to gear or not to gear

[cueshark]

i have read that you do not usually gear steppers. that said i am using 5/8x8 acme screws. i was thinking of changing over to 1/2x8 2 start or1/2x10 5 start screws. what would be the difference if any [i can not see a difference] of just using the toothed pulleys & belt to go either 2 to1 or 4 to1 instead? thanks in advance for any answers or opinions. greg

[bobw63]

Hello Cueshark,
Acme screws have a low efficiency of converting torque to linear motion. This loss is due to friction - sometimes as low as 45%. That means half of your available torque is not available for moving anything. The benefit of going to an increased pitch is two fold; increasing the efficiency (most of the time) and allowing a higher inches per minute from the router. All screws have a limit on how fast they can spin without producing whip, a destructive oscillation of the screw itself. If you have reached this critical speed on your existing machine you can go to a multistart screw which will yield more inches per minute at the same screw rpm. On the downside,you lose mechanical advantage of the higher pitch screw and may not be able to accel/decel as fast, if at all.
Bottom line - if you can lose torque to gain screw rpm and still move your load while not exceeding the critical speed of the screw, it should work. Increased screw speed may be rough on the nuts (acme).

[harryn]

I will throw in my marginally informed opinion, which will certainly bring on some real knowledge from others.

There are a number of reasons to gear / not gear down your drive system. The first step, IMHO, is to figure out what resolution of motion you desire. As a simple example, suppose your screw advances the nut by 1 inch every 8 turns - in some ways, this can be thought of gearing it already.

Many stepper motors have 200 steps per revolution, so

200 steps / rev x 8 rev / inch = 1600 steps per inch.

In simplistic terms, this means that if you use "full steps", then the best possible theoretical ability for your system to resolve a motion is 1/1600 or about 0.0005 inches.

If you add microstepping into the equation, now you have (example, 10 microsteps / step)

1600 steps / inch = 10 microsteps / step = 16,000 micro steps / inch. In THEORY, you now have the ability to resolve motions of 0.00005 inches (approx).

Obviously, you really don't have this in the real world. Part of the reason, is that stepper motors and gears are not made perfectly round, and don't take perfectly even steps as they go around. The typical stepper "step" is +/- 5% variation as it circles, a few will spec +/- 3%. There are also plenty of other imperfections in the typical DIY build which will limit motion control.

Gearing down the motion of the stepper also helps gear down these errors. It also can introduce its own errors, as belt systems almost always have some backlash, and if you are like me, your DIY hole drilling is not perfect. Lastly, the stepper motor often have kind of mediocre bearings which are not designed for both axial and radial load. Some exceptions to this are the from Danaher Motion - such as the CT series stepper motors. These have very high quality bearings, and are designed to directly couple to the shaft.

Depending on how you add it up, sometimes you can put more $s into the motor, and save on other parts.

There is also a torque requirement to make your motion work. Some motors have better torque at low speed (often) but I have been told that many motors do not run smoothly at low speeds. This sort of forces people to run at higher rpm than they prefer to, and forces gearing down. I am still not so clear on how big of a problem this is, and which motor / driver combos this comes from.

There is another effect I have heard of here on the forum (I think from Al the man ) related to a ratio between motor rotor inertia and load inertia. I am not clear on if this is only a servo motor effect or also a stepper motor effect, nor its real significance.

[ger21]

I'm with Bob. Multiple start is more efficient, so even though everything appears equal, the mulitple start screw will deliver more force due to it's higher efficiency.

[cueshark]

thanks for the info. i had foregotten about the efficiency part. the multi start [5 start] is near 2x as efficient as my single start. thanks greg