Questions about milling machines with linear scales

More
2 years 4 weeks ago #311330 by cprhodin
I'm having a hard time making round circles with my milling machine.  When cutting an OD path 1.25" in diameter I'm seeing 0.006" diameter variation.  If you have a machine with linear scales I have a couple of questions:

1) What kind of accuracy are you getting when contouring a circle?

2) What does the PID error look like in the scope?


Chris
 

Please Log in or Create an account to join the conversation.

More
2 years 3 weeks ago #311557 by tommylight
What conroller, drives, etc?
If analog drives, tuning might be loose, so adding more P term could help.
Same goes for steppers with scales.

Please Log in or Create an account to join the conversation.

More
2 years 3 weeks ago #311567 by Todd Zuercher
Also if the mechanics of the machine have backlash, you will struggle with imperfections at the points of direction reversals, even with linear scales. And that backlash will likely cause instability and preventing you from tuning the PID loop stiff enough to get good results.

Please Log in or Create an account to join the conversation.

More
2 years 3 weeks ago #311622 by cprhodin
I have AC servos that accept stepper input at a configured resolution of 100,000 steps per inch.

When I use the HAL scope to trace the X and Y error signals I see a following error of 1 to 2 tenths except during the direction changes where I see 5 tenths on Y and 1 thou on X. I don't think these numbers account for the 6 thou error I see. I'm left with two possible causes; backlash in the linear scales and slop in the axis ways. I've adjusted the gibs multiple times trying to resolve this.

I'd like to know what other people get when they cut an OD so I can know what reasonable results look like.

Please Log in or Create an account to join the conversation.

More
2 years 3 weeks ago #311650 by Todd Zuercher
Slop in the ways or gibs can easily cause very large imperfections at direction reversals, due to the table rocking on the slop. The biggest problem is when the ways are worn unevenly, and they are too tight at the ends, but to loose where most worn.

Please Log in or Create an account to join the conversation.

More
2 years 3 weeks ago #311656 by OT-CNC
Get a good indicator and check for mechanical backlash on the machine, .0001" per division.
There should be no backlash on the scales. Not sure sure what you mean. Reader should follow the machine. 
Is this a ball screw driven machine? AC bearings on the screw? Quill on the Z? Everything that moves has some play and will add up especially under load. I usually write down all play values for each axis/components and have that as a reference for wear down the road....

Please Log in or Create an account to join the conversation.

Time to create page: 0.057 seconds
Powered by Kunena Forum