G92 usage question with z-zero block

[Michael Burnside]

I had a friend give me a very nice z-zero setup gauge. It measures exactly 1" when "zeroed". I figured I would give it a try but wanted to get some feedback from any CNC experts. My CNC is a ShopSabre running WinCNC. My thinking is that I would insert a tool, measure with my tool height sensor and set XY as per normal. Then, with the gauge, "zero" it to my machine/spoilboard. To set Z0 I would use G92Z1, which would effectively set my zero 1" above actual. Then run program as normal. I haven't had a chance to try it, but this is my general thinking. I don't think this would change how my tool height offsets work and I believe the order is still correct.

I plan to give this a whirl this weekend, and if it works out, maybe I'll add this to my WinCNC INI to do this with a press of a button if I ever use the block so I don't have to type in the g-code directly.

Any feedback or ideas are welcome.

[Jim Becker]

I use the FTC (fast tool change) too measuring switch on my Camaster machine (also WinCNC) and haven't done a manual "touch plate" measure since I replaced the spoilboard two years ago and had to do a little magic to cut in for my tee tracks that are embedded in the machine bed. All my files use the top of the spoilboard as Z-zero. That only changes if I surface the spoil board and it takes about a half minute to do that. The singular downside is that I must measure material thickness with a caliper for job setup in the Vectric software, but I'm so used to doing that, it's no big deal. I'll design with an approximate and correct to actual thickness before writing out the final TAP file for WinCNC to do its thing with.

[Michael Burnside]

Thanks for the reply Jim. I pretty much zero to the board as well for most jobs, not sure why so many use material, but that's another topic . I usually just design for "nominal" thickness so I rarely even measure with anything more than a tape. That said, I have a couple of drawer fronts I've designed that I wanted to give a go with material zeroing. Turns out I was correct. Using G92Z1 did the trick. And this zero block is insanely accurate. Repeat measures resulted in less than 0.001 (basically sub 0.001) error! This is the gauge I'm using: https://www.mscdirect.com/product/details/67641191

I ended up adding a custom button to my WinCNC config next to my standard Z Zero called Z1 Zero, so if I ever use the gauge, it's a quick button press and and the offset is automatic!

[Jerome Stanek]

I use the FTC (fast tool change) too measuring switch on my Camaster machine (also WinCNC) and haven't done a manual "touch plate" measure since I replaced the spoilboard two years ago and had to do a little magic to cut in for my tee tracks that are embedded in the machine bed. All my files use the top of the spoilboard as Z-zero. That only changes if I surface the spoil board and it takes about a half minute to do that. The singular downside is that I must measure material thickness with a caliper for job setup in the Vectric software, but I'm so used to doing that, it's no big deal. I'll design with an approximate and correct to actual thickness before writing out the final TAP file for WinCNC to do its thing with.

If you use the top of the spoilboard why do you need to measure the thickness. Say you set the material to .75 but it is only .5 the cut will only go to zero or the top of your spoilboard. the first pass would be an air pass if you are doing 3 passes.

[Jim Becker]

If you use the top of the spoilboard why do you need to measure the thickness. Say you set the material to .75 but it is only .5 the cut will only go to zero or the top of your spoilboard. the first pass would be an air pass if you are doing 3 passes.

Because material is never nominal thickness. When using the spoilboard top as the Z-zero reference you need to know the "exact" thickness of the material so that depth of cut is accurate relative to the actual top of the material. If you are just doing cutouts, as you mention, you'll get the job done regardless, but if you are pocketing, v-carving, 3D carving, etc., you must have accurate material thickness when using the spoilboard as the z-zero reference.

[Jim Becker]

Ah, but even if you are making cabinet parts you must measure the material thickness to be able to accurately dimension the panels so that the structure matches the intended overall dimension. THis is no different if you were cutting the parts at the table saw. Sheet goods are never their nominal thickness, regardless of the measuring system used. The cutouts don't require exact thickness and I set that to the same as the thickness in the job information, but sizing parts that interface to, say...create a box of a given outside dimension...does. So since I need to do that thickness measurement anyway, it's my habit to put the accurate information into the file so it's never forgotten on a file that it really does matter. Consistency in process really helps with CNC work because the machine always does exactly what you tell it to do, including cutting things, um....err....wrong.

[Michael Burnside]

If you use the top of the spoilboard why do you need to measure the thickness. Say you set the material to .75 but it is only .5 the cut will only go to zero or the top of your spoilboard. the first pass would be an air pass if you are doing 3 passes.

Correct. I only measure if I’m doing delicate vcarving or I’ll just use the material as my reference. Pocketing or fluting or other tasks are designed against relative thickness. If I design something for approximately 3/4” I build it to ~3/4 but I don’t measure it or modify my vcarve setup. Anything custom is already measured to exact dimensions and designed that way in vcarve. If it’s tricky or dependent on both then I’ll surface a few thou off the part but that’s very rare.

This is especially true if I’m building cabinet parts with dados or rebates. I don’t care how deep they are (18mm or 17.90mm or 18.10mm is all the same to me), using the spoilboard will guarantee dimension of the piece. I don’t care how deep the pockets are.

[Kevin Jenness]

Michael, can you post a picture of the gauge in use? I mostly use the Camaster FTC feature like Jim. For zeroing manually I use a taper gauge between the spoilboard and bit.

[Michael Burnside]

Michael, can you post a picture of the gauge in use? I mostly use the Camaster FTC feature like Jim. For zeroing manually I use a taper gauge between the spoilboard and bit.

Done. See my second post for a link.

[Jim Becker]

Are you just jogging the tool down to that indicator like you would for measuring direct to the spoilboard? (Just curious)

[Michael Burnside]

Are you just jogging the tool down to that indicator like you would for measuring direct to the spoilboard? (Just curious)

Yep. I get close and then jog at 0.001 until the dial reads zero. It’s nuts too, my CNC and this gauge both read exactly 0.001 each step!

[Kevin Jenness]

Done. See my second post for a link.

So the device sits face up on the spoilboard and the bit depresses the circular button?

[Michael Burnside]

So the device sits face up on the spoilboard and the bit depresses the circular button?

Correct. And then as you jog down the needle moves. When it hits zero the bit is exactly 1” above.

[Carl Beckett]

May not be relevant - but am wondering if you even need to jog?

A lot of software you can INPUT the z height. So if your dial was on the reference surface - you just read whatever the value is and input that into the software as the offset. Whether 1.007 or 1.243 doesnt matter in absolute terms, since you type it in.

(I may not be understanding...)

[Michael Burnside]

You are understanding mostly. The issue is setting Z without moving the spindle or having to manually adjust Z given the current machine coordinates. G92 is meant for this purpose. I created a button now in WinCNC that does all of this automatically whenever I use the gauge. It eliminated “stupid human” mistakes