Inconsistent Z

[jbworden]

I'm having a problem maintaining an accurate Z setting. I'm making several ADA signs with raised lettering and Braille. I zero Z to the surface of the substrate then drill my holes for the Braille rasters. I then swap to the 60 degree V-bit and rezero again to the surface of the substrate. Using the Braille holes as a locator I can then cut small pieces of the 1/32 Rowmark used for the raised lettering and adhere it to the substrate. When I cut the letters the first few cut okay but as I move out towards the center of the table in the Y direction I am no longer cutting completely thru the letter material. It's like the table has a slight sway in the center. After several tests it turns out that I have to increase my depth of cut from 0 to .011 inches to cut thru to the surface of the substrate. I've checked the substrate and it's pretty consistent thickness. I just resurfaced the table (we've cut one small job since) so that shouldn't be an issue.

Any idea what might be going on here?

[Gary Campbell]

Jim...
When you suface the table vacuum may not suck down the bleeder all the way if it is not glued completely. When you put material on the table vacuum is incresed and the spoilboard actually sinks a few thousandths. (MAYBE?)
Gary

[jbworden]

I wish I had that problem, Gary. My vacuum motors are sitting in a box at the house. My spoileboard is plywood. I bolted a 3/4 sheet of plywood to the table then glued another 3/4 sheet to that. So far I've surfaced about 2/3 of the top sheet away so (I think) it should be level.

My subtrate is .25" PVC clamped to the table.

[harryball]

Couple of things to try...

Use a regular 1/4" or 1/8" bit, flat bottom. Zero to the table and then move down .005" enough to leave just a trace mark on your table top. With the bit at -.005" run an X with a box around it on the entire table, on 1/4 of the table and on 1/8 of the table, however far you need to go. You are looking for differences in the traces. If they dissappear in some places and cut too deep in others you can investigate why.

Cut 2 test pieces one after the other without rezeroing. If the first piece cuts through then does not as the toolpath progresses and the second piece doesn't cut through anywhere your bit (or something) is slipping.

Let us know how it goes.

/RB