Help evaluating this servo motor

[WoodSnarfer]

I found three CMC servo motors, new old-stock, model 3515, winding type "C". Here's the datasheet, if anyone is interested:

http://www.torquesystems.com/downloads/platform3500.pdf

I talked to a tech at CMC, and he said that at 50VDC, this motor will run at approx 3500 RPM, and produce 4.0 in-lb of torque. I'm trying to figure out if this is good enough for my (approx) 33" x 48" x 7" wood router design.

So, here is what I figured:

Power supply: 50VDC, 20amp

unloaded RPM: 3500
80% of RPM: 2800
motor torque sensitivity Ke: .99 in-lb per amp (~ 16 in-oz/amp)
maximum continuous current: approx 4 amps
maximum continuous torque: 64 in-oz (4.0 in-lb)

example gearing: 4:1
example screw lead: 0.20" (5 TPI)
example screw efficiency: 0.907

geared torque at screw: 256 in-oz (16 in-lb)

screw rotational speed (loaded): 700 RPM
axis linear speed (loaded): 140 ipm

screw rotational speed (unloaded): 875 RPM
axis linear speed (unloaded): 175 ipm

axis thrust (pi/8 * TPI * in-oz * efficiency): 456 lbs
power delivered to load (RPM * in-oz / 1351): 133 watts

So, 456 lbs of thrust sounds like pleny, right? Next, I checked the Nook catalog and see that for a typical ballscrew, they say it takes 0.035 in-lbs of torque to raise a 1 lb load.

Okay, so that means after my 4:1 gearing, if I'm using this ballscrew, I should be able to raise 456 lbs, which equals my thrust calculation (I used this to back into the screw efficiency value, just to double-check my math).

My gantry could end up weighing around 200 lbs. I realize that I'm not "lifting" the gantry, but just "pushing" it on low-friction linear rails. I will be lifting the z-axis, but that should weigh around 35-40 lbs.

I'm more worried about inertia.

Is this motor going to work?

Thanks,
Chris

[ViperTX]

Chris, any more of these motors available. Took a quick glance and it looks correct, you've selected a torque that is in the middle of the useable bandwidth and available from low to high rpm.

Inertia is what smooths (eliminates cogging), and it's also what you need to overcome whenever you stop or want to change direction. I'm estimating that it will only affect your estimates by less then 1 percent.

[Al_The_Man]

I have used these motors fairly extensively, they come up on ebay all the time, they also show up as EG&G I don't know who actually makes them, maybe they were bought out by CMC, they also sometimes come with a tach or encoder, it is best to look for the ones with the rear shaft as it is very easy to fit a Renco encoder to. They also work well with the A-M-C line of DC amplifiers. Usually the Galil amplifiers that come up on ebay are made by AMC and are usually plentifull.
Al

[Torsten]

Your Powersupply seams a little big for 3 of those motors you
would only need about 8-10 amps.

[WoodSnarfer]

Great! I am trying to hit my design specs without slipping into my usual (expensive) over-engineer everything out of fear...so a 10A power supply is even better.

(ViperTX...do an ebay search on CMC servos, that's how I found these piecemeal, over a two month period of looking...Al says they are also EG&G, a good piece of info to have when searching).

Thinking about this some more...if I convert to SI, the 456 pounds of thrust is 2,028N.

If the gantry weighs 200 pounds, that's 75Kg, or 735N of downward force due to gravity.

If using linear slides, the coefficient of friction is a ridiculously low number (something like .004), so the force of friction is pretty much negligible...maybe 3N...so net motor (applied) force on the gantry is 2,025N.

If I calculate acceleration, Force = Mass * Acceleration, or A=F/M, and A = 2025/75, or 27 meters/second/second

If my desired jog speed is 175ipm (or 0.074 meters/second), it should take about 3 milliseconds for the gantry to accelerate to the jog speed. Turnaround time should also be similarly "instantaneous".

Boy, I hope I've done this correctly.

It would seem that the only 'real' forces I have to worry about are:

1. Friction of the cutting bit through the material being cut, at the desired cutting speed.

2. Overcoming gravity for the z-axis.

Intuitively, it would seem that I've got plenty of power at the router for woodworking or light aluminum. Provided I'm not doing something stupid like a 2" deep cut in maple, using a 3" molding bit.

What was worrying me was some of the reading here I've done about people buying much larger servos. I think I saw specs like 200 oz or more. Maybe they were talking about peak torque?

My little motors are 64 oz / 256 oz after gearing...so either I missed something, or we are talking about continuous versus peak torque, or everyone else is over-spec'ing their motors.

[Al_The_Man]

According to my calculations, if you are using a .2" lead ball screw to move the gantry, with a 4:1 reduction at 2800rpm will give you 140"/min feed. If your gantry weighs 200lbs the important thing is to keep the inertia ratio between motor and load to under 10:1 at the required accel. decel. rate.
But the advantage with reduction is that the drive torque required due to inertia decreases by the square of the reduction, so at 4:1 you have reduced the drive torque required by 16.
Al

[WoodSnarfer]

Yep, 140 ipm was my 'feed' rate, 175 ipm was the 'jog' rate (in my original post).

I looked at Galil stuff out on eBay and instantly got confused. I had planned on using Geckos...but if anyone has a recommendation for an all-in-one package that can drive 3 axes for less than $342, I'd love to hear it

Would this do it? (and listen to Mach2?):

http://cgi.ebay.com/ws/eBayISAPI.dll...tem=3871207413

So, Al, do you "like" these 3515 motors, or are they dogs?

-Chris

[Al_The_Man]

Chris, I have had no problem with them, Where I have used higher voltage supplies, I have used the AMC amps and they limit the current to whatever your motor is rated at, I forgot once and burned one out when it stalled. due to a failed overtravel limit, otherwise no problem.
Al

[unterhaus]

Originally Posted by Al_The_Man I have used the AMC amps Al

Al,
What have you driven these amps with? Galil?

[Al_The_Man]

Galil and Acroloop PC based motion control cards.
Al

[2muchstuff]

A couple of years ago I tried searching for information on these motors and came up empty handed. Finally some kind of information. I have a whole box of these, model # ME-3515-131E. I don't know what the 131 stands for but I would assume that it is an "E" wind. They have 4 brushes spaced at 90 degree intervals unlike a regular motor at 180 degrees apart? I will say this, they have plenty of torque. At 12 volts with a 10" wrench on the shaft, you cant' hold it back. The shaft encoder that is attached on the back of the motor is also different.

[WoodSnarfer]

Yes, you have the 'E' winding...even more torque sensitivity.

I'm not sure about the four brush configuration, maybe the motor was set up for some sort of switchable voltage / direction wiring from an external source (?)

I'm on the road right now, but when I get back home I'll look at the encoders on mine. I'm trying to figure out the pin-outs, myself. When you say your encoders are 'different', what do you mean by that?

Thanks,
Chris