[KEVINSTJ]

I have a few questions about steppers
1. What controls the speed of the stepper
2. How fast can you get steppers to go on rapids.
3. Why does everyone use pulleys on the axis motors, why not direct couple the motors to the screw.

Has anyone seen the video of the mini Fadal on you tube, I've watched the video a bunch of times and I try to figure out how he gets the spindle speeds and rapids. Here is the link.

[ame=http://www.youtube.com/watch?v=ISzkthutKnA&feature=youtube_gdata_player]Scratch Built mini version of Fadal's VMC 15 - YouTube[/ame]

[T.W.O.]

Hello

1) The speed of the stepper is controlled bij the stepperdriver.
2) The max speed of the axis depends on the voltage of your power supply, the step of your screw and the quality of your build (linear rails, allignment,....)
3) Pulleys are mostly for increasing torque and/or precision But you can connect the motor directly to the screw without problems.
Youtube: i don't think he uses steppermotors, i think he uses servomotors. With servomotors you can achieve more speed and acceleration but they are more expensive.

greetings

[wizard]

I have a few questions about steppers
1. What controls the speed of the stepper

Stepper speed is controlled by the rate of the step signal going to the drive (in a step and direction drive).

2. How fast can you get steppers to go on rapids.

This depends upon many factors.

3. Why does everyone use pulleys on the axis motors, why not direct couple the motors to the screw.

Actually many stepper driven systems are direct drive. A few reasons not to do that include:

1. The need for higher resolution.
2. The need for greater torque.
3. Belt drives are easier to "adjust".
4. Physical requirements.

Has anyone seen the video of the mini Fadal on you tube, I've watched the video a bunch of times and I try to figure out how he gets the spindle speeds and rapids. Here is the link.

Pretty impressive isn't it. I believe he was using linear ball slides, that helps a lot. Up front engineering and quality drives go a long way.

[datac]

I just wanted to comment more on the question of using belts rather than direct drive.

The nature of home brew CNC machines is such that the builder more often than not simply does not have the engineering experience necessary to really come to an overwhelming successful conclusion. People have got very lucky at times.

Seriously, there is an awful lot of mathematical calculation necessary to even come close to arriving at perfection. Actually sitting down and attempting to calculate all of the specifics related to the goals of the machine.... weights, inertia, loads, temperatures.....

Then add in the fact that many home brew machines are built with a few parts gathered here, a few gathered there..... often parts not initially desired but purchased because they got them cheap. Nothing wrong with that !

Now enter in the cogged belt and pulleys. One poster mentioned that it was easier to "adjust". This is where they are a benefit. If by chance you get your machine completed with perhaps screws that were not your first choice on available pitch, with a few pulley changes and a new belt, you might have a chance at squeaking out a little better performance, be it faster speeds or the other way, finer resolutions.

One final aspect of using belts is that the machine can often be made a little more "compact". Even some OEM machines over the years left a motor or two sticking out annoyingly past the footprint of the machine or work area. YUCK !!!

Any comments about belts reducing accuracy on a machine typically discussed and used in this forum is slightly misleading. indeed, there is going to be a "measurable" amount of added slop for lack of a better word in a machine. However, that measurable amount is so MINIMAL when using a decent pulley and belt brand its almost silly to even mention it, again, with the typical Router, Engraver or Milling machine discussed here. If your really a NASA engineer probing for ideas on a new telescope lens grinder, then you might have a concern.

[arizonavideo]

My servos are direct drive, well the X is.

There are many ways to reach a given performance level. It is perhaps best to work backwards from the work piece to the type of machine needs to produce that kind of part with the tolerance needed.

Some stepper systems can be vary fast or slow depends on the money in and the mass moved.

Money being the key part.

[wizard]

It should be noted that I spent years working on lathes doing single point diamond turning. Every one of the machines I worked on (3 different brands) had belt drives on each and every axis. These where each servo driven. Adjustment of belt tension was key to getting the results wanted.

Now in some cases the belt drives where there to increase resolution as the old CNC controllers resolving 0.001mm were not good enough to reliably produce the parts. However that was only two axis. In the end though that belt drive allowed for adjustment that you would not have with a servo directly coupled to the lead screw shaft. If the coupling in question has a resonance issue there is effectively little one can do to adjust for it. Note that these were with DC servo tech and frankly did not have any of the issues a stepper might have.

Speaking of math, for one of those machines I had the good fortune to spend some time with one of the engineers that designed the hardware. A most impressive individual - way out there. However even a correctly engineered system will give you trouble if the guy working on it doesn't have the chops to adjust it properly. It is surprising that after all the design effort getting something to go is often a function of feel.

Today things have changed a bit. Back then even the encoders where analog and there was much to tune. Today components are much more forgiving. Encoders approximate square wave output and there is little to tune in the feed back loop. Motor drive amps are often digital these days and give far more reproducible results without highly skilled techs tuning them.

I'd have to say that one of the reasons the DIY crowd has so much success these days is that the components are much easier to use and are coupled with far more reliable software systems. Combine that with systems often built with a common set of parts; that is specific Chinese iron, EMC, Gecko stepper drives and specific steppers, you thus have most of the engineering distilled out of the equation. It isn't surprising that these conversions work at all as the task is mostly an issue of assembly.

It is a whole different ball game if one is doing a one off machine build, such as a large router. There engineering is a fresh opportunity. Design issues are often seen in the types of questions we often see in the forums.

I just wanted to comment more on the question of using belts rather than direct drive.

The nature of home brew CNC machines is such that the builder more often than not simply does not have the engineering experience necessary to really come to an overwhelming successful conclusion. People have got very lucky at times.

Seriously, there is an awful lot of mathematical calculation necessary to even come close to arriving at perfection. Actually sitting down and attempting to calculate all of the specifics related to the goals of the machine.... weights, inertia, loads, temperatures.....

Then add in the fact that many home brew machines are built with a few parts gathered here, a few gathered there..... often parts not initially desired but purchased because they got them cheap. Nothing wrong with that !

Now enter in the cogged belt and pulleys. One poster mentioned that it was easier to "adjust". This is where they are a benefit. If by chance you get your machine completed with perhaps screws that were not your first choice on available pitch, with a few pulley changes and a new belt, you might have a chance at squeaking out a little better performance, be it faster speeds or the other way, finer resolutions.

One final aspect of using belts is that the machine can often be made a little more "compact". Even some OEM machines over the years left a motor or two sticking out annoyingly past the footprint of the machine or work area. YUCK !!!

Any comments about belts reducing accuracy on a machine typically discussed and used in this forum is slightly misleading. indeed, there is going to be a "measurable" amount of added slop for lack of a better word in a machine. However, that measurable amount is so MINIMAL when using a decent pulley and belt brand its almost silly to even mention it, again, with the typical Router, Engraver or Milling machine discussed here. If your really a NASA engineer probing for ideas on a new telescope lens grinder, then you might have a concern.

[underthetire]

Wizard, the true diamond turn machines waaaay back in the day were using laser feedback in vacuum. All the slides were hydrostatic, and even the old dog controls were still able to control even pitch and yaw. The entire machines were flooded with oil kept at 68 +\- .001 deg. I hated working on those. The new ones are all linear motors with Sony laser feedback scales, hydrostatic ways with air spindles. We probably have 15 of those. Interesting to find someone else doing diamond turn.

[KEVINSTJ]

Thanks for the replies. I was looking around and I really don't see anyone using Baldor servos and drives. I get a pretty large discount from Baldor and was gonna start looking into them. Anyone have any comments about Baldor.
Thanks

[RICHARD ZASTROW]

We used Baldor on most of all of the axis servo's and drives on machines we've built. Cutter spindle and work heads on our newer machines have hollow shaft frameless or cartridge motors.

We are starting to use direct coupling motors that "swallow" the end of the ball screw and have built in clamps that grip onto the end of the shaft. One caveat: these motors have no bearings and must rely on the shaft end to accurately locate the motors rotor relative to the stator.

Dick Z