Page 1 of 1
Z zero seems to change when I run gcode file
Posted:
Tue Feb 07, 2023 7:03 am
by DoolinGuitars
I'm carving archtop guitar tops with a CNC router with a UCCNC controller. I used VCarve to generate the file, using the Mach3 post processor. I'm using a 1/2" ballnose bit. I home the machine and zero to the material surface, then load the file and run it. When it gets to areas of Z0.0000, it cuts into the material about .015". If I jog the machine to Z0.0000, run the spindle, and jog in X or Y, the bit does not cut into the material, it just barely grazes the surface, as one would expect. Somehow running the gcode file seems to add about a -.015" offset. I checked in the Offsets page of UCCNC and there is no Z offset.
Here's the start of the file up to the first cut at Z-.125, and then much later the first cut at Z0.0000 (which shouldn't cut at all but does) -
G00G20G17G90G40G49G80
G70G91.1
T8M06
G00G43Z0.8000H8
S11000M03
G94
X0.0000Y0.0000F150.0
G00X14.0576Y3.5837Z0.2000
G1Z-0.1250F35.0
G1X13.8063F150.0
.
.
.
.
G1X4.1985Y9.3540Z0.0000
G1X4.1761Y9.3413
.
.
Can someone help me with this? Or if this is the wrong place to post it please let me know where it should be posted. Thank you!
Re: Z zero seems to change when I run gcode file
Posted:
Tue Feb 07, 2023 9:41 am
by thegreatwaldo
Check your safe z in the vectric program. I had something simalar where my safe z was too high for the cnc machine and cutter and would pull up and loose some steps
Re: Z zero seems to change when I run gcode file
Posted:
Tue Feb 07, 2023 10:51 am
by Battwell
in your tool table. is there a length set for tool 8?
G43 H8 in your code is telling machine to offset to tool 8 length.
if you are using fixed tools with fixed lengths this is correct.
if not zero all tool lengths in the tool table.
Re: Z zero seems to change when I run gcode file
Posted:
Tue Feb 07, 2023 1:42 pm
by DoolinGuitars
Thanks guys. But my safe Z is only .2" which puts me about in the middle of the 4" Z range with the spoil board and workpiece, and I don't hear it missing steps. And my tool table is all zeros. I don't have a tool changer, just a wrench. Also, I'm zeroing with a Triquetra and have tweaked the settings to within .001", and have calibrated the Z axis to within .0001"
Re: Z zero seems to change when I run gcode file
Posted:
Tue Feb 07, 2023 8:43 pm
by ger21
and I don't hear it missing steps
You can lose steps gradually, without ever hearing anything. Especially with 3D carving. Try lowering your Z axis acceleration 25%.
Re: Z zero seems to change when I run gcode file
Posted:
Wed Feb 08, 2023 12:56 am
by DoolinGuitars
That sounds like a good idea anyway, what's the hurry?
Re: Z zero seems to change when I run gcode file
Posted:
Wed Feb 08, 2023 1:14 am
by Delco
Its in your Gcode G1Z-0.1250F35.0 so its coming from something you have set in your cam program.
Re: Z zero seems to change when I run gcode file
Posted:
Wed Feb 08, 2023 6:49 am
by DoolinGuitars
No, G1Z-0.1250F35.0 is just the first cut at 0.125" depth. That's in reference to z zero, 0.125" below that. My problem is when the code says G1X4.1985Y9.3540Z0.0000, it should raise the bit to z zero, so it doesn't cut at all. But instead, it raises it to something like z-0.0150, so it cuts 0.015" into the wood. That in turn means that G1Z-0.1250F35.0 is actually cutting at a 0.140" depth.
Re: Z zero seems to change when I run gcode file
Posted:
Wed Feb 08, 2023 8:40 am
by Delco
DoolinGuitars wrote:No, G1Z-0.1250F35.0 is just the first cut at 0.125" depth. That's in reference to z zero, 0.125" below that. My problem is when the code says G1X4.1985Y9.3540Z0.0000, it should raise the bit to z zero, so it doesn't cut at all. But instead, it raises it to something like z-0.0150, so it cuts 0.015" into the wood. That in turn means that G1Z-0.1250F35.0 is actually cutting at a 0.140" depth.
How much backlash do you have in that axis , can only be that , a slipping coupler or lost steps
Re: Z zero seems to change when I run gcode file
Posted:
Thu Feb 09, 2023 9:46 am
by DoolinGuitars
Sure enough, ger21 called it. I reduced the Z axis acceleration. I think it was losing steps in that first Z0.8000 move. Problem solved!