round holes not exactly round, slight oval shape
10 Aug 2016 17:34 #78581
by bschiett
round holes not exactly round, slight oval shape was created by bschiett
Hi all
I have a self built CNC gantry style router I built. It's a pretty heavy machine (about 160kg) with a t-slot table about 1 meter x 0.5 meter in size. The ball screw spindles are driven via belts by stepper motors. I'm running linuxcnc to control the machine (v2.4 if i remember correctly).
All is well but i've noticed that i'm getting slightly oval shaped holes when cutting a hole that is supposed to be perfectly round. It's possible to see it with the naked eye if you look carefully but it is not immediately obvious when you first look at it the array of holes I've cut.
The hole itself is 20mm diameter and i'm cutting it as a pocket with a 10mm slingle flute endmill. Material if acetal, held down by tape.
I'm helixing down into the hole.
My machine acceleration is set at about 300mm/s2. The feed rate i'm using is about 2500 mm/s.
I'm wondering what could be the source of this problem? Is it that i should cut the holer slower (i.e. lower feedrate), or is this due the acceleration being set too low, or is it because there is a problem with the steppers, ballscrews or belts?
Any help greatly appreciated!
thanks
B
I have a self built CNC gantry style router I built. It's a pretty heavy machine (about 160kg) with a t-slot table about 1 meter x 0.5 meter in size. The ball screw spindles are driven via belts by stepper motors. I'm running linuxcnc to control the machine (v2.4 if i remember correctly).
All is well but i've noticed that i'm getting slightly oval shaped holes when cutting a hole that is supposed to be perfectly round. It's possible to see it with the naked eye if you look carefully but it is not immediately obvious when you first look at it the array of holes I've cut.
The hole itself is 20mm diameter and i'm cutting it as a pocket with a 10mm slingle flute endmill. Material if acetal, held down by tape.
I'm helixing down into the hole.
My machine acceleration is set at about 300mm/s2. The feed rate i'm using is about 2500 mm/s.
I'm wondering what could be the source of this problem? Is it that i should cut the holer slower (i.e. lower feedrate), or is this due the acceleration being set too low, or is it because there is a problem with the steppers, ballscrews or belts?
Any help greatly appreciated!
thanks
B
Please Log in or Create an account to join the conversation.
- Todd Zuercher
- Offline
- Platinum Member
Less
More
- Posts: 5008
- Thank you received: 1441
10 Aug 2016 18:54 #78585
by Todd Zuercher
Replied by Todd Zuercher on topic round holes not exactly round, slight oval shape
For a hole that size with a tool that size, there are a long list of possible causes.
The first think I'd check is backlash.
Slop in one of the ways (especially the Z axis), Be sure to check this for each axis in all directions and for rotation.
Axis out of square with each other.
Scale wrong on one axis
Lost steps
Flex in the machine
Part moving (Highly probable if only taped down)
Many of the possible cause have a little different look, but on a small circle they can all sort of look the same. A larger circle might show better where the possible source of error is coming from. A backlash problem may not show up much on a larger circle (may look more like a bump or offset), but the amount of error stays the same no mater the circle size so becomes more significant the smaller the shape is. Lost steps at reversals will be similarly more significant the smaller you go, and may disappear on larger shapes because of lower acceleration rates on larger arcs.
The first think I'd check is backlash.
Slop in one of the ways (especially the Z axis), Be sure to check this for each axis in all directions and for rotation.
Axis out of square with each other.
Scale wrong on one axis
Lost steps
Flex in the machine
Part moving (Highly probable if only taped down)
Many of the possible cause have a little different look, but on a small circle they can all sort of look the same. A larger circle might show better where the possible source of error is coming from. A backlash problem may not show up much on a larger circle (may look more like a bump or offset), but the amount of error stays the same no mater the circle size so becomes more significant the smaller the shape is. Lost steps at reversals will be similarly more significant the smaller you go, and may disappear on larger shapes because of lower acceleration rates on larger arcs.
Please Log in or Create an account to join the conversation.
Time to create page: 0.062 seconds