed will decrease because the lower Voltage will overcome the lower resistance. The Voltage going down will be the greater factor and the lower resistance the lessor factor so the current will decrease. And that smaller current will translate to a smaller amount of torque.

Another key fact here is the assumption that I made about the load being the same. If a MP, DC motor has it's speed decreased by a change in the load, ie, the load is increased, then the driving Voltage will remain the same, the back EMF will decrease, and the current will increase. That increase in the current will them translate to an amount of torque that is actually greater. So the way that the speed of the motor is also a factor here. I suspect that you have heard this and were confused by it.

There are motor controllers that work around this problem and do keep the torque of a PM, DC motor the same or nearly the same as the speed is decreased by the controller. The most common way for doing this is to drive the motor with pulses of DC. These pulses are at the full rated Voltage of the motor or even higher while they are only ON for a short time and then OFF for a period of time. The torque is related to the current which is determined by that full Voltage, so it remains high. But the motor's speed is slower because the short duration of the pulses is not long enough for it to accelerate to it's full speed.

The ratio of the ON time of these pulses to the time from one pulse to the next one is called the duty cycle and it is that duty cycle that determines the speed of a DC motor. Almost all modern, DC motor speed controls use this principle in one form or another.

This pulse method works well over a large range of speeds but it will start to fail when width of those pulses becomes small enough that the inductance of the motor is enough to prevent the Voltage, and therefore the current, from building up to the full value before the pulse ends. At that point, which will be different for each DC motor and each controller circuit, the torque will then start to decline. This is why some motor controllers will use a supply Voltage that is larger than the nameplate rated Voltage of the motor they will be used with. The more sophisticated ones will only use that higher Voltage at the slower range of speeds: this is due to possible heating problems if it were used with higher duty cycle factors.

The way to go is to use a motor and a motor controller (driver) that are matched to each other and that will provide the needed torque (or HP) at all points on the needed range of speeds. Don't try to re-engineer the world, just buy a matched pair of components. And a DC power supply to match them if it is not included in the controller.

If you only want a single speed just get your gearratio right Adding a timing belt gives you the possibillty to finetune the RPM
Mecanicly slowing down increases the torx anyhow . I would take a AC motor With a VFD if you need variable speed

Peter

First of all, you need to get a servo class motor. Cheap ones will have a small number of slots in the armature and corresponding commutator bars. If you feed them a low voltage to run them at a low speed and put a load on the shaft, the speed will vary, known as cogging. The best motors have a number of features that reduce cogging and they cost much more. There are a lot of them floating around the surplus market, removed from obsolete printers and tape drives. If you give me an idea of the speed and torque you need, I may have a motor that I can send you.

If the power level is low, I would stay with a simple transistor system like the one I made for a 4th axis for my CNC mill. Someone gave it to me because no one knew what to do with it. It had been made for a single model of mill and used a different control system from others. I had worked with the type of Baldor servo motor and made a simple controller with complimentary transistors, which works perfectly.

A couple of general statements-

Permanent magnet and shunt wound motors are virtually equivalent, the PM ones are easier to use.

With a good motor, torque does not drop off with speed. The stall torque with the motor stopped and the voltage adjusted for rated current will give rated torque.

The operating voltage does not change performance but the typical 90 Volt ones sold by people like Grainger are made to a price and lack the smooth operation I think you need here.

Bill

These chaps have a lot more experience with dc drives than I do, but what little know says that the system would work best when the motor is near its center operating speed. As such why not get a gear reduction unit that is closer to your desired speed than try to operate on the fringes.

Tom

Not only is a better class of DC motor needed, Optical etc, etc, etc, you ALSO have to eliminate the gears in a gear motor, the cogging of synchronous toothed belts or of several forms of chains as well.

That leaves resilient pinched friction rollers, flat belts, PolyVee, MicroVee, steel, kevlar, or similar bands, monofilments, or woven cable for zero "cogging". Nor are the lumpy-bumpy imperfectly predictable/repeatable friction and hysteresis variations in legacy "deep" Vee belting good enough.

Not a new need. Plenty of prior-art.

Some of it (woven linen belting) going all the way back to ignorant phonograph technology where the variation became detectable impairment to the fidelity of sound, not-even of far, far pickier light.

A gearmotor is going to want to "watermark" yer work with a subtle pattern. Or maybe not so subtle at all.

That basic. I did say "prior art" existed? Bigtime. "Dig it."

Phil in Montana said:

It the lower speeds the tork will drop off, just oversize the motor and all will be fine....Phil

Click to expand...

DC Not AC. DC typically needs right about HALF the rating of an AC motor where Torque is the prime directive. Can't "slip".

The RESERVE of Torque is actually constant below base RPM. Draw on that reserve? Delivered at once.

Have power enough on-tap back of it? It can be pushed UP to easily 4, 6, even NINE times nameplate at or near locked-rotor. And break stuff.

Or so sayeth Reliance Electric & Engineering on paper. And my impressed eyeballs, and SE grin.. in the real!

A five hoss Reliance RPM III Type TR here has the same shaft size as a TEN hoss 3-P. Is twice the size. And masses five times as much as the 3-P one does.



But yes. ANY "variable speed" application, a good rule of thumb is to oversize by 150% to 200% of whatever a fixed-speed unit calculated to need.

A(ny) nameplate data at a fixed-speed is "BEST CASE". The Goldilocks porridge. Variables are forever at risk of being operated anywhere EXCEPT the "best case".

And then they get s**ty cooling when in direst need of all of having some.

See "blower duty" DC motors. We ain't talkin' no iddy-biddy "computer case" Pabst "muffin" fans here.

Rig on a "blower duty" Dinosaur Current stem-winder is often more powerful than the average residential whole-house central HVAC air-handler gets.

Thanks, all, for the useful input. I clearly have some more readin' to do. Regarding the fineness of the finish, it's maybe less of an issue than might be surmised. The pitch layer gets conditioned after planarizing (like a VTL facing op, basically) by running the polisher with a very flat and heavy disk (granite, zero-expansion glass, whatever specs require) running on the surface of the pitch to smooth and flatten so the workpieces can come off the machine polished to desired flatness and finish. Therein lies the art, when you need 30nm flat or better....

I am looking for a gearmotor that is down in the right target RPM range, but where "re-purposed" is concerned, you don't always have unlimited choice.

specfab said:

Thanks, all, for the useful input. I clearly have some more readin' to do. Regarding the fineness of the finish, it's maybe less of an issue than might be surmised. The pitch layer gets conditioned after planarizing (like a VTL facing op, basically) by running the polisher with a very flat and heavy disk (granite, zero-expansion glass, whatever specs require) running on the surface of the pitch to smooth and flatten so the workpieces can come off the machine polished to desired flatness and finish. Therein lies the art, when you need 30nm flat or better....

I am looking for a gearmotor that is down in the right target RPM range, but where "re-purposed" is concerned, you don't always have unlimited choice.

Click to expand...

Made a few sheckels "back in the day" investing not in the Intel's nor AMD's of the wurld, but rather in the companies that made the pre-pep and finishing machines that manage the silly-cone for intentional dirtying with doping and slicing up and such afterwards. All that high-cost gear goes obsolete at an astonishing rate, fast as they keep shrinking "features" and bumping up percentage yields.

30 nm?

You might need to "get out more" to ascertain what has historically worked well or has never?

Howard Hughes to Noah Dietrick: "Find the experts".

Hacker sez: Now see how TF you can get 80% of their goodness for under 20% of their COSTS and in TEN percent of the time. Do it again, better, faster, and cheaper, repeat, repeat, etc.

Because ultimately? It is TIME.. not money... as is the real "MOFU"

Master Of the F**king Universe.

Doubt me? Go cut a PO for replacement of whatever TIME was wasted two year ago.
Now try to get that order filled with deliverable product.

Not "on backorder".

i dont know much about motors, but would be interested in the whole set up. what media is used for 30nm finish?

dian said:

i dont know much about motors, but would be interested in the whole set up. what media is used for 30nm finish?

Click to expand...

The good stuff.

dian said:

i dont know much about motors, but would be interested in the whole set up. what media is used for 30nm finish?

Click to expand...

Standard optical polishing process with cerium oxide slurry, on the pitch lap. Various sizes and types of polishing machines can be had for this type of process.

This is typical of people who ask for help then refuse to give anything useful about the application, like how big are the parts being polished, what surface speed, how much torque, what speed motor, how many horsepower? Dian asked about the setup. Your response was "cerium oxide". I even offered to send you a motor if I had a suitable one. No answer. Don't bother to post an answer to this because I will not open this thread again.

Bill

What HP and RPM are you looking for? I have a few parallel gear motors kicking around.

If it were me, which is ain't, I'd steer clear of any DC setup and jump on a AC unit with a VFD for fine tuning, plus all the other stuff a frequency drive offers. I am not a fan of DC motors and drives.

Stuart

9100 said:

This is typical of people who ask for help then refuse to give anything useful about the application, like how big are the parts being polished, what surface speed, how much torque, what speed motor, how many horsepower? Dian asked about the setup. Your response was "cerium oxide". I even offered to send you a motor if I had a suitable one. No answer. Don't bother to post an answer to this because I will not open this thread again.

Bill

Click to expand...

Yah. OK. "We" - all of us in our own time and circumstance - DO "come to PM" in hopes of better answers, sooner ...and cheaper.... than doing OTHER research.

But.... "we" DO GET such answers "often enough" that a bit of friction and annoyance - and even the silliest of irrelevant politics - is paid back to us for whatever we had in trade!

Usually several times over ... by end of any given year. Year AFTER year, too.

"Stone Soup" - or "It's PM, dammit!", just work that way.

Yah bring whatEVER yah GOT, yah run with whatEVER yah GET... and nobody has to go away hungry. At the very least, they know more as to where to go NEXT to find enlightenment.

Unless, of course, they get overly fussy about the odd turd in the soup?

Then again? The soup is only "virtual".

So are the turds! Can't actually HARM yah, can they?

So. the trick to going along to get along is simple enough.

Try not to BECOME one of those "virtual turds".

Or at least not all THAT more often than the average?

"Hang in there", Bill. We needja, still yet!

Meanwhile? Could you temporarily up the cranberries in yer diet? Better yet, Johannesbeeren?

Gots us an Austrian recipe for "Hirsch" from the old Gasthaus zum Post - just over the line from Berchtesgadnerland .. and it NEEDS the berries to make that rich brown gravy come out "just so"....

One of you other lot shall have to supply the Venison.

All yah can expect from thermite is "in the book". A fire virtually impossible to extinguish.

And Spring or Demeyere cookware...even Griswold's better Cast Iron .. to cook neat stuff OVER it!

I did say "It's PM, dammit?"

Can't limit yerself to roasting the odd free-range "virtual" ass at a distance, can yah?

Not a bit more nutrition in a virtual ass than in a virtuous one!

Nor pompous?

thermite, do you have such a good memory or do you write stuff down?

9100 said:

This is typical of people who ask for help then refuse to give anything useful about the application, like how big are the parts being polished, what surface speed, how much torque, what speed motor, how many horsepower? Dian asked about the setup. Your response was "cerium oxide". I even offered to send you a motor if I had a suitable one. No answer. Don't bother to post an answer to this because I will not open this thread again.

Bill

Click to expand...

Well, I guess I feel a bit at sea here. I have small scraps of time to look at this information and formulate replies, sometimes just skimming until I can sit down and read the entire set of responses, good, bad, indifferent. My comments regarding the polishing process are at best tangential to the actual issue described in my original post, which is finding a drive scheme using a gearmotor for controllable low speeds. I was perhaps less detailed than I could have been regarding the highly specific application, which I did mention as planarizing a pitch layer on the polishing machine.

For those of you that aren't familiar with polishing of optical flats or other flat optical components, this is typically (and best) done on a continuous polisher, which uses a rotating platen with an annulus for the polishing working surface. For high-end stuff, the polishing surface is a pitch layer, applied on top of the platen, which is generally granite, but in older machines might be cast iron. The pitch get melted, then distributed to desired thickness on the pre-heated platen while it is rotating. At some point, there is enough pitch on the platen, and it cools, then it needs to be roughly planarized. This is essentially a facing operation with a pitch-cutter mechanism, which could be manual, but is better done with motorized drive for consistency. The pitch cutter is what I am building. The cutting tool will be driven across the rotating pitch-covered annulus, from OD to ID, with the machine running at about 5 RPM (48" diameter). To achieve good flatness and P-V finish, I probably need to be able to control the tool advance to about .020"/rev, so this is a process that will take some time. It is simply the first step in preparing the pitch lap for actual work. So, as I noted earlier, the overall poishing process and part sizes have no bearing on this aspect of the process.

I greatly appreciate the offer of a motor. I don't yet know enough to know what I need exactly. Thanks for your insights and offer of assistance.

atomarc said:

What HP and RPM are you looking for? I have a few parallel gear motors kicking around.

If it were me, which is ain't, I'd steer clear of any DC setup and jump on a AC unit with a VFD for fine tuning, plus all the other stuff a frequency drive offers. I am not a fan of DC motors and drives.

Stuart

Click to expand...

Can an AC gearmotor be driven with a variable transformer, and have a "reasonable" amount of torque as voltage is reduced? I think I need shaft output in single-digit RPM range to start with but also some speed control. I assume with VFD, I should get pretty close to desired output RPM in the basic motor specs, so I don't end up with trying to run the motor at 8 or 9 Hz, which I assume(further) might lead to some cyclic on-off motion(?)