Full Steps -vs- Micro Steps

[DJB282000]

My Question here is how do i know if i sould use full step drives or micro step drives.

I am looking at buying Geckos but i am not sure if I should buy 202's or 212

I would have posted this under the Gecko threads but i think that it applies to any drive.

Tks for the help.

[ger21]

Generally, microstep drives run much smoother than half/full step drives, because the microstepping can greatly reduce or eliminate the low speed resonances that can lead to lost steps.

The Gecko G212 is a G202 with a step multiplier. The only reason you would need the step multiplier, is if your computer can't output steps fast enough to reach the machines top speed. For instance, with a G202, and a 10 tpi leadscrew, you'll need 20,000 steps per inch. 200 steps per rev, times 10 microsteps per step, times 10 turns per inch. Mach3, at it's lowest speed setting, can output 25,000 steps per second.

That would give you 1.25inches per second, or 75 inches per minute. If that is the max your computer can output, and you don't want to change to a higher lead screw or use gearing, then the G212 can multiply each step the computer sends and allow faster speeds.

Does that make sense?

[DJB282000]

I am using lead screws that are for the most part 1" /rev and stepers that are 200 s/r. If I under stand you correctly i can go with the 202 have no problems and still gear down?

What would u consider a computer that is too slow an would actually need the micro steps?

[ger21]

Depends on the software, and the entire setup. Some guys use 1/4-20 leadscrews and need it.

[DJB282000]

I have been reading on this site for a few months now and I am trying to get parts together but I think like most of the members on the zone I really can not afford to buy parts just to see if they are the correct ones or not.

I understand the math and can figure that but is there a systematic way of figuring out if you should be using full, ½, 5, 10 steps.

I understand that this is going to depend on each persons own set up. But what should I be looking at as I try to make the decision. Final Speed, TPI, Gearing, Computer, End Use (plasma, torch, mill, etc.), Precision.

Do more steps truly give more accuracy (not considering the rest of the machine)?

I am still confused.

[ger21]

More steps are more accuracy, more microsteps are not.
10 microsteps are best when using Geckos.

What kind of machine are you building? What size? How fast do you want to go?

[DJB282000]

Im looking at about 4' X 4' cutting area for plasma and router. I am not that concerned with speed (some would be nice) but i would like to keep the acuracy as high as posible taking in to concideration funds and all.

If i read the last post correctly it would be better to gear down and use more steps per/" than go to micro steps. with the understanding that you still want you final speed to be within the stepper motors range.

[MrBean]

Since I have tried both microstep drives and full step drives on the same machine, I would now not consider using a driver without microstepping.

The difference was dramatic. Even if you use gearing the motor/assembly, will still want to resonate using full steps. Microstepping helps to eliminate some of that resonance.

Terry.

[DJB282000]

I think that I am going to end up spending the extra 40 buck and buy the 212 that way i can use them like the 202 or any of the micro step configs.

Does a micro step actually break up the steps in the motor? So one factory stepper motor step (1.8 deg.) would be broken down into 10 smaller steps (0.18 deg)?
--or--
Does the drive mulitply the steps coming into it and then output those steps * 10? 1 step into the drive = 10 out to the motor (18.0 deg)?

If you just slap me this will most likely all clear up.

[MrBean]

if you run a 200 step/rev stepper using 1/10 microstepping, it takes 2000 step pulses to make one revolution. Thus a 1.8 deg stepper becomes a 0.18 deg stepper.

[ger21]

Originally Posted by DJB282000 Im looking at about 4' X 4' cutting area for plasma and router. I am not that concerned with speed (some would be nice) but i would like to keep the acuracy as high as posible taking in to concideration funds and all.

Everyone thinks they don't need speed, but it's quite important. I've never used a plasma cutter, but from what I've read, I think you need at least between 100-200ipm. And for a medium size router like that, again at least 100ipm should be your goal. Anything less can get painfully slow, but more important, is that you'll get poor tool life, and burning isf your not cutting fast enough.

It's easy to get .001 accuracy without needing to go to the G212. Use a 5 turn per inch screw. With a 200 step per rev motor, that's 1000 steps per inch, or .001 per step. 200 ipm when using 10 microsteps would be 33,333 steps per second. Mach should be able to do that no problem with about a 1.5Ghz PC. A much bigger concern, though, would be if you can spin that screw at 1000 rpm with your steppers, and will screw whip be a problem at that rpm?

If i read the last post correctly it would be better to gear down and use more steps per/" than go to micro steps. with the understanding that you still want you final speed to be within the stepper motors range.

No, you always want to microstep to get smoother running motors. When planning the machine's resolution, don't even take the microstepping into consideration.

I think you're placing too much importance on accuracy. In the example above, I pointed out that you can easily get .001. That's probably much more than you need for plasma, and even .002 would be fine for a wood router. Keep in mind that you also have to build a machine that doesn't flex .001 anywhere, and is in perfect alignment to within .001 everywhere. If it flexes, or is not setup up accurately, then the leadscrew resolution is meaningless. And a router can easily generate enough force to move a machine .001 or even much more.