Calculating steps per inch!

[Pplug]

When I first set up my converted CMM router I had some weird screws with an unknown pitch. I had to use Mach3's built in calculator to find the steps per inch value. That worked fine for the work I was doing on it but now I have the chance to make the numbers absolute.

I have replaced the unknown lead-screws with 10mm pitched ballscrews. I have also upgraded the motors and controller to a g540 and 200 step per inch motors. My big question is..... Mathematically, how many steps per inch do I need?

10mm = 0.39370079 inches
200 steps per inch motor @ 10 microsteps = 2000 steps per rotation?

2000steps = 0.39370079 inches
5080steps = 1 inch

Is my math correct?

[ger21]

yep.

[JD_Mortal]

I am curious how you are getting 10 micro-steps?

You would need 5-coil stages to get that from the controllers I know of.

Normally, there are 4-coil stages, and thus four wires, two paired sets (A,B), yielding the 4-coil stages. (+ -, + -) (- +, + -) (- +, - +) (+ -, - +) repeat.

That is single-step operation or 1x.
Then there is half-step, which uses 1/2 current between each phase-switch.
1/4, 1/8, 1/16... etc.

I wasn't aware there was a 1/10th step controller. (Would that be the "metric" of controllers? Sounds interesting.)

Does it let you select individual steps? 10/10, 9/10, 8/10... etc...

[ger21]

All Geckos use 10 microsteps. Most are fixed, but they do have models with step multipliers if required.

[cornbinder23]

I believe the g540 is a 10 microstep driver

G540 4-Axis Motor Control

JTCUSTOMS

[hemsworthlad]

In my experience there are more that do 10x times than dont..

Pplug: I would still use the set steps per feature in Mach, If used with an accurate measureing system it will be more accurate than just setting the steps per @2000 because it helps account for any inaccurecy's in the screws/motors etc and counts the actual steps taken to move a set distance.

[Tmau]

Hello.

I have a 4 axis cnc mill. I have a problem and manufacturer said that it is problem with steps per but I am not shure if that is the case.

When I mill ring( I mill jewelery) The tools starts at the botom of it and than goes around and finishes where it started. but not exactly where it startdds it is a mistake that is big for jewelery industry. about 0.25mm. the toolpath lenght is abou 4 meters. that happens mostly in y axis direction (longest toolpath is in y) and a little bit in z axis direction.

Should I change steps per or is there any other problem.

I would be glad if anyone helped.

Hope I write it under right category.

Thank you, Matic

[hemsworthlad]

Originally Posted by Tmau Hello. I have a 4 axis cnc mill. I have a problem and manufacturer said that it is problem with steps per but I am not shure if that is the case. When I mill ring( I mill jewelery) The tools starts at the botom of it and than goes around and finishes where it started. but not exactly where it startdds it is a mistake that is big for jewelery industry. about 0.25mm. the toolpath lenght is abou 4 meters. that happens mostly in y axis direction (longest toolpath is in y) and a little bit in z axis direction. Should I change steps per or is there any other problem. I would be glad if anyone helped. Hope I write it under right category. Thank you, Matic

Hello Matic,

I would be pleased to help but really you need to create your own thread asking this question.! It will take meny posts to help resolve this issue and it's not fair to Pplug for us to hi jack his thread.

Make a thread then we'll look at it for you.

[awerby]

Most rotary tables have some. You run into it when you change directions. Try running the rotary axis a revolution or two in the direction it turns the part, then zero it without changing direction.

Andrew Werby
ComputerSculpture.com — Home Page for Discount Hardware & Software

[Pplug]

It all came together last night! I put in 5080 into the steps per inch and the darn thing works! I used my digital calipers to measure the travel and it is spot on! I'm sure I could get more accurate with a dial but for a quick spot check it worked!

Thank you all for validating my dusty 15 year old high school math skillz!