Indexer Wizards version 2.0 Now Available

[gravirozo]

hello all


with a simple example, a grid with 1x1 inches sides.....
in "flat" mode already everyone can imagine it.... to turn, wrap or anyway you call, on a cilynder the router need same direction like in"flat" mode... the postprocessor in aspire can be changed, as after positioning, the postprocessor will write all Y to A or any letter you like...

also a solution, to make 2 initfile for mach, and the first will positioning, as it send signals to the Y, after positioning you can close the first, and start the second, which one configured as the Y axis pointing on the A, or forth with other words...

third solution can be after positioning , turn off controller and just simply plug the rotary motor to the Y controller unit///


the problem of this all, to generate the file.... a simple model, that generated by one conturcurve carry out no difficulties...
with other words objects that can be turn...

only a few critery need to be fulfilled... the largest point from the center need to be used for "full turn"...
all lower point be in "better" resolution...
the positioning can be just like in "normal" use with aspire...

this indexer a pretty application, looks working even without programs, making fluting or spirals, tasks can be described matematically...

so, it is only up to you you want to use this indexer, or just make your own way...

anyway, in october hopefully it will be a class... and everyone can give their experience...


thanks
viktor


actually, i made file already, hopefully james will help to show it...

[rgoldner]

so we just change the post proccessor in vectric to read a instead of y ?

jim

I assume you are talking about Greolt's method of wrapping. Unfortunately, it's not that simple. Each Y value has to also be converted into a number of degrees that will move the surface of the work the number of inches specified in the Y number. You must also know the diameter of the work to do this. It's not particularly hard to do, but I don't think it can be done in the post processor. This is exactly what the wrapper program does. And of course this method will not work with arcs.

[gravirozo]

hello


no, i don't want to count ondegrees... as on the beginning gaveexample, i want to use that method... to converting to angles, looks unnecessary... the Y movement, as steps, be given to the rotary motor... and the rotary axis will acting like Y axis...

your program probaly calculate some angles... i have no doubt works well... on my side i prefer solutions i can understand...


thanks
viktor

converting to angles makes too difficult all procedure... that's why the circles doesn't work with cordinatas converted to angles...

[rgoldner]

hello


no, i don't want to count ondegrees... as on the beginning gaveexample, i want to use that method... to converting to angles, looks unnecessary... the Y movement, as steps, be given to the rotary motor... and the rotary axis will acting like Y axis...
This is how my system works, Y generates the correct number of steps for the rotary axis motor. Angles are never calculated

your program probaly calculate some angles... i have no doubt works well... on my side i prefer solutions i can understand...


thanks
viktor

converting to angles makes too difficult all procedure... that's why the circles doesn't work with cordinatas converted to angles...

[Greolt]

I think you may be missing my point. Arcs or no arcs is not the problem.

Oh I was mostly talking about arcs because you mentioned that earlier as a concern.

I think another poster mentioned it as a concern also. Where as arcs are not much of an issue in rotary axis work. IMO of course.

When you throw velocity and acceleration settings and backlash compensation into the mix it starts to make things interesting.

If your macro is elegantly handling those things when pin swapping then I take my hat off to you.

I can't imagine how that stuff can be established if you are not even looking at the gcode.

Greg

[cabnet636]

this is the machine we are finnishing for the fourth axis, we are modifying the frame to run the lathe perpendicular to the gantry instead of parallel as it was originally designed then we will have more length on the lathe!! the table is removable and it has a 12" run on the z. i will do my best to have this ready in octoger and have been told by camaster it will be so!!

jim

[rgoldner]

I think you may be missing my point. Arcs or no arcs is not the problem.

Oh I was mostly talking about arcs because you mentioned that earlier as a concern.

I think another poster mentioned it as a concern also. Where as arcs are not much of an issue in rotary axis work. IMO of course.

When you throw velocity and acceleration settings and backlash compensation into the mix it starts to make things interesting.

If your macro is elegantly handling those things when pin swapping then I take my hat off to you.

I can't imagine how that stuff can be established if you are not even looking at the gcode.

Greg

Using the existing rotary axis steps/degree, velocity and acceleration and the column diameter I calculate the three settings for the swapped Y axis. As long as the existing settings are in tune my new settings work. I do not handle backlash as I do not think it is much of a problem on most rotary axes.

[cabnet636]

these machines are out there

[moto633]

Jim are you building yours in your shop??

I am going to locate an old lathe and convert it with a stepping motor and see what happens!!!!

Nick

[cabnet636]

the fellows at camaster build a 5-1 gear box for the lathe setup!! after that is is a drive and a motor, mach is setup for it, i'll get you some better pics

jim

[gravirozo]

hello...

for wood lathe two affordable option...



http://www.harborfreight.com/cpi/ctaf/D ... mber=34706


and a little larger
http://www.harborfreight.com/cpi/ctaf/D ... mber=98676


the first can be taken apart, and the head and tailstock can be on a 2x8 steel pair.... if need longer..... or just fixed on the router table if it fit under the gantry...

thanks
viktor

[Greolt]

Using the existing rotary axis steps/degree, velocity and acceleration and the column diameter I calculate the three settings for the swapped Y axis. As long as the existing settings are in tune my new settings work. I do not handle backlash as I do not think it is much of a problem on most rotary axes.

Fair enough about the backlash compensation. Probably not going to affect so many users. Funny thing is, of my three machines running Mach, the only axis that requires backlash compensation is my router's rotary axis.

As an example my Y axis rapids at 10,000mm/min and my A axis rapids at 3000degrees/min. If the rotary axis was pin swapped and sent Y pulses then it would stall every rapid move.

I know twiddling pin assignments is easy but I am surprised that you can adjust an axis velocity and acceleration settings via a macro.

For my own education I would love to learn how to do that. How you have done it, is quite rightly your business. I will have to ask the Mach code gurus about that one.


Greg

[rgoldner]

Using the existing rotary axis steps/degree, velocity and acceleration and the column diameter I calculate the three settings for the swapped Y axis. As long as the existing settings are in tune my new settings work. I do not handle backlash as I do not think it is much of a problem on most rotary axes.

Fair enough about the backlash compensation. Probably not going to affect so many users. Funny thing is, of my three machines running Mach, the only axis that requires backlash compensation is my router's rotary axis.

As an example my Y axis rapids at 10,000mm/min and my A axis rapids at 3000degrees/min. If the rotary axis was pin swapped and sent Y pulses then it would stall every rapid move.

I know twiddling pin assignments is easy but I am surprised that you can adjust an axis velocity and acceleration settings via a macro.

For my own education I would love to learn how to do that. How you have done it, is quite rightly your business. I will have to ask the Mach code gurus about that one.


Greg

Actually, it's no secret. In fact my macro is in plain text. Anyone who has the Indexer Wizards can open the macro file and see how it's done. Mach3 contains many, many internal values that can be changed by its VB scripting code. My macro is all script, not a gcode to be seen. Specifically, Mach3 has a GetParameter and a SetParameter VB command that can get/set these internal values. If you promise not to spread it around I'll send you the macro. Send me an email (rgoldner AT iname DOT com) requesting it. Of course, besides kinowing that it can be done you need the knowledge and technical ability to do it.

This is the whole problem in making my Indexer Wizards work with non-Mach3 controllers. The other controllers, to my knowledge, do not have the capability to get into their internals the way that Mach3 does.

BTW, 3000 degrees per minute is awfully slow (less than 9 RPM). Why can't you get it to go faster?

[Greolt]

BTW, 3000 degrees per minute is awfully slow (less than 9 RPM). Why can't you get it to go faster?

Good question.

I have made my rotary using a small industrial right angle gearbox. (hence the backlash) 40 to 1 reduction. This I am driving with a Nema 23 stepper which I had on hand.

None of the above is ideal and if I set out to build a rotary axis with some dollars it would be different.

However it cost me almost nothing and does a surprisingly good job. I am having a lot of fun exploring the capabilities.

I see more and more people interested in getting rotary axis or maybe getting started using one they already have.

Programs like your "Indexer Wizard", "Wrapper", "D2NC" and "Indexer Virtual Tool" all help with "hobby" type users being able to inexpensively get started in rotary stuff. Which is a good thing.

Thanks for your offer with the VB code. Now that I know that motor tuning parameters are settable I will do a bit of research off my own bat.

Do you have a mechanism that will reset pins if a catastrophe happens and gcode is not completed. (Tool break for eg.)

It has been pointed out to me that my example above was poor,

My Y axis rapids at 16666 steps per second and my A axis rapids at 11100 steps per second. That makes more sense.

Thanks for having the patience to discuss this stuff with me.

Greg

[rgoldner]

BTW, 3000 degrees per minute is awfully slow (less than 9 RPM). Why can't you get it to go faster?

Good question.

I have made my rotary using a small industrial right angle gearbox. (hence the backlash) 40 to 1 reduction. This I am driving with a Nema 23 stepper which I had on hand.

None of the above is ideal and if I set out to build a rotary axis with some dollars it would be different.

However it cost me almost nothing and does a surprisingly good job. I am having a lot of fun exploring the capabilities.

I see more and more people interested in getting rotary axis or maybe getting started using one they already have.

Programs like your "Indexer Wizard", "Wrapper", "D2NC" and "Indexer Virtual Tool" all help with "hobby" type users being able to inexpensively get started in rotary stuff. Which is a good thing.

Thanks for your offer with the VB code. Now that I know that motor tuning parameters are settable I will do a bit of research off my own bat.

Do you have a mechanism that will reset pins if a catastrophe happens and gcode is not completed. (Tool break for eg.)

It has been pointed out to me that my example above was poor,

My Y axis rapids at 16666 steps per second and my A axis rapids at 11100 steps per second. That makes more sense.

Thanks for having the patience to discuss this stuff with me.

Greg

Actually, Mach3 will reset the pins when it is started. However, the tuning numbers for Y must be reset. There are directions in my manual on how to do this.