Strange offset during milling
- Twizzlermill
- Offline
- Senior Member
Less
More
- Posts: 40
- Thank you received: 5
03 Oct 2021 00:49 - 03 Oct 2021 00:52 #222067
by Twizzlermill
Strange offset during milling was created by Twizzlermill
Hello all,
I finally got my router in a working state – or rather almost working, since apparently, there's still some kinks to deal with. Today, I milled a test piece; MDF, so nothing particularly taxing. However, during the facing operation, I had a weird offset in the piece I can't quite explain:
(I don't know how to post images here, so here's an imgur link: imgur.com/CcamjLW)
It's a zig-zag facing operation at an 45° angle. As far as i can remember, for the pass in question, the endmill started at the top right and moved down, so it started at the wrong height. It's a self-built cnc router with cloded loop steppers. All cables are shielded with the shielding connected to ground.
What's weird for me about this is that apparently, the machine "corrected" itself at some point. If it was a physical issue (i.e. something pushing up the z-axis), I would expect that every subsequent pass had the "wrong" height, but as you can see from the picture, the z-axis went down to the correct height after a while and stayed there for the rest of the operation. Apart from this pass, the facing looks great.
It doesn't seem like I can reliably reproduce the issue.
Does anyone have an idea what might've caused this?
TIA
I finally got my router in a working state – or rather almost working, since apparently, there's still some kinks to deal with. Today, I milled a test piece; MDF, so nothing particularly taxing. However, during the facing operation, I had a weird offset in the piece I can't quite explain:
(I don't know how to post images here, so here's an imgur link: imgur.com/CcamjLW)
It's a zig-zag facing operation at an 45° angle. As far as i can remember, for the pass in question, the endmill started at the top right and moved down, so it started at the wrong height. It's a self-built cnc router with cloded loop steppers. All cables are shielded with the shielding connected to ground.
What's weird for me about this is that apparently, the machine "corrected" itself at some point. If it was a physical issue (i.e. something pushing up the z-axis), I would expect that every subsequent pass had the "wrong" height, but as you can see from the picture, the z-axis went down to the correct height after a while and stayed there for the rest of the operation. Apart from this pass, the facing looks great.
It doesn't seem like I can reliably reproduce the issue.
Does anyone have an idea what might've caused this?
TIA
Last edit: 03 Oct 2021 00:52 by Twizzlermill. Reason: Clarification
Please Log in or Create an account to join the conversation.
- tommylight
- Away
- Moderator
Less
More
- Posts: 18725
- Thank you received: 6295
04 Oct 2021 22:06 #222226
by tommylight
Replied by tommylight on topic Strange offset during milling
Had a look twice, can not figure out what is going on from that picture, so more pictures ..
To upload here:
-click on "attachments" just below the editor text entry
-click on "add files"
-choose the file and double click or choose and click "open"
To upload here:
-click on "attachments" just below the editor text entry
-click on "add files"
-choose the file and double click or choose and click "open"
The following user(s) said Thank You: Twizzlermill
Please Log in or Create an account to join the conversation.
- Todd Zuercher
- Offline
- Platinum Member
Less
More
- Posts: 5034
- Thank you received: 1429
12 Oct 2021 13:23 #222935
by Todd Zuercher
Replied by Todd Zuercher on topic Strange offset during milling
Looks like a mechanical issue. The closed loop steppers did their job, detecting the error and correcting it. But that a mount of error for that amount of time probably should have caused them to set a following error alarm or fault. You should probably look into adjusting your settings for that. As to your mechanical issue that caused the Z-axis to jam or stick you will have to look into that yourself. (We can't see and feel your machine and it's design.)
The following user(s) said Thank You: Twizzlermill
Please Log in or Create an account to join the conversation.
- Twizzlermill
- Offline
- Senior Member
Less
More
- Posts: 40
- Thank you received: 5
13 Oct 2021 21:13 - 13 Oct 2021 21:14 #223040
by Twizzlermill
Replied by Twizzlermill on topic Strange offset during milling
Hey, I totally missed your answers til now, sorry.
Yes, it may be a mechanical issue. Maybe the dust shoe put too much pressure or got caought in an edge (I'm still playing around with the design on that part, that iteration was just too thick and the bristles too stiff).
Interesting observation regarding the stepper settings. I'm not sure where exactly to find those settings, but I'll take a look at it. I'm assuming the correction is happening within the drivers themselves; I haven't set up a position output from the drivers to the card. So can I even change the settings to this in LinuxCNC?
@tommylight:
The piece doesn't exist anymore, at least not in that stage, so there's no more photos of it.
The machine got pushed up around 0.1-0.2mm at the start of one pass, but corrected itself back to normal.
But hey, good to know that the Closed loop steppers did what they're supposed to!
Thanks!
Yes, it may be a mechanical issue. Maybe the dust shoe put too much pressure or got caought in an edge (I'm still playing around with the design on that part, that iteration was just too thick and the bristles too stiff).
Interesting observation regarding the stepper settings. I'm not sure where exactly to find those settings, but I'll take a look at it. I'm assuming the correction is happening within the drivers themselves; I haven't set up a position output from the drivers to the card. So can I even change the settings to this in LinuxCNC?
@tommylight:
The piece doesn't exist anymore, at least not in that stage, so there's no more photos of it.
The machine got pushed up around 0.1-0.2mm at the start of one pass, but corrected itself back to normal.
But hey, good to know that the Closed loop steppers did what they're supposed to!
Thanks!
Last edit: 13 Oct 2021 21:14 by Twizzlermill.
The following user(s) said Thank You: tommylight
Please Log in or Create an account to join the conversation.
- Todd Zuercher
- Offline
- Platinum Member
Less
More
- Posts: 5034
- Thank you received: 1429
14 Oct 2021 12:12 - 18 Oct 2021 14:15 #223079
by Todd Zuercher
Replied by Todd Zuercher on topic Strange offset during milling
Those settings would be in the drivers not Linuxcnc.
If you did set up position feedback to Linuxcnc, then Linuxcnc could set a following error alarm. But without the position feedback, you would still need an alarm feedback from the drive to let Linuxcnc know that the drive has detected a problem for it to do any good.
If you did set up position feedback to Linuxcnc, then Linuxcnc could set a following error alarm. But without the position feedback, you would still need an alarm feedback from the drive to let Linuxcnc know that the drive has detected a problem for it to do any good.
Last edit: 18 Oct 2021 14:15 by Todd Zuercher.
The following user(s) said Thank You: Twizzlermill
Please Log in or Create an account to join the conversation.
- Twizzlermill
- Offline
- Senior Member
Less
More
- Posts: 40
- Thank you received: 5
14 Oct 2021 20:25 #223114
by Twizzlermill
Replied by Twizzlermill on topic Strange offset during milling
Yeah, I figured… Hum, I had hoped I didn't have to mess with calibrating the drivers directly. Well, I guess I'll just let the machine work for a while and see what happens.
I did connect the alarm outüputs to LinuxCNC, but I really want to try it out to see whether it does what it's supposed to. Not sure how to trigger the alarm without risking damage though.
I did connect the alarm outüputs to LinuxCNC, but I really want to try it out to see whether it does what it's supposed to. Not sure how to trigger the alarm without risking damage though.
Please Log in or Create an account to join the conversation.
Moderators: piasdom
Time to create page: 0.125 seconds