servo vrs. stepper

[ints99]

Based on experience, what is the typical speed difference of similar torque (lets say, 100 oz.in) steppers compared to servos?

And, what are the optimal reduction rates or servos, for steppers?

Rgrds,
Indrek

[ViperTX]

Servo have their rated torque over their RPM range. Steppers's torque goes from max at low RPM to minimum at highter RPMs.

[ger21]

100 oz is 100oz. But, like Viper says, steppers will have less torque as rpm's increase, where as a servos torque is constant. Steppers are usually best driven directly, unless VERY slow speeds are required. The optimum reduction ratio for a servo would depend on the machine. Load it has to move, speed required...and other factors, all come into play.

[ints99]

100 oz.in. is torque, not speed?

Ok, if the only/most effective way to use steppers is 1:1 reduction (and in any case no more than 2:1, but can it be useful in other way end, i.e. like 1:2 ?), and as in the same time servos are having constant torque over rmp range, but they have (can have) generally n times higher RPM (but using m times reduction changes the actual outcome), then we can state, that similar torque rated servo is having n/m times faster actual speed but has m times more torque? What are typical rates for m and n ?

Or other way end:
if preferable stepper is 600 oz.in, what kind of servo (oz.in.) and reduction I should use to have similar end speed? Typically, as probably every single stepper and servo motor has own exact specs? But are there some kind of "general rules"?

Or:
of course every mechanical setup has on need for torque and speed. but in servo case I can use not only bigger motor, but also play with reduction to select suitable motor. How I will decide, when to stop adding RMP and select bigger torque servo? Or why not use 1:1 reduction with servos too?

[Kookaburra]

IMO,

There is no comparison. Servo's are the only way to go. More expensive at the beginning, yes, but cheaper in the long run as you will not loose steps and destroy a job. Steppers are a thing of the past. Servos are cheap enough now to consider.

[ynneb]

There is no comparison. Servo's are the only way to go. More expensive at the beginning, yes, but cheaper in the long run as you will not loose steps and destroy a job. Steppers are a thing of the past. Servos are cheap enough now to consider

Its funny how they are bringing out steppers these days that they then put encoders on them, and call them Smart steppers / hybrid steppers etc. Essentially they are working the same as servos. Why would you bother with the effort when you can just use straight servos anyway.

I can testify first hand how my servos saved a cutting job I did when the spindle got bogged. When the spindle raised, the machine jumped back to where it should have been and continued to cut the job. I just wound down the feed rate and the rest of the cuts were fine.

Kooka, I am still waiting very patiently for your new servo drives to be released
Whats the hold up?

[ger21]

Benny, can't you also testify how the following error on your servos was giving you that problem?

[ger21]

A properly set up stepper system will never lose steps. An improperly setup servo system won't run very well. There is nothing wrong with steppers. Mariss from Gecko says that up to ~150W (I think, can't remeber the exact #), steppers are a better choice. If you need more power, than servos become a better choice. Steppers are cheaper, easier to setup, and don't need the added gear reduction, which costs more and adds more complexity.

[ynneb]

Hmm, I am not sure what you mean Jerry. When I was doing the cutting it was into 1 inch poly something. As this stuff heated up it actually bacame hard and wrapped around the top of the spindle stopping the machine from moving along the material.
To say that "A properly set up stepper system will never lose steps" is not like you. We have both heard of many instances where this has happened. Servos will correct the loss though, if they are freed., steppers will not.
I dont see how that could be called impropperly set up?
Is this what you mean?

[ints99]

A properly made setup for steppers or for servos - what this actually means IRL? Just about selecting right sized motor and reduction rate pluss ensuring that everything runs smoothly plus measuring correctly existing backlash, or what?

Or, what in addition you must take care speaking about servo setup?

Rrds,
Indrek

[imserv]

To somewhat compare a stepper to a servo, compare the stepper rating, say 250 oz-in, which is holding torque, to servo max torque divided by about 2. Use the belt/gearing reduction when you make the calculation.

Here are the numbers for our Globe servo motors:

20 oz in continuous, 178 max, 2800 rpm
2.8:1 belt reducer yields 60 oz-in continuous and approx 500 oz-in max, at 1000 rpm.

1000 rpm happens to be a sweet spot for desktop machines like the Sherline, Taig, the mini and micro imports, which use 20 threads per inch screws. It is also just about perfect for a 10 tpi acme screw which is a good size to use on a benchtop CNC router. 50 ipm on the mills and 100 ipm on the routers is about as fast as you can go when using brass or delrin antibacklash nuts on these machines.

If you have ball screws and rolling slides ( linear ball guides/roller bearing wheels ), which both have reduced friction, a 5 tpi thread is very common and that multiplies the above speeds by about 4, all from the same 1000 rpm servo motors w/reducers.

some additional comments:

1) These motors are often driven by our 150 watt servo drivers and provide a stiff and accurate positioning system, well within the desired tolerances for all the machines I mentioned, and at about 1/2 the price of other servo drivers.

2) Steppers usually have so little torque at 1000 rpm that they can barely move the machine table or gantry. The result is that a stepper makes a nice slow machine with little high speed cutting capability.

3) Servos have additional dynamic capability that does not show in the above calculations. At any time a need for sudden acceleration or additional cutting power in any axis can temporarily draw on the surge capability in a linear power supply. This does not cause a loss of position on the machine

Fred Smith - IMService
http://www.cadcamcadcam.com

[ger21]

Originally Posted by ynneb Hmm, I am not sure what you mean Jerry. When I was doing the cutting it was into 1 inch poly something. As this stuff heated up it actually bacame hard and wrapped around the top of the spindle stopping the machine from moving along the material. To say that "A properly set up stepper system will never lose steps" is not like you. We have both heard of many instances where this has happened. Servos will correct the loss though, if they are freed., steppers will not. I dont see how that could be called impropperly set up? Is this what you mean?

I can't find the thread with the problem you were having. Maybe I was mistaken, but I thought your servo's were lagging a bit behind due to the following error, resulting in an incorrect cut path in 1 spot of your program. Or was it a different problem?

A properly "designed and setup" stepper system should not lose any steps. If steppers miss steps, it's because either the stepper is too small for the application, or the machine is binding somewhere (incorrectly setup). Most of the people you mention that are losing steps are using small (~100 oz) motors, and are trying to go faster than they are capable. Try to use too small of servos, and the drives will fault, which may still ruin your parts.

There is nothing wrong with steppers, if they are sized properly for the application. You should be able to run a stepper machine all day long and never lose steps. If you do, either the steppers are too small for the machine, or there is another problem. Steppers are NOT the problem.