ShopBot - Cutting Plywood video

[hespj]

www.youtube.com/watch?v=XteEyLYvtXA&feature=c4-overview&list=UUF0Y_ypEky7cd0DWyLFoU1w

The presenter shows how he Z-zeros on the spoilboard rather than the top of the ply and gives some reasons why. I've been Z-zeroing on top of the ply for the past eight years, and whilst it's true that I sometimes accidently don't cut right through I think this is outweighed by the fact that I can keep Z-zeroing on top of the ply as long as it stays held down. And yes, I can add a little depth to the cut to make sure I do go right through (but not so much as on the spoiboard he showed).

I'm not saying I'm right, and if the arguments are more in favour of Z-zeroing on the spoilboard I'll change - I always Z-zero on the spoilboard for 3D stuff, but then, I've never made a 3D object which took up the whole spoilboard.

So those of you who Z-zero to the spoilboard when cutting sheets - how do you Z-zero if you're using more than one bit or if you have to re-Z-zero the first bit? I'm talking about ful sheets that take up the whole spoilboard.

John

[steve_g]

John

I think the important thing is to be consistent! Do it the same way every time. When V-carving, Zeroing to the top of the material will give the correct results without worrying about material thickness variation...

I tend to design to "standard" thicknesses but the material is seldom standard! To allow for this I tend to Z-Zero to the bed and raise the head the design thickness and re-zero. This makes my perimeter or through cuts the right depth but may make my pockets and grooves off a little bit. I'm not making cabinets, so this method usually works well for me. This method does pose a problem when re-zeroing when the bed is completely covered, but lucky for me... My Bot is a dual Z machine. What this means is I have 6" of bed on the negative side of the X axis. This area is available for zeroing unless I'm indexing my parts that direction. Some people use a "touch" plate located off the bed. This plate is a known distance from the bed and compensated for in the zeroing routine ... This compensation, either physical or digital has to be changed each time you resurface the bed.

I haven't watched the video... yet. But if your method has been working for you the past eight years, I wouldn't change. For every reason to do it one way there's a reason to do it the other! The important thing is that either way is available to you if something becomes compelling for whatever reason.

SG

[hespj]

Steve, that all makes sense. Thanks. I think you're right - be consistant or some expensive mistakes will happen.

I've often thought the touch plate would be a good idea, not just because of the reasons discussed here, but because it's probably more consistant than the moveable z-zero plate, but I've never really looked into how to allow for the difference between the surface of the touch plate and where I want my zero to be.

John

[steve_g]

"but I've never really looked into how to allow for the difference between the surface of the touch plate and where I want my zero to be."

John...
The only touch plate I've seen in action was on a "Joes 4X4" running Mach III. Hopefully a ShopBotter will let us know where and how he has implemented this zeroing feature!

SG

[blackhawk]

John - Here is a thread that I started. I setup a dial indicator to zero between bits.

http://www.talkshopbot.com/forum/showthread.php?t=13049

[hespj]

Brad, that was an interesting thread. I wonder if instead of the adjustable bolt a block of soft aluminium could be fasted adjacent to the spoilboard. When the table is resurfaced it resurfaces the aluminium too. I've never machined aluminium so don't know how practical that would be.

I guess the problem with this (apart from the machining) would be that the spoilboard might expand or contract so that it was no longer flush with the top of the aluminium. But then, if it started doing this it would probably no longer be flat, which opens up a whole other can of worms.

John

[jTr]

I wonder if instead of the adjustable bolt a block of soft aluminium could be fasted adjacent to the spoilboard. When the table is resurfaced it resurfaces the aluminium too.

Many have utilized the aluminum block or similar aluminum shape in this way. I wanted a quick fix, so simply used a wooden block and still utilize the original zero plate. Modified my setup to go to a favorite ZZero point of -.40 in X and 5" in Y. In general, especially for plys, I do not run more than a 3/8 Diameter bit, so the bit fits past the plywood edge, even if table is completely covered. -.40 is just enough to do this without tripping your prox switches, as long as you've got them tuned for a tight response.

I zero to bed almost exclusively, and have experienced greater consistency. I do numerous tool changes when cutting plys, and things work well this way.
Since doing so, I no longer sweat over actual thickness of plywood, which is always in flux, even with a large stack from the same batch.

This fluctuation is transferred to the inside of plywood cabinets when zeroing to the bed. For instance, a quarter inch deep dado is always executed at a 1/2" Z height. CNC will cut fixed shelves (cab floor, stretchers, etc.) exactly to specified length, so if Z is at a fixed depth and ignores variables, you can always rely on exterior of box to be dead on desired width.

If You zero to top of ply, the fluctuation is transferred to the outside of the cabinet, as your ply thickness variable +or- is transferred to that Z height.

It is your choice where you want that inconsistency to land. Since I often do a "hybrid" frameless cabinet, I rely on the exterior being the proper dimension, so this works best for me.

Often times, this results in 1/32 to 1/16 wider inside cab. For side mount drawer glides, things work better when you have this relief built into the cabinet width.

jeff

[knight_toolworks]

I use a block of aluminum threaded for a bolt with a big brass knob. have a wire running to the z block in my bot. so a little routine jogs to it zeros drops the bit .121 then rezeros then asks for the thickness of the material. so it is the best of both worlds get the accuracy of the bed zero with the ease of using the material. if I do vcarving and such I may zero on the material.

[myxpykalix]

John,
"And yes, I can add a little depth to the cut to make sure I do go right through (but not so much as on the spoiboard he showed)."

When you say that does that mean you go back and re-toolpath to add a little depth to finally go thru the material?

If so here is a quicker and easier way to accomplish cutting thru.
If i find that i'm not cutting all the way thru what i do is just lower my Z to 0 (material surface) then lower it by .05 or .1 and REzero to that point and run the file again. Takes only a few seconds.:D

[hespj]

Jeff, thanks for the detailed description. It all makes sense.

Steve, that sounds like a neat system although I don't understand it completely. Is the bit zeroing on the end of the bolt and if so have you previously adjusted the bolt so that it's level with the bed and how have you done that? Then there's the routine - not something I could easily do.

Jack, yes sometimes I go back and recut if the bit hasn't gone all the way through, but that's usually an error. I sometimes do as you suggest, and sometimes specify a deeper depth of cut than material thickness in the software. VCarve Pro throws up a warning if you do this, but that's okay, and RhinoCam lets me visually see where the toolpaths are and how deep into the spoilboard I've gone. It also allows the last cut level to be specified differently - for instance on 12.5mm ply I might use three 4mm deep cuts with a final 1mm deep cut = 13mm.

John