5 or 10 lead ballscrew?

[15mgtar]

Guys which one is better a 5 lead or a 10 lead ballscrew for stepper motor? I'll be using them for y axis 2 meter long. I've got 5 lead ballscrew on my x axis and z axis already was thinking to go with 10 lead for faster drive.

[Zappautomation]

They will give you about twice the speed, but at the same time half the accuracy.
A stepper motor has a +/-5% accuracy on the full step, and microstepping is not linear so the best you can hope for on a 5mm lead screw is 0.025mm. (+/-5%)
Having a 10mm lead will double that to 0.05mm

Originally Posted by 15mgtar Guys which one is better a 5 lead or a 10 lead ballscrew for stepper motor? I'll be using them for y axis 2 meter long. I've got 5 lead ballscrew on my x axis and z axis already was thinking to go with 10 lead for faster drive.

[kostas1]

Hi,
Zappautomation: Given that we use 10mm lead (ie 1610) in 2m long, how about also using some microstepping of G203V drives?Will it regain the lost accuracy due to 10mm lead?

Kostas.

[Zappautomation]

Microstepping is not linear or accurate.
A stepper drivers works in open loop, and between the full step positions is not that accurate.
A stepper motor is only accurate on the full step position.
so by doubling the lead on the screw will half the accuracy.

Originally Posted by kostas1 Hi, Zappautomation: Given that we use 10mm lead (ie 1610) in 2m long, how about also using some microstepping of G203V drives?Will it regain the lost accuracy due to 10mm lead? Kostas.

[kostas1]

Ok, but looking for what is actually microstepping, I found in wikipedia the following opinion:

Microstepping What is commonly referred to as microstepping is actual "sine cosine microstepping" in which the winding current approximates a sinusoidal AC waveform. Sine cosine microstepping is the most common form, but other waveforms are used [1]. Regardless of the waveform used, as the microsteps become smaller, motor operation becomes more smooth. However, the purpose of microstepping is not usually to achieve smoothness of motion, but to achieve higher position resolution. A microstep driver may split a full step into as many as 256 microsteps. A typical motor may have 200 steps per revolution. Using such a motor with a 256 microstep controller (also referred to as a "divide by 256" controller) results in an angular resolution of 360°/(200x256) = 0.00703125° or 51200 discrete positions per revolution. However, it should be noted that such fine resolution is rarely achievable in practice, regardless of the controller, due to mechanical stiction and other sources of error between the specified and actual positions. Step size repeatability is an important step motor feature and a fundamental reason for their use in positioning. Microstepping can affect the step size repeatability of the motor. Example: many modern hybrid step motors are rated such that the travel of every Full step (example 1.8 Degrees per Full step or 200 Full steps per revolution) will be within 3% or 5% of the travel of every other Full step; as long as the motor is operated with in its specified operating ranges. Several manufacturers show that their motors can easily maintain the 3% or 5% equality of step travel size as step size is reduced from Full stepping down to 1/10th stepping. Then, as the microstepping divisor number grows, step size repeatability degrades. At large step size reductions it is possible to issue many microstep commands before any motion occurs at all and then the motion can be a "jump" to a new position.

.

From this I sum up that microstepping affects cnc's accuracy despite the fact that I know there is much discussion and arguments in this theme....

Kostas.

[Zappautomation]

Yes Microstepping effects your accuracy.

I had a customer do an experiment.
He used a standard 1.8 degree nema 23 motor, connected a mirror to the shaft and used a pen laser pointer to reflect the red spot onto a wall about 5 meters away.
(The laser was pointed at the mirror and reflected onto the wall)

The driver was set to a high microstepping resolution, and then commanded the driver to move less than on fullstep, and back again a few times.
Each move he put a mark on the wall with a pen, and doing this he could see the position was not consistent, with a few CM difference on each move.

Saying that even 0.025mm is still a very fine amount of movement, and to achieve higher accuracy and repeatability, will require more than just a better ball screw and motor resolution, the whole machine will need to be of a very solid construction and a lot more put into the design.



ollowing opinion:

.

From this I sum up that microstepping affects cnc's accuracy despite the fact that I know there is much discussion and arguments in this theme....

Kostas.[/QUOTE]

[Al_The_Man]

Often forgotten is that B.S. dia. and pitch also affect the torque proportionately.
Al.

[kostas1]

Hi Al,
could you please explain how this can be done?

Kostas.

[Al_The_Man]

My area of expertise is not mechanical engineering, but I have always understood that the action of a ballscrew is likened to a class 1 lever, this has the fulcrum point between the load and the effort.
If you imagine the centre of the BS as the fulcrum, the pitch would represent the load and the effort the dia.
As you know, on a lever, if you increase the length from fulcrum to load (increase pitch) then the effort advantage is reduced.
Likewise if you increase the length from fulcrum to effort (increase dia), you increase the effort advantage.
I believe I have the interpretation right.
Al.

[Zappautomation]

This is a work in progress spreadsheet i am working on to help work out the rotational torque to linear force.
It is work in progress so not all the functiones work yet and no formating has been done.
Also it does not take into account the inertial mass of the ball screw.
the wider the diameter of the screw, the higher the inertia and more torque will be needed to overcome the inertia.

[kostas1]

But you forgot that G203V perform morphing from microstepping to full stepping...
Microstepping occurs when motors turn up to 3-4revs/sec and above that limit they turn into fullstepping without user understand it...
So in slow speeds you cannot lose resolution and with a 2510 ballscrew you problably have accuracy of 0.005mm = 10mm/2000 steps...
In full steps and high speeds, you return back to 10mm/200 steps = 0.05mm...

Kostas.

[Zappautomation]

Gecko drivers suffer just like all the other drivers on the market.
Microstepping is not linear because it is open loop. so you can only rely on the full step position as an accurate and repeatable position.
Also Morphing has nothing to do with it, Gecko use this, and i cant think of other manufactures that do, but it only changes from microstepping to full stepping after a set speed and then reverts back to microstepping when it slows down again.
I think this is done to give slightly more torque at the higher speeds and a anti resonance function?
I have no experience of this, but i am sure that the new range of drivers that can tune to the motor like you do with servo drivers would have a better response to anti resonance than this technique?
Im not knocking Gecko, they have some good drivers, but i just dont see what all the fuss is about with morphing.

Originally Posted by kostas1 But you forgot that G203V perform morphing from microstepping to full stepping... Microstepping occurs when motors turn up to 3-4revs/sec and above that limit they turn into fullstepping without user understand it... So in slow speeds you cannot lose resolution and with a 2510 ballscrew you problably have accuracy of 0.005mm = 10mm/2000 steps... In full steps and high speeds, you return back to 10mm/200 steps = 0.05mm... Kostas.