Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: rcrabb on December 30, 2006, 11:29:02 PM

Title: Steps per inch (Uneven numbers)
Post by: rcrabb on December 30, 2006, 11:29:02 PM
how do I deal with an uneven number for my rotary axis. Ive calulated 17.7777777777777777777777777777777777777777777777777.... steps per unit? Please help
Title: Re: Steps per inch (Uneven numbers)
Post by: fer_mayrl on December 31, 2006, 02:48:52 AM
just put that number in the steps per inch dialog, put as many decimals as it will let you.
I think what mach will do is round as much as it can. It should work, although the number is quite small, so maybe resolution will not be that good.
Regards
Fernando
Title: Re: Steps per (uneven numbers)(fractional numbers)
Post by: PeterF on January 10, 2014, 07:13:31 AM
I too think about it. Our 4th axis is 40:1, which means 80'000 steps per 360 degrees, which means 222.222 steps per degree. How does Mach3, current version e.g. R3.043.62, handle the fractional part? Will it ensure that every 80'000 steps is 360 degrees, such that after many rounds it is still good. Assuming, I enter just 222.22, after 470 rounds I miss 376 steps. When I enter 222.22222, I miss 0.376 steps. When I enter 222.22222222, I miss 0.00376 steps.

I'm not CNC, but IT and Math, needing to set it up properly. What is good CNC respectively CAM practice, and what does matter?

(470 rounds is my sample piece, that's why.)
Title: Re: Steps per inch (Uneven numbers)
Post by: cncalex on January 10, 2014, 10:03:17 AM
When you travel in one direction the error accumulates as you calculated. When you travel back to the start point the error is 0 because Mach travels all the same way back.
Normally not a big problem as long as the diameter is not too big or the amount of rotaton is huge and you are in G90 mode. If you are in G91 ( incremental ) mode and
working with lots of sub recalls and do a short rot on 360 then the fraction of the last step will be lost ( not really lost - only within your operatin ) whitch can cause a greater
error at the end. The error is alway 0 in G90 mode at the home ( G53 ) position regardless what you did before. That all is what I found by testing no guarantee that is 100% correct   ;D
You can type in 222.2222222 ( 10 places )  When your diameter is 100mm then the circumference is 314 mm you said you miss 0.00376 steps after 470 rounds that is
314 / 360 = 0.87222 mm per degree / 222.2222222 steps = 0.00392 mm per step *  0.00376 steps = 0.000145 mm error on the circumference  after 470 rounds at 50mm radius.
Not too bad I think. Anyway, the best thing of course is to have the mechanical setup so that gives you a even number. If you doent want to change the mechanics there is one way
left to get it 100% precise, you can coose your circle  eg. 400 degree or what ever gives you even numbers, but short rot on G0 or 360 rollover will not work with this setup.

Alex
Title: Re: Steps per inch (Uneven numbers)
Post by: PeterF on January 10, 2014, 05:14:33 PM
Alex I do not construct the 4th axis. The gear 40:1 and the steps 2000 are made by the respective producers. Or what are the numerical precisions of the "steps per degree" value? I did not yet find the 10digits specification.

What is good CNC / CAM practice for fractional steps per degree, even as steps per revolution is integral. how is this handled?
Title: Re: Steps per inch (Uneven numbers)
Post by: cncalex on January 10, 2014, 05:58:58 PM
The 10digits is what Mach3 accept in motor tunig. You can type in more but only 10 will be accepted. The fractional part beyond the 10digits are lost. ( as far as I can see )
Code: [Select]
`how is this handled?`exactly as you calculated it before
Quote
after 470 rounds I miss 376 steps. When I enter 222.22222, I miss 0.376 steps. When I enter 222.22222222, I miss 0.00376 steps
You can set 222.2222222 steps per degree you lost  0.0376 steps after 470 rounds.
Quote
What is good CNC / CAM practice for fractional steps per degree, even as steps per revolution is integral
That depends on what you want to do. If your 4th axis is only traveling a few rounds forward and backwars in G90 mode the whole day that should not be a problem.
But when you travel thousends of rounds in one direction or have lot of subcalls in G91mode that may be a problem. Depending on how precise your work should be.

Alex
Title: Re: Steps per inch (Uneven numbers)
Post by: PeterF on January 10, 2014, 07:16:52 PM
I want to set it up so that it works precisely, without "depending on". I set the machine and then it is on its own. I do not machine (i'm the IT, not the CNC). That's why I care, and because the design is 80000 steps per revolution what cannot be entered precisely.

In case you know more from what is practice in the CNC world, any information is useful. Is there a reference to the 10-meaningful-digits-rule, if it is somewhere in the machwiki, or otherwise, it will be useful, of course.
Title: Re: Steps per inch (Uneven numbers)
Post by: cncalex on January 11, 2014, 02:43:06 PM
I understand where you are heading but it is as it is. You need a even Steps per unit to get it precisely. Everything else will be more or less precisely always  "depending on" what you are doing.

Quote
Is there a reference to the 10-meaningful-digits-rule, if it is somewhere in the machwiki, or otherwise

No, I think its not.

Alex