Microstepper drivers

I am trying to retrofit a small cnc router for wood and soft materials and are not quite shure about wich driver to choose. Gecko G201x drivers have a fixed 10 microstep wich they claim to be the right one because of tolerances in the stepper motor, they have support and documentation. Arc Eurotrade have another type 4,2A with 128 microsteps an with no support and sparce documentation. It seems to be the same models available on Ebay all made in china.

My stepper motors have a rated current of 2,8A.

Does the microstep resolution on 10 or 128 give any benefit or is anything over 8 or 10 not usable with steppermotors.

Is there a difference in stability between these two types.

Do anyone have experience with any of theese or any other suggestions.

Having put a Gecko G540 on my Sieg KX1-NU; I'm not an expert, but not a real novice, either:

- steppers "snap" to each major step.

- micro-stepping is a way of subdividing these steps by keeping the magnets going in a "certain way".

I'd be really surprised if the Gecko (or, anything else) could accurately subdivide by balancing magnets. Sure, maybe those 10 "microsteps" are really part of the 1 major step; but the probability that they are accurately subdivided under load is not something that I'd bet on.

128 subdivisions? wow.

In my understanding, the driver does the microstepping, the stepper motor just follows along, so the motor has little effect on the micro stepping. (I could be wrong)

With the EMC2 (LinuxCNC) software that I use, I had to enter the "10" microsteps into the config in order to get it working.

Got to run - showing my KX1 off to another club member soon!

Another JohnS.

Microstepping is not a precise angular subdivision of whole steps, it is an approximate subdivision which has the considerable advantage of smoothing the rotation of the rotor, reducing noise and vibration and avoiding some resonance issues. For most uses 8 or 10 steps is about the sensible maximum.

Most motors have about a 5% tolerance on whole step angle, however the microstep interpolation accuracy is probably much worse, the in-between steps depend on the variation of flux between whole steps which varies with different motors. Don't think of microstepping as an electronic gearbox, it is more an intermediate stage to make the stepper look like a two phase motor with quasi sine/cosine drive.

Billy.

Remember as well that increasing microstepping reduces torque.

Also the more steps you put out the faster the computer and controller has to be. Some machines using servo drives that work in 1,000's of steps can't use a program such as Mach without resorting to an add on board that can handle the speed.

Anything over 8 microsteps is a waste of time.

John S.

Yes and no. Torque depends on the angle between the present position and the commanded new position according to a sine function. Thus for example if you are 1/8 microstepping and wish to move one microstep the torque will be reduced to 20% of holding torque. However if you wish to move say 4 microsteps at once the torque available will be 70% of holding torque and if you wish to move 8 microsteps the full holding torque will be available. Remember the motor will only produce torque as a result of a position error.

Russell.

The torque depends on the method of microsteping. There are two main methods, full current to one coil then step the next coil incrementally to full current for the half step position so both coils are at max current in the half position and approximatly sine and cosine drive where maximal current is on the full step positions.

The torque is not just the current, the half step position has maximum air between the poles so the reluctance of the magnetic path is maximal and torque therefore lower as John said. The sine and cosine drive concept is just that- it does not determine the rotor angle, just a simple concept. Stepper motors are designed to step and hold, what happens between steps is not pecisely controlled at all which is why microstepping is not fine angle control but smoothing the motion between whole step positions. The drive manufacturers can only approximate to the required wave drive because they don't know the exact function needed which will vary between motors.

Billy.

Hi Ralph,

I have recently built a drive for my lathe top slide based around a 2M542 controller from fleabay. They are microstepping and the number of microsteps is selectable - I used 8 microsteps giving a total of 1600 pulses per revolution of the motor.

I haven't any other experience but this driver worked fine, was relatively cheap and seemed to be quite well made with good specs, i.e opto-isolated inputs and adjustable current limiting.

Regards, Hugh

Thank you for all the input. I have now decided to buy the m542 microstepper from ebay and will run it with Mach3 and a new ethernet controller from CS lab so i can get rid of the old lpt port.

Ralph - good for you.

I agree with the lpt port "issue", but, just have integrated an MPG pendant from cnc4pc into my EMC2(LinuxCNC) machine, so now i have *two* lpt ports.

I was using a USB game controller (less than the cost of 2 pints of beer) for hand-held control of my mill, but there's an interesting disconnect issue for the human-factors people to study, so I went with a rotary dial controller. The "joystick" approach did not work for me.

The MPG pendant is a fantastic bit of kit, but it (generally) needs a parallel port!

Oh well. Good luck with your project.

The Canadian JohnS.