vacuum hold down table / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

vacuum hold down table

1 2 next

Move Thread

LAN_403

Ben Solon

Fri Aug 24 2012, 12:49AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

i've been doing a lot of pcb milling lately, and as i see it the only way i can squeeze more accuracy is to make myself a vacuum hold down table. right now i have a recessed rectangle that i align the pcb to the edges of, but to hold down the material i have to drill and tap a hole in the wood and then use a nylon bolt to clam the edge. but i get it off by a fraction of a mm each time. this often leads to the holes being off from the pads on one side.

what i want to do is machine myself some sort of vacuum hold down plate that i can hook up to a pump.your average shop vac can pull ~55 inches of water which converts to a tad under 2psi. the pressure at the elevating of my town(200ft) converts to ~15psi. i haven't been able to find any equations that i can use to tell me the holding pressure, nor would i know where to start. airflow reduces suction, so that needs to be factored in, as does the surface area in contact with the vacuum.

any ideas? my thoughts where on an thick aluminum 'box' with alignment edges on the work surface and a grid of holes leading into the vacuum from the work surface. the problem is that when drilling and cutting board outlines i don't want to shred the work surface.

for size reference i use the small "3020" machine found on ebay. my z travel is ~40mm, so it either needs to be thin or a replacement of the existing surface

Patrick

Fri Aug 24 2012, 02:47AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

ive had the very same issue with milling PCBs, unfortunatley my experiments with vacuum tech didnt proove useful, and i had a much bigger table than you (which i think made it easier for me than you). however what i found worked well, was to have a x and y groove towards one corner in a sacrificial wood table, then use epoxy or CA near the periphery of the PC board to be milled. then to free the part, the wood was milled down and the part released, it was inefficent since your always making several of these "sacrificial" table parts, which have to be milled to precise dimensions too. then your obliterating a thin section each time you make a PC board, but it works well for me.

It avoids all the problems of screw and clamping as you and i both know.

Ben Solon

Fri Aug 24 2012, 03:15AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

Ok, so I guess the vacuum is out unfortunatly. The problem for me is that I use emc(linuxcnc) so while I can use eagle, I can't do any 3d cad/cam using a parametric modeling program. I enjoy autodesk inventor on my windows box, but wine doesn't support the .net framework that it(and all other good cads) work on.

I read about tacking town the pcb, but I think in the end drilling and tapping two holes is a little easier.

I have a block of wood that I bolt down to the work surface. What i did was use my end mill bit to cut a recession out of it roughly the same depth as a rs pcb. Then I line up the pcb in one corner, bolt it down then drill a hole at 0,0 as well as all the other pcb holes. Then I route that side. Then I flip the board and line it up in the other corner. Set my 0,0 to the hole drilled at that point, then route that side. Then use MDI commands to cut ou the pcb. Whew!

I'm usually off by a fraction of a mm... I guess I am going to have to come up with some ingenious new way to clamp down the pcb without obstructing too much surface area. Ideally using an aluminum edge so it can stay in good condition.

I'll post what my setup looks like once I get out my camera and laptop. But let me know if you think of anything.

2Spoons

Fri Aug 24 2012, 03:50AM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Here's a really cheap way to make a vacuum table.
You need a piece of MDF (Medium Density Fibreboard), know in NZ locally as Customwood, or as 'Glit' to a mate of mine (glue + shit ). A piece 1/2" to 3/4" thick will do.
Seal the edge and back face with paint, then drill a deep hole in through one edge. This is where the vacuum is attached.

Believe it or not the stuff is porous enough internally to distribute the vacuum and pull air through the unpainted face.
I've used this trick to hold down pieces of glass. Best result if the entire vacuum surface is covered - in your case you could just use a bit of clingfilm to cover the unused surface.

And if you reverse the flow and pressurize the board you get the equivalent of an air table, and flat things float beautifully.

description of MDF on wikipedia

Patrick

Fri Aug 24 2012, 04:03AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

well with your MDF comments maybe im all turned around on the vacuum idea, ill have to try it.

Ben Solon

Fri Aug 24 2012, 04:06AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

Alright, that sounds like a good way to at least get something going. I think it'll take a bit more work than that because if you try to mill out the board outline and you don't have enough pressure especially at higher feed rates the board will slide slightly(or a lot). I might try distributing the vacuum to just below the surface through multiple air pathways that run parallel to the surface.

I think this would work a lot better than an aluminum structure because there won't be a problem or running the bit into the block at pcb milling/drilling feed rates. *SNAP* XD that's why I use nylon bolts btw!

Edit: patric, if you've done any small smd stuff like tqfp or smaller what type of bit did you use? I use a .6mm cutting edge diameter v tipped engraving bit. I think it's 45 or 60 degrees. I'm going to get a 10 degree .1mm v bit unless you have different recommendations. I have pcb-gcode set to .2mm just so it'll attempt the smd gaps, but it also takes too much away from the traces if I do that.

2Spoons

Fri Aug 24 2012, 05:15AM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

You may run into trouble with small boards: hold force will be directly proportional to area. May work if you can slow the feed rate enough (=less tool force).
Multiple bore holes for vacuum distribution is definitely worthwhile .

ConKbot of Doom

Fri Aug 24 2012, 03:45PM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

I used to operate CNC mills, and on some parts. They had a tall "skirt" on the side, and we had to do some rather aggressive milling on the top, we used vacuum fixtures to hold them down. Typically they were laid out like so

the shape of the hold down would be dictated by the shape of the skirt and internal features of the part, but a flat one would work fine for a PCB, obviously.

Your hold down force was calculated by the pressure delta from the vacuum channels to the atmosphere multiplied by the surface area of the vacuum channels, since the vacuum really isnt pulling down much on the areas touching the reference plane.

You said your shop vac pulls 2 PSI, so if you had 5 square inches of vacuum channel under the PCB, then you would have 10 lbs hold down force.

We used rotary vane or piston vacuum pumps and a reservoir and piping to all the machines, and were usually pulling 25+ inHg, which was ~12.7 PSI

For you, you could probably do the bed in wood or metal. and line in your PCB, and then lay plastic over the whole bed, and turn on your vacuum. Mill though the plastic and copper.

IVe seen the vacuum table + sheet over everything method used on how its made for various stuff, and I think an automated saw we used at another plant I worked at used it also.

Ben Solon

Fri Aug 24 2012, 06:34PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

thanks, that helps a lot!

the shop vac's pull the pressure *down* 2psi, thats not the delta. atmosphere(at 200ft) is 14.6psi. for a typical rs 4*6 board, the sa is 24in^2. so that equates to over 300lb of holding force! are you sure that those numbers are correct? it sounds a little on the high side.... but if this is the case, then i'm definitively going to be looking farther into this.

=========================================== ====================================

edit:

with a test using a spring-scale and my shopvac, the .994 in^2 inlet could hold .944 lbs verticly before unlocking. this gives a vacuum delta of .949. so something isn't right here... BUT. i'm not using an REAL shopvac, its a craftsman clean n carry. this could be the difference.

though using the .949 as a 'vacuum coefficient' to the surface area, it still yields accurate results. ie, the 300lb holding force becomes ~23lbs. this is still a verticle force though, so we can use that rating as a synthetic normal force to replace the insignificant gravitational normal force and then find the coefficient of friction of whatever material that its being held against to calculate how much sheering force(or torque) the piece will take until it starts sliding.

using a real vacuum pump would help of coarse. along with the pcb drill holes reducing the vacuum delta.

Ben Solon

Sun Aug 26 2012, 12:01AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

off topic, but do you know why you need a perfectly flat pcb to use a higher degree v tip? and i read an article a while ago that i cant find now that says how the degrees correspond to fine cuts, but i would assume that a 10 degree .1mm v tip would give a very good cut. i'm looking for a carbide v that'll cut out my tqfp's with plenty of space to spare(as opposed to the one in my avatar where i used a .6mm bit to cut a .4mm gap in two passes), and i have my eye on one of the dimensions mentioned abpve...

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob