Z axis zero changes in middle of job

Questions/answers/discussion about initial setup of your CNC Shark

Moderators: al wolford, sbk, Bob, Kayvon

sudo
Posts: 23
Joined: Wed Jan 11, 2017 3:06 pm

Z axis zero changes in middle of job

Post by sudo »

I recently received the Shark HD4. Everything is on the most recent firmware versions. Ive been using the laser engraver thus far and today is the day I fire up the router.

I have a simple job that im trying to vcarve.

I zero out x,y,and z and run the job. It cuts about 5%, then I notice that the vcarve bit is about .25 - .375 above the work piece. Its plunging and running the code, but no longer is making contact with the work piece. I e-stop the job, and use the pendant to send it back to x,y,z zero. X, and y are fine, but the z axis has raised about .1 inch. All axis read zero.

Since this is on the CNC itself, I doubt its something in the gcode vcarve produced.

Anyone seen this before?

Im also having scale issues where its cutting/engraving smaller that it should be, but Ill tackle that one after getting this resolved.

User avatar
Kayvon
Posts: 552
Joined: Tue Oct 21, 2014 11:46 pm

Re: Z axis zero changes in middle of job

Post by Kayvon »

I've had this occur in two situations:

1) I didn't secure the bit in the router's collet tightly. This can go either way--sometimes it ends up too high, not making contact with the piece, and sometimes the rotation pulls the but out farther, destroying the work piece rather than traveling above it.

2) The router itself isn't secured tightly and lifts up. This happens only when I've reassembled the machine after a move and I only mention it because it's your first time using the router aspect of the machine.

It's also possible for the Z alignment to mismatch as a result of bottoming out against the table/workpiece or topping out against the max height of the gantry.

SteveM
Posts: 168
Joined: Thu Sep 08, 2016 1:29 pm
Location: Franklin, Wisconsin

Re: Z axis zero changes in middle of job

Post by SteveM »

I also agree that the bit may not be tight in the collet.
Once the bit started cutting, it slipped into the collet and you just didn't see it move.
Do you use a zero touch plate or just a piece of paper to zero out the z axis?

sudo
Posts: 23
Joined: Wed Jan 11, 2017 3:06 pm

Re: Z axis zero changes in middle of job

Post by sudo »

Thanks for the responses. The bit is tight. I ran the job twice to reproduce the issue and it steps up at the same time on all tests.

The bit is not slipping.

Its like to gets to a certain point then just decides to move up. Can something in the gcode reset the z zero?

To reiterate the issue:

* I clamp down the work piece. In this case a .5" piece of birch ply.
* Set xyz zero in the appropriate location (bottom left) manually. I do not have a z zero touch plate (yet)
* Run the program. This is a vcarving using a .375 90 degree vcarve bit.
* The program runs for about 8 minutes then steps up.
* I stop the program and send the router back to xyz zero.
* x and y are in the appropriate location, but z has raised up about .1 of an inch.

I think its worth mentioning a few things:

* when the bit 'steps up', it is cutting a part of the same 'model', but it is moving to a new vector. All of these vectors, however, have the exact same cut settings as they are all on one cut pattern.
*Up to this point, I have been doing laser engravings. To do this, you move the laser so that its touching the work piece, reset xyz zero, then move z up .1 inch and reset xyz zero again. Its as if this is happening in the middle of the v carving.

Thanks in advance for all your help.

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Z axis zero changes in middle of job

Post by Rando »

Thanks for the detailed description of your process.

Is that material attached directly on the bed? Is the gantry mounted on the upper set of bolt holes on the side-plates?

I ask these questions, because the behavior you're citing is as if it was "grinding its gears" at the BOTTOM of the Z-axis carriage. Which tends to mean that when it's making some deep cut, the z-axis carriage is hitting the bottom of its range, and "losing steps".

When the board is not in the machine, can you insert that bit and SAFELY jog all the way down to where the bit is just touching the bed? My guess is that you'll stop a ways before that, because you'll notice it about to hit at the bottom of that z-axis carriage. If that's what's going on, raising the stock on a bed of MDF or similar will likely fix it. Heck, theoretically you'd only need to move it up by that odd amount it seems to be missing :D.

Another thing I'd look at is using a "back plotter", which shows the path of the cutting tool, without attempting to simulate the resulting material. The one I use is the CNCCookbook GWizard Editor, which has a free-trial, and is quite good at doing sanity checks, stepping through, and detailed examinations of GCode. If you get a trial, load the file and click the "Right" view. You can then see just how deep the cutting is trying to go, at it's deepest point(s), to see if that Z value can actually be reached with that bit (by jogging), without losing those precious steps. If you don't want to, feel free to send me the gcode (tap) file via email (thom@tctrandolph.com), and I'll take a look at it.

The likelihood is VERY low, but it is possible that an error crept into the GCode output by Vectric, if there was some latent error in the post-processor file, and this toolpath for some reason hit it where others did not. Careful examination of the code around the area where you're seeing the jump might reveal some truly odd thing happening. After all, this is all software...and we all know how incredibly reliable software is ;-).

Though I **hope** it will, mostly I just **wonder** whether this will help :D.

Regards,

Thom
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

sudo
Posts: 23
Joined: Wed Jan 11, 2017 3:06 pm

Re: Z axis zero changes in middle of job

Post by sudo »

Thanks Thom.

I have yet to attach a spoiler board and have the work piece clamped down to the table directly. I checked the z axis to see if it was bottoming out, and indeed it was. Good call!

And of course, it stepped up right after the deepest cut was made. Im going to get the spoiler board attached today and give it another go.

Ill be sure to update this thread if the issue is not resolved.

Thanks again to everyone for opining.

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Z axis zero changes in middle of job

Post by Rando »

Oh, that's AWESOME! I too have watched in horror as a supposedly normal cut sent an expensive endmill plunging through the work. In my case, it was through a 2" thick block of aluminum, when the machine decided that 5mm safe height was really 5 inches...when it only had like 7mm of space at the top of Z. Luckily the material was up on machinist parallels so the vise didn't get injured, but that freaked me out for a couple days, literally afraid to start a cut until I could trace EXACTLY what caused it.

Cheers!

Thom
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Z axis zero changes in middle of job

Post by Rando »

sudo wrote:Thanks Thom.

I have yet to attach a spoiler board and have the work piece clamped down to the table directly. I checked the z axis to see if it was bottoming out, and indeed it was. Good call!

And of course, it stepped up right after the deepest cut was made. Im going to get the spoiler board attached today and give it another go.

Ill be sure to update this thread if the issue is not resolved.

Thanks again to everyone for opining.

To be fair, Kayvon suggested this issue at the very bottom of one of the earlier posts. I just gave more details about figuring out whether that was the issue :).
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

sudo
Posts: 23
Joined: Wed Jan 11, 2017 3:06 pm

Re: Z axis zero changes in middle of job

Post by sudo »

Indeed :)

Thanks again to everyone for their help.

User avatar
Kayvon
Posts: 552
Joined: Tue Oct 21, 2014 11:46 pm

Re: Z axis zero changes in middle of job

Post by Kayvon »

Glad you got it figured out!

Like most people here, I have a small ritual of sanity checks I do before making any cuts because, like most people here, I've made just about every mistake possible at every step. ;)

Post Reply