[Frank_NY]

Hi everyone. New guy here.

I have some projects to do and was planning to farm out the CNC work but with all the prototyping and setup charges it might be more cost effective to get a machine myself. Of course, once I started looking on eBay and some of the commercial sites one thing leads to another and now I'm thinking of building my own.

So here I am now, with a design for a CNC router dancing in my head and a bunch of questions. I'm not a mechanical engineer and have never done anything like this, so I think I'll fit right in.

My question (right now) is about motor selection for the Z axis. Hopefully this is the right forum for this question (too many forums to choose from).

It seems from everything I've seen so far that all the motors on a CNC machine are sized the same. I can understand that the X and Y axis would need equal amounts of power (or torque) to pull the router through the material being cut. However, is it really necessary for the Z axis to be sized the same?

That is, does plunging a bit into a given material require the same effort as moving the bit laterally through that material?

[Zumba]

The issue is gravity, which pulls the Z-axis down.

You'd be surprised at how little effort it takes to push a heavy gantry running on high quality ball bearing linear motion guides. Often less than what it takes to hold the Z-axis up. Some people just add higher gearing to the Z-axis (3 to 1 pulley ratio for instance), but I think it's important to make the Z-axis as fast as possible, especially if you're carving 3D stuff. A slow Z-axis will add hours to an intricate carving.

Besides, upgrading motors is cheap. If you've already spent $300-400 on Geckos, $100-200 on the power supply, and $150-250 and the X and Y axis motors, it'd be pointless to save $25 and downgrade the Z-axis motor.

[Frank_NY]

Thanks for the response, Zumba.

So the larger force to overcome is movement of the axis, not resistance from the material being routed?

I would have expected the opposite. Particularly if, as you point out, a low-friction, quality bearing is being used in the design.

[Al_The_Man]

Alot of attention seems to be paid to side forces, although they exist and should be considered, the torque demands on servo a motor exist:
When at rest.
When accelerating at the demanded rate.
When moving at required constant feed speed (side cutting forces may exist).
And when decelerating at the demanded rate.
In most cases you will find that the greatest demands occur at the accel/decel times.
Cutting forces are not usually as great when sharp tooling at the correct feed/speed is used.
In the case of a vertical Z axis, there exists the above forces, but the weight of the axis has to be considered when moving downward as this can actually cause problems of overhauling load due to inertia.
Likewise, when moving up, the axis is now required to accelerate with a different inertial condition caused by the weight of the axis.
This is why many heavy Z axis are counter-weighted.
When a motor accelerates, it not only has to contend with the inertia of the load, but the rotor has its own inertia.
And the general rule is that a system should be designed so that the motor to load inertia does not exceed 10:1 and ideally be 5:1 or less.
Apart from the system weight (table+part weight/gantry weight) , the factor often influencing this to a greater extent is the accel/decel rate.
Fortunately when gearing, the inertia is reduced by the square of the reduction.
Motor sizing should always consider the accel/decel rate.
Al.

[Frank_NY]

Alot of attention seems to be paid to side forces, although they exist and should be considered, the torque demands on servo a motor exist: ...

Thanks, Al.

There's obviously a lot more involved from a mechanical engineering perspective to correctly size the axis motors, rather than taking the shotgun approach of picking a package from HobbyCNC and keeping your fingers crossed.

I, of course, not being a mechanical engineer, will have to use the shotgun method.

[ger21]

I, of course, not being a mechanical engineer, will have to use the shotgun method.

Which gives you 3 motors all the same size.

[Al_The_Man]

Almost every servo manufacturer and major supplier has a Motor sizing program, usually free, Allen-Bradley, Seiko-Denki, Kollmorgen etc
http://www.danahermotion.com/design_..._selection.php
Al.