Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: nikolatesla on March 21, 2012, 09:15:59 PM

Title: A axis trouble calculating help!.
Post by: nikolatesla on March 21, 2012, 09:15:59 PM
Hi guys,

I have my chinese machine working fairly well untill  I installed the A axis motor

gearbox is 50:1  motor is 200 steps per turn ..1.8 DEG  microsep is 8 and I want the a axis to be angular.

I am trying to set a axis steps per  to the right value

I have set a axis angular .In tool path a-rotations enabled

I did (200 x eight) x 50 /360  

I tried running these numbers with no success . 222.222222 (the result) is nowhere near right in practice
I then tried the trial and error and found that 229 is almost right but has a small error.

x,y,z are all spot on ..well .01 mm.. ok for wood

I am surely doing something wrong or not clicking something I should .
Its driving me nuts knowing that my machine is not really "notational l control "

Lindsay

Title: Re: A axis trouble calculating help!.
Post by: RICH on March 21, 2012, 10:45:10 PM
Lindsay,
The caluculated steps per degree for A is correct. Make sure you remove any backlash before rotating the A when checking. You may need to refine if the gear reducer is not exactly 50:1 reduction. For an indicator to check you just measure the table, say 6" dia,
then 3.14(6)=18.85" circ x 1/360=.052" for 1 deg       .026" for 30'     .013" for 15'   etc.
So with a small pencil line away from zero mark you can refine quickly. Also do more than one revolution as number  of steps your off
will multiply.

RICH
Title: Re: A axis trouble calculating help!.
Post by: nikolatesla on March 22, 2012, 03:29:50 AM
Thanks Rich,

I have calibtared using a spirit level on the top of the chuck it is very  accutare to .2 deg.
Just rechecked my 50:1 again ,all good
motors checked again ..all good

No backlash setting applied and all tests are  in one direction

The only way was to use the calibrate axis option ,sending it for a 1440 angle then  measurung the error applying it then doing eight turns ..I wound up with 228.67456354 or something like that. 

It is wonderful that I was able to achieve useful accuracy , but frustrating that the numbers failed to do the angular calc as I expected .

I must be doing or setting somethnig wrong in mach.  I am new at this but Ive spent considerable time checking myself .

Im thinking microstepping fault on controller?// More likely noob has not clicked something

Lindsay
Title: Re: A axis trouble calculating help!.
Post by: nikolatesla on March 27, 2012, 08:22:41 AM
I best answer/conclude  my own post here ,for others who's numbers dont add up

The controller board was faulty swapped with another axis and all ok ...
These chinese machine have many issues ..wires break easily(replace them all !) , vfd noise, lose sliderail bearings, swarf everywhere.

What you are buying is a bunch of parts of differing quality thrown together as quickily as possible.Never tested as bought .But ok for hobby use The machine itself is going your hobby for a good while .
Title: Re: A axis trouble calculating help!.
Post by: GytarrMann19 on April 04, 2012, 08:35:31 PM
Not trying to hijack your thread...but quick question: How does Mach handle fractional steps per unit? I.E. 200.52 steps per degree?

Kevin
Title: Re: A axis trouble calculating help!.
Post by: Hood on April 05, 2012, 03:41:25 AM
Obviously Mach can only send out a full pulse so if you command a move that will include a decimal of a pulse then Mach will move to the nearest. A track of this missing decimal is kept and worked out in the next move.
Hood