Forum Navigation
Please or Register to create posts and topics.

DIY Autoprobe and Vacuum Hold Down

(Original thread started on 12-31-09 by Mark L.

Okay fellow CNC'ers! I don't know if your aware of this, I know a few are not, but this is something you gotta take the time to make. I made mine in a few hours tonight after a trip to the hardware store.

 

Too many times I've had something happen and lost the x,y coordinates to start a job over or wanted to start a tool path in a exact spot and never get it quite close enough. Not to mention 15 minutes putting around with jogging to get it as close as I can. Now I can just put my viewfinder on the exact spot, push a button on the Mach3 console and in 30 seconds, I'm sitting dead center where I want to be. To see what I'm talking about, watch the following video and go to the CNCZone forum to see how it's constructed. There's a few with some other mods as well that are really cool.

 




 

http://www.cnczone.com/forums/showthread.php?t=56079

 

Since I made this the other night, I've used it quite a bit on some parts I've made. I can't imagine not having this again.

 

Using the flat angle to set the Z is what I use the most. My offset is .122, no more futzing with a piece of paper to set the Z. Let it touch the plate and I'm off and running.

 

Scott and Ron, if you build this, our controller doesn't have the handy terminal block on the board so you either have to get one and solder it on or just solder the wires to the controller board like I did. I also used pin 13 to leave 15 available if I put a E-Stop on. The instructions also mention you might need a capacitor for debounce, I bought one but did not install it and have had no problems to date.

 

One another note, this little piece of information makes a difference and only reading thru 40 pages of the thread would you find it. Once everything is built and tests OK, to get the Z to calibrate correctly, you enter the thickness of whatever you are using for your touch plate, in the box under the 'Z' button so it can correctly calculate and set your Z. The trick is, when you enter the numbers on the Mach3 screen for the probe section; ie: .122 press the Enter Key! If you don't and press the 'Z' button, it will go down, touch the plate and that's it. It thinks it is where it's supposed to be because it never got the offset which happens only when you press Enter after entering the value. When it works correctly, it calculates zero to be down from it's current position plus the offset you entered and sets that as zero then moves the Z axis up the offset value plus .25, in my case I always end up at +.372 above my material after probing the Z axis.

 

Just touching off the Z axis to auto set the top of material is worth it. I used to stick a piece of paper between the bit and material, then switch the jog to step mode and jog down until the paper was indicating I was close enough, set the Z and then reset back to normal mode etc etc. Now I just throw up the touch plate and press the button and I'm done. The X,Y positioning though is really sweet for moving to an exact location. I plan to also build a isolation probe I can use to plot with eventually, Mach3 has a plugin for that!

 

Additionally, I needed a way to hold my copper clad as flat as possible, one way to do this is with a vacuum plate. Since aluminum would have been great, but not in the budget, I opted for a 3/4" piece of plywood.

 

I milled out the backside .5" with a few staggered supports to keep it strong and a recess along the edge and then made a cover plate that I sealed it up with along with some silicon sealant. I then flipped it over and drilled the holes for the vacuum hose and then the holes themselves. After that I milled the surface area with a .5" straight bit at 0 and 90 degrees:

Markl 5

 

This is the part I don't get. If I milled it out flat, hooked up the vacuum source, put on my copper clad (which stayed in place brilliantly I might add!), why then when I got to the middle, did it cut through the .01" clad board, but on the edges cut only .005" like it was supposed too? I had verified that the entire surface was sucked down flat, yet it milled thru the center. If the material was thicker in the middle, there should have still been .005" left. I know I didn't spend $10k - $30k for this, but I really thought it wouldn't vary like that after just having been milled flat.  Any ideas?  I should add, that as long as I don't have a .005" tolerance to adhere to, this worked great.

 

(Posted by Ron Rollo on 08-11-12)

My first thought was that the suction from the vacuum was pulling down on the board. But forget that. I don't think the vacuum would be strong enough to do anything in this case and if it was strong enough, it would have the opposite effect.

 

So my real thought is that your cooper clad in not perfectly flat. You might find that if you hold the piece up and look straight down one of the edges, you'll find some "Line integration" or a slight bow or arc. It will be ever so slight and maybe even invisible to the naked eye. But we are talking about .005" over several inches.

 

I would see if I could detect this arc and face the convex side down onto your board and the use some painters tape around the perimeter to help pull that arc out and make it as flat as possible. Hope this may ring a few bells for you.

 

(Posted by Jason Hite on 08-11-10)

Most copper clad board is not perfectly flat. Heat and mechanical stress can cause it to warp. I would suggest using double sided tape to help hold the material flat. I have had lots of issues with this myself and this is the only thing that assures the copper stays flat on the table.

 

I love the vacuum table though, I think I will have to do this as well. Can you post some pictures of the reverse side of the vacuum table if you ever disassemble it?

 

(Posted by Mark L. on 08-11-10)

Well, I understand that it could have bends etc, that's why I went with the vacuum path. Before engraving, I checked at different points and it seemed tight to the surface. There's a lot of pull from the vacuum I hooked up. .010 copper clad is not very strong stuff to resist the vacuum in my opinion. But something definitely caused it.

 

The reason I didn't use double sided tape, is I tried that before doing this. Got the same result, actually a little worse. If you don't tape the entire area, you have pockets that will sag as the bit goes over it. Remember, this is .01 scissor cut I'm using, very thin stuff. So I'm looking for single sided, thick enough to be a little stronger and yet flexible enough to make the bend for the TQ cover.

 

Making the table was easy, I centered a 12"x12" area on the 14" wide board (depends on width of your machine) and milled down a .5" with a .25" lip. On one end where I connect my vacuum hose, that was also milled out the same with the lip. Note, just make this area for connecting the vacuum far enough away so the router won't run over it. I then used some .25" Masonite (peg board material without the holes) and cut it to the milled out area so it would sit flush in the lip. It's held in place with the sealant. On the top side, I drilled staggered .125" holes 1" apart (CNC is great for that!) and cut out the hole for the hose. I had a vacuum cleaner that died, so I stole the hose from it which allowed me a nice connector for a twist and lock onto the plate and the other end hooks up to the vacuum cleaner that works.

 

(Posted by Geneb on 06-30-11)

Here's a great method for doing this kind of thing:

http://www.talkshopbot.com/forum/showthread.php?t=11481&page=4

 

Here's my result in using the same process:

http://www.geneb.org/shopbot/cutting_alu.html

 

It would work equally as well for doing copper clad board.

 

(Posted by Mark L. on 06-30-11)

I actually haven't used this but twice after I made it. Mainly because of tolerance issues, since I have now replaced all the axis nuts with real 'anti-backlash' nuts, I just may have the tolerance I was missing before. Gene, the links you provided though also peak my curiosity and I may have to try that.