Hello Guest it is July 14, 2020, 07:59:37 PM

Author Topic: New Mill  (Read 7520 times)

0 Members and 1 Guest are viewing this topic.

New Mill
« on: July 23, 2010, 07:40:59 PM »
I just received my new Syil X4 and have purchased a Mach3 license and am using the .XML file that Syil supplied.  Everything seems to be going smooth, except for one thing...

When I cut a pocket, it is under size by about .002 in/in and bosses are over size by about .001 in/in, (am assuming) the variation is probably the spindle (or) collet run out plus what ever backlash I have...

I have checked the spindle run out (about .0005) and have check the X Y movements, which shows about (on a average) .006 short in 4"...

Art, what area gets changed and how do I do it...

Kendall Miller
Fun Dimensional, Inc.
Leander, Texas

Offline RICH

*
  • *
  •  7,411 7,411
    • View Profile
Re: New Mill
« Reply #1 on: July 23, 2010, 08:29:52 PM »
Ken,
Make note of the original steps per unit. Now go to the settings page and calibrate each axis ( info in the manual ).
( the steps per inch, velocity and acceleration can be changed in CONFIG>MOTOR TUNING )
See if there is any difference and be meticulous in what you do. You may want to tweak the steps per unit.
I would suggest that you check it out some by first seeing what the steps per unit should be for say a max travel, then 1/2 the max travel,
and then see what it looks like at a range of motion you may normaly be working at.

Don't know what kind of balls screws are in your machine and the associated specs for them. Even lower spec'd ones will be in the +- .003" / ft
and they would be very repeatable if backlash was neglegable.

You need to have confidence that the steps are correct, machine is adjusted nicely, and then you can adjust machining  practice to achieve desired results.

RICH

Offline ger21

*
  • *
  •  6,282 6,282
    • View Profile
    • The CNC Woodworker
Re: New Mill
« Reply #2 on: July 23, 2010, 10:03:39 PM »
If your tool is .001 undersized you'll get similar results.
Gerry

2010 Screenset
http://www.thecncwoodworker.com/2010.html

JointCAM Dovetail and Box Joint software
http://www.g-forcecnc.com/jointcam.html

Offline Hood

*
  •  25,846 25,846
  • Carnoustie, Scotland
    • View Profile
Re: New Mill
« Reply #3 on: July 24, 2010, 05:33:46 AM »
R

Offline Hood

*
  •  25,846 25,846
  • Carnoustie, Scotland
    • View Profile
Re: New Mill
« Reply #4 on: July 24, 2010, 07:00:01 AM »
Oh sorry for my post above, was using the netbook to view the forum and dont know how the hell I managed to do that ;D

Hood
Re: New Mill
« Reply #5 on: July 24, 2010, 08:50:58 AM »
As soon as I saw the problem, I purposely chose a large stubby tool that would not flex and it measured exactly .375, I ran a program that would cut a 1.500 x 1.500 pocket, and one that would would cut leaving a 1.000 x 1.000 boss standing...

I am checking the pocket with gauge blocks and the boss with a 1" mic...

The machine has class P5 ball screws, since they brag about it, I assume that is good...

So it sound to me that you are saying that I need to decrease the number amount of steps in the [Config>Motor Tuning], correct?  Is there any other item that this will effect and need to be addressed in the [Config] when I do this?

Regards,
Ken

Offline RICH

*
  • *
  •  7,411 7,411
    • View Profile
Re: New Mill
« Reply #6 on: July 24, 2010, 10:23:29 AM »
Quote
The machine has class P5 ball screws, since they brag about it, I assume that is good
Yes they are good...P5 lead accuracy is +-23um  / 300mm

The tighter the tolerance on milling, the more practical machining items like, coolant / cutting fluid, type end mill, cut passes / direction, etc come into play.You will find that there can be a difference by just not using a cutting fluid for the same material or the condition of the end mill.
If the spindle run out is a half thou, and then the collet has some eccentricity, end mill sticks out some, then before you know it you could have a thou run out at the tool cutting edge. Actualy machining something is good, but, you don't change the steps per units based on it.

 I said note the original settings, as i will assume that their steps per inch are correct, and thus you want to confirm.
So take your time and reafirm table movement by using a dial indicator or whatever as long as the method is accurate.

Quote
decrease the number amount of steps in the [Config>Motor Tuning], correct
Yes....but why not use the calibrate axis available in Mach ( it is fast and very accurate ).
 
Quote
Is there any other item that this will effect and need to be addressed in the [Config] when I do this?
Quote
For the small step changes that you may make it will not have any affect on the accel or velocity

RICH

Re: New Mill
« Reply #7 on: July 24, 2010, 11:01:18 AM »
As I stated in my first Post (1st [Post] ever):
RE: with more detail

I checked the X movement out measuring 6" ((2) 1-2-'3' blocks against a stop, measuring leading edge to leading edge with an electronic indicator (same direction of travel)), Mach3 DRO = 5.994 (measured several times, always varied by about .001, this seemed to be the average)...

RE: set up
The 1-2-3 blocks were trammed straight with an indicator, secured, and checked with mic's to make sure the over all length was correct before the test measurements were taken.

Measured the Y movement over 4" travel ((2) 1-'2'-3 blocks), Mach3 DRO = 3.995 (average, using same set up as above))...

The test cuts were made in (2) materials, Delrin & Glass Filled PC, with the exact same results, the pockets were measured with gauge blocks, and the boss was measured with 1" mic.

What is my next step to correct this problem?

Offline Hood

*
  •  25,846 25,846
  • Carnoustie, Scotland
    • View Profile
Re: New Mill
« Reply #8 on: July 24, 2010, 11:15:15 AM »
Is your machine stepper or servo and what are your steps per unit?

Hood

Re: New Mill
« Reply #9 on: July 24, 2010, 11:24:23 AM »
I am a novice, but all I see in the doc's is "Stepper per rev = 200", so I assume that I have Stepper Motors...