Small CNC Machines For Making PC Boards

[tunedwolf]

The problem, of course, with micro-stepping, is that less torque is available from the motor, and this gets worse the more you divide up the step. For a stepper motor on any sort of machine that requires positional accuracy, this can be a bad thing. There are many factors that can easily overcome the holding ability of the motor which in turn will translate into missed steps, inconsistent positioning of the machine or just plain old stuttering and stalling. Even when the motor chosen is large enough to overcome the actual machine requirements, there's little accounting for the loads placed on it by the user. There are ways to help mitigate this effect, in the actual mechanical drive for the machine, like using ball or lead screws, rack and pinions for motion. For the most part, these motors are capable and do a reasonable job so long as the user doesn't place demands on it that are outwith the design parameters of the machine. Which is an art all in itself and really down to the operator's machining experience.

Personally, on all but the smallest of CNC machinery, I prefer using AC Servo's and drives, but that's a whole other can of worms

 

[MrAl]

Basically. I didn't care much about the logistics as you did, but the current is changed to HOLD the motor between full positions and it pretty much has to be in the gizmo that accepts the direction/~enable signals which usually interfaces to the motor. e.g. 64 pulses is a full step

Hi,

Yes. I was going on the general principle of modernization, where everything is moving to the computer for the main control and only the hardware that is truly needed be used externally. The idea is to move the controls into a more flexible environment where we have more control over the changeability of the total system that way.
This doesnt mean it has to be done that way but if it was i'd like to know if and when it happens. That would make my driver boards obsolete

An example would be a driver that interfaces with the computer and the computer software provides 2x microstepping, then with a software upgrade it provides 16x microstepping with no new hardware to buy. But again this doesnt have to happen, i just wanted to know if it did yet.

I was using old ink jet printer parts and an old line printer for the main table, with CD drive guts for the Z axis. It was fun to build but i should have continued with it before other things started coming up.

 

[MrAl]

The problem, of course, with micro-stepping, is that less torque is available from the motor, and this gets worse the more you divide up the step. For a stepper motor on any sort of machine that requires positional accuracy, this can be a bad thing. There are many factors that can easily overcome the holding ability of the motor which in turn will translate into missed steps, inconsistent positioning of the machine or just plain old stuttering and stalling. Even when the motor chosen is large enough to overcome the actual machine requirements, there's little accounting for the loads placed on it by the user. There are ways to help mitigate this effect, in the actual mechanical drive for the machine, like using ball or lead screws, rack and pinions for motion. For the most part, these motors are capable and do a reasonable job so long as the user doesn't place demands on it that are outwith the design parameters of the machine. Which is an art all in itself and really down to the operator's machining experience.

Personally, on all but the smallest of CNC machinery, I prefer using AC Servo's and drives, but that's a whole other can of worms

Hi,

I thought the torque/position curve had the minimum in the very center of a full step. That's where any teeth are farthest from the coils.
I have read that it is less though.

 

[ClydeCrashKop]

The way i see microstepping is like this...
The current to one winding goes lower, while the current to another winding goes higher, the two currents hold the rotor in a position in between the direct step positions, so a stepper with 100 steps could step 200 steps if the rotor is held between two full steps. If the stepper pulls less one way than the other, we get 1/4 step, then 1/8 step, and so on and so forth. If the controller can do linear, it could rotate the rotor smoothly just like a motor as it puts sinusoidal waves on the coils.

Yes. It is all done in the driver.

 

[jpanhalt]

ClydeCrashKop
Nice conversion. I have an almost unused Bridgeport with variable speed drive and got the stepper mounting kits from Mr. White in Nevada a few years ago, but I have not proceeded to convert it to CNC. I was impressed by your comment about dealing with 0.030" backlash was no problem. There is a lot of discussion about backlash on every CNC forum I have visited.

What are your thoughts on dealing with such backlash and/or worn lead screws? Does the software allow correction for both or just backlash? Do you monitor actual position or just count steps? (My machine has DRO on 3 axes.) Did you consider ballscrews, and why did you reject them?

John

 

[Tony Stewart]

A buddy of mine built one which he used to make simple 1-of boards. He also rebuilt an old Bridgeport II with new bridge drivers and AutoCad/Cam interface for large custom tooling. The only help I made was one day making the serial interface talk to the native OS command line interface.


The board mill is limited by the quality of the router bearings, so his trick was to use a dummy 1/2 hp spindle in series with very low ~ < 1 mil runout, between small motor and head.

Similarily his Bridgeport that he and Pal refurbished, has 1 mil features and 3 mil absolute accuracy.
You can get cheap USB to parallel ports or an old surplus PC with one. Your choice for board router as parallel ports have 50kB/s minimum BW.

The Bridgeport works well on a serial port with optical end stops for datum and safety and only simple commands to seek XYZ at any speed for example. CAM SW translates the AutoCad drawings into GERBER like protocol proprietary to Bridgeport. He shipped 2 old units worth over $500k new for something like,$15k and $10k in retrofits plus 6mos NRE, early 80's vintage dinosaurs but survived the ice age.

He has a wide range of talents in invention, design and any fabrication technology from vacuum moulding to 5 ton plastic composite fabrication to 10kW miniature water cooling of copper sponge electrolysis heatsinks for IGBT's and MOSFETs to Stainless Steel welding of Pulse Jet engines, RC vehicles etc.

See if you can spot the results here. https://umdcanada.com/microcontroller-ecu-interface/

Tony Stewart

 

[ClydeCrashKop]

Thanks John
It just lets you count steps. If using it toward the ends of the travel where the lead screw isn’t worn as much, it would not be accurate. If I start using it more, I might replace the lead screws and nuts but the rest of the machine is worn too much to make ballscrews practical.
It’s a good thing that you brought this up. I am meeting with a man to convert an old WEBB milling machine for a maker space.
He also wants me to setup the Mach3 backlash in his personal machine.

5.6.3 Configure Backlash
Mach3 will attempt to compensate for backlash in axis drive mechanisms by attempting to approach each required coordinate from the same direction. While this is useful in applications like drilling or boring, it cannot overcome problems with the machine in continuous cutting.
The Config>Backlash dialog allows you to give an estimate of the distance which the axis must back up by to ensure the backlash is taken up when the final "forward" movement is made. The speed at which this movement is to be made is also specified See figure 5.18

Note: (a) These settings are only used when backlash compensation is enabled by the checkbox.
(b) Backlash compensation is "last resort" when the mechanical design of your machine cannot be improved! Using it will generally disable the “constant velocity” features at “corners”.
(c) Mach3 is not able to fully honor the axis acceleration parameters when compensating for backlash so stepper systems will generally have to be detuned to avoid risk of lost steps.