Advanced Search

And we're back to your X limit values in INI file being wacky.

See the error about "Line 17" in your screencap?

Line 17 in the pawn file is the first move with an X axis value.  Specifically:N20 G0 X55.56 Z10.And if you look at your INI file:MIN_LIMIT = -48.0
MAX_LIMIT = 3.00
Something isn't going to work out when you tell LCNC to move to an X position greater than the MAX_LIMIT unless you have a tool offset active that drives the commanded position back inside the MAX/MIN limits.

Problem #1 is that your MAX should be something like 51, and MIN should be something like 0.  Or maybe closer to 40 and -14 if you want the turret centerline tools to be at machine "0" after homing.  I've no idea what the actual stroke length is for X.

Problem #2 is that you probably have F360 programmed with a HUGE clearance diameter/radius.  F360 has no idea about your machine's travel limits (unless you define them in the Machines Definitions), so it doesn't alert you when you tell it you want a gigantic clearance.  Try setting your clearance to 1 or 2mm from the stock OD and re-post.  Line #17 will probably look more like:

N20 G0 X35.56 Z10.

So circle back to my earlier post about adjusting your X MAX/MIN values, and once you've got them where you want, re-do all three F360 operations to have a minimum clearance value and re-post.  You can then look at the g-code and immediately see whether F360 is trying to drive your X-axis beyond the max limit.

Uthayne is your man. He is very knowledgeable in this area.

No, there are too many dependencies that cannot be met. The move from python 2 to python3 is the biggest one.

You would be better off upgrading to Debian 12

JT wrote:

The ini entry INPUT is now THEME gnipsel.com/linuxcnc/flexgui/ini.html#themes

Which file browser, the keyboard or the touch screen?

Good point on the flood and mist and I see no reason to disable them at any time and that has been fixed.

JT
 

THEME = touch overrides my qss. I get the touch file browser back be it's using whatever qss the theme uses instead of the one I specified with QSS =. Can I add the touch style file browser to my qss?

As for the file browser showing the current directory, just some bar at the top of the browser or maybe even a label that can display it above the file browser. Right now you see files in the directory that you are in....but you have no idea what directory you are in. I keep most of my NC files on a NAS so sometimes I move around a bit to get to the file. It's alot easier to navigate if you can see what directory you are currently browsing.

BTW, thanks for answering these questions. I can probably figure some of this out myself but that probably wouldn't help anyone else looking to work with Flex.

No way lol, it's amazing what you see when there's a finger pointing right at it. That was stupid on me it's so obvious.
What was throwing me off was when ever I touched off the x and z . The program wouldnt run because it was exceeding the limits. Even though there's plenty of space there. So there has to be an issue with my limits etc ?

I give up!

  I

 

hmmm

LOL i think i need one trying to figure this out.
hal , test file and pawn attached
When i touch off g54 on X it drops the model down in the preview. Its thinking Im cutting from the top down and also the Z is inverted on the model. no matter what I do in fusion changing flipping axis or flipping the chuck placing its still the same.

O.M.G.

I just watched your video.  @Aciera was right, as usual.

What you are pointing at is the tool path, not the pawn.  I've circled the 'pawn' in red.  What you are pointing at and claiming is the 'base' is, in fact the pawn's head after the cuts are made.

White = tool path (are tool will cut)
Black = pawn after material is removed

This still doesn't explain why positive SEARCH_VEL moves left during homing but we can circle back to that.

 

One other thing about a supported rotary set up is if you have a stock width that will fit the bore of your set up you can reduce you cutting distance compared to cutting from plate. when you look at things like duty cycle that's a big plus when your cutting @ 60% duty. On top you don't have to wear your slats down either, so you reduce your operating cost. so, it makes sense on small parts to keep this in mind.

when you get to a point where you are processing a lot of sheets the skeletons of the sheet add up to quite a bit of wasted material. this is just something you notice if you are around it all the time. i know time is money and that's why it happens but looking for savings can't be forgotten about at the same time.

Hi.

One question:

The actual master , can compile on Debian 10 ?

I'd be willing to test this out and provide some data. I currently use the BCL-AMP setup with QTPlasmac and while it works, it could definitely be faster in its tracking. The sensor outputs the nonlinear signal which is also noisy. After filtering (hardware and software), it works, however I feel as though it is not sensitive enough now due to being averaged and filtered so much that any abrupt changes in height will cause the nozzle to touch.

A cleaner, linear response signal could be the ticket, as I have fiddled with the PID parameters quite a bit as well as my Z axis velocities and accelerations and it is still not really where I'd like it to be.

Let me know how I can be of help.

Thank you for the tip.  I'll have a look this evening at index-enable.

If it makes any difference, I'm not using orient.comp for the main spindle... although I am using orient for the sub-spindle.

Main spindle is using caxis.comp but there aren't any references to index-enable in it other than a "//detect index-reset" statement.

If it is an index-enable issue, I wonder what would cause it to stay high?

Relevant HAL section for what it's worth:
# INDEX-ENABLE
net SPIN0-INDEX-ENABLE      <=> hm2_[MESA](BOARD).0.encoder.04.index-enable
net SPIN0-INDEX-ENABLE      <=> spindle.0.index-enable
net SPIN0-INDEX-ENABLE      <=> joint.2.index-enable
net SPIN0-INDEX-ENABLE      <=> pid.s0.index-enable
net SPIN0-INDEX-ENABLE      <=> pid.c.index-enable
 

Can you post the g-code file as well as your HAL?

I'm not quite ready to suggest you call in a priest.

You could sure make all of this SO much easier by simply providing the basic information you have been asked for.