[Joe Jones]

Hi,

I am planning to build a vacuum table for CNC routing of flat material. I need some advice from those of you who have made the mistakes already.

The table would be 5x10 feet, and I believe I would make chambers so that only the region(s) I need to hold down the material would be active. I am NOT an engineer, and I do not play one on TV. So I am willing to spend some money and "do it right" but I do not know what that means yet.

What should I use a the source of the vacuum? I would think first "a shop vac" but I am not sure that would produce enough vacuum to work properly. Also, a shop van relies on the flow of air to keep it cool, so using a shop vac to produce the vacuum seems illogical, since it would heat up and die an early death.

Is there a dedicated vacuum machine that can produce the proper amount of vacuum? I know Harbor Freight Tools sells something, but I am not anxious to "Go Chinese" for this project.

For the surface, I am told that 1/2" MDF is porous! That surprises me, but I was thinking more of a sheet of material with holes drilled pegboard style. But what diameter? And how close together? Should the hols be conical shaped? I would block off the holes that are NOT covered by the material with strips of plastic or whatever.

My thought was to make the chamber perhaps two inches thick, with internal ribs to prevent the surface from caving under the vacuum stress, but would a thicker chamber work better? Thinner? I just don't know! I am guessing that a manifold with ball valves would work to create zones on the table surface that would have vacuum while others are turned off.

All advice and suggestions from the pros will be greatly appreciated. I want to do this ONCE, and do it RIGHT!

Joe

[Joe Jones]

I am watching You Tube videos and checking out other posts here, as well as other sources. So much information!

I believe the biggest challenge will be to achieve a flat and level CNC router ... using a PlasmaCam table. The rails on the PlasmaCam are about as INACCURATE as one can imagine. There are no machined tracks or sexy bearings to maintain a completely horizontal movement of the gantry or carriage. The rails wear as the edge of the soft 11 gauge metal mushrooms down by being beaten upon with the hardened steel cam followers that roll along the edge of the steel channels, so there is a built-in degree of failure due to the fact that there is no height control while using a router.

For plasma cutting, the rails do not need to be perfectly level, or even share the same plane. The height control built into the PlasmaCam will compensate for slight variations. When using a router, the height control is disabled, and this makes the issue of the rails and table surface NOT sharing a single plane much more pronounced.

I was drooling over a large cnc router table I saw at FabTech 2019 in Chicago, but I could not convince myself to spend the money they wanted. Still, it would have provided a much more accurate cutting surface than I will ever be able to achieve with the PlasmaCam bolt-together table.

Wish me luck!

[machinedude]

Plasma cutting can be crude in table design largely because the process is not a high precision process and the tables tend to depend on THC way to much in my opinion. they also are not built to be rigid on the gantry's since there is no cutting forces there.the tables are built to be tough since they get hammered with loading sheets and plates but are not built very accurate. the gantry's are usually built light so they can accelerate at a high rate so corners can be turned on a dime at a high feed rate. not all plasma tables are built the same so the high end tables might be ok if things are solid enough for a dual purpose machine but that does not sound like the case here. sounds like a lot of things will have to be improved and beefed up taking a guess before you get to where you will be happy with the results.

i would take a hard look at the motors driving the axis's and make sure they can handle extra cutting loads placed on them and i would do some drive calculation's to make sure you can get to a resolution you can live with. there are all kinds of things to consider here and a lot of things you can look into before you spend all kinds of money and get in too a big project. i would check the machine backlash on all the axis's as well to see if your getting into more than you wish as well.

just a few point to consider and evaluate over all before taking the dive.

[joeavaerage]

Hi,
there are a good number of oil lubricated single stage vane vacuum pumps for sale. They produce a high ultimate vacuum
and flow rates commensurate with size. The down side is that they produce oil vapor in the output gas stream. If you don't
want oil mist in your workshop you will have to avoid them. Its a shame because they are a cost effective solution.

Note that the oil vapour occurs when there is high gas throughput of the pump. When you first hook up and 'pump down' your
plant there will be a high gas flow rate and the inevitable oil vapour results. Once the system is pumped down there is no gas left
to 'suck out' of the system and therefore the oil vapour problem stops.

If you can be assured that your vacuum system will maintain high vacuum integrity then a vane pump may be OK. If however you
anticipate the pump will have to maintain high gas flow rates for protracted periods of time then a vane pump is not a solution for
you.

In industrial practice two high volume vacuum pump designs are worthy of research. The so called 'liquid ring' pump and 'rotating
lobe' or 'positive displacement' pump are worth considering. Another type, it is a variety of positive displacement pump, is the scroll
pump. Often used in lab situations where oil vapour is forbidden, they have good ultimate vacuum but are expensive for their flow rate.

Craig

[machinedude]

as for vacuum tables a shop vac is not going to produce enough vacuum on a table that big. not to mention the level of vacuum is not as strong as a typical commercial vacuum system. with wood working you need a lot a volume and a lot depends on how well the system is built. for wood working 15 hg of vacuum would be ideal. and the surface area of the work piece has a lot to do with it as well. the smaller the piece the less hold down force you have. the cost of a vacuum supply on a table your size would be substantial. if your thinking of pulling vacuum through a spoil board even if the table is zoned you will need something big to get that result.

[ger21]

You need to be more specific about what you are doing.
The minimum fir a 5x10 table might be 4 industrial vacuum motors, up to a $10k 10hp rotary vane pump.

[machinedude]

You need to be more specific about what you are doing.
The minimum fir a 5x10 table might be 4 industrial vacuum motors, up to a $10k 10hp rotary vane pump.

i agree completely with you on this. this is a big table so the system would need a lot of CFM to keep up very expensive to say the least. i think maybe a pod type system might be more realistic on a budget project. you would avoid the high CFM requirements of a commercial system. and go with a strategic leak free hold down approach. the vacuum chucks for metal working don't require much CFM for a source but they always use gasket to seal them and don't leak. so maybe a good t slot table and pod system might be a better approach? like you said it's hard to tell knowing so little about requirement's.

[Joe Jones]

Thank you for the replies! I have been spending the day educating myself on this, and it is looking more and more like I will just end up buying a Shop Sabre Pro 510 table.

PRO Series - CNC Router | CNC Router Machines, NC Router Tables | ShopSabre CNC

It is already set up to do what I want, and I won't have to re-invent the wheel. It is $37,000.00 but it looks like it would be a THOUSAND times more accurate than the PlasmaCam table setup, even after I struggle to make a flat vac table for it.

Joe

[Joe Jones]

You need to be more specific about what you are doing.
The minimum fir a 5x10 table might be 4 industrial vacuum motors, up to a $10k 10hp rotary vane pump.

I have done a lot of routing with my Samson 510 CNC Plasma table, using assorted routers mounted in place of the plasma torch.

The table is somewhat accurate, but I wouldn't try to build a space shuttle with it. The biggest problem is getting the entire surface LEVEL. For simple signs, it isn't much of an issue, but if you are trying to create a large panel with an inset pocket for a sheet of mirrored plexiglass for example, the depth of the router cut at one end of the table might be 1/8" or more DEEPER than the other end. That is a problem when you need uniformity of depth for the cut.

I did put a spoil board on the grates, and planed it down to the rails, but even then, I saw a variation in the depth of the cut, simply because the Z-axis motor on the PlasmaCam tables is NOT that accurate. If it cuts a ring, and then raises to cut another overlapping ring, you may end up with a step in the material. I believe the cnc router tables of a more stout construction like the Shop Sabre would not have that problem.

[kmagyoyo]

Don’t overlook regenerative blowers, they are less expensive than rotary vane. I had a 15 HP regenerative blower on my 4x8 CAMaster and it worked quite well.

I have never seen anything other than rotary vane or regenerative on any industrial flatbed router. Almost all are rotary vane.

[Joe Jones]

Well, I believe I am going to take a shot at building a vac table insert for this machine. A hollow flat box with ribs to prevent the top from caving in under the vacuum pressure. I will study your suggestions and decide on one (or two?) vac devices that will provide me with enough suction to negate the need for screws and clamps. First I have to get through Christmas and New Years. They are both little more than obstacles to my progress. I don't even bother with Christmas decorations any monger.

[joeavaerage]

Hi,
I noticed in another thread that you said that you have a rotary screw compressor?

Would it be possible to hook your vacuum apparatus to the inlet side of the compressor? After all that's what a rotary lobe vacuum pump
is.

I use a small second hand refrigeration compressor for a vacuum pump for vacuum bagging fiberglass components. It is a reciprocating
piston pump. It can achieve about 20 in.Hg vacuum, not great but more than enough for my needs. I got it free when they re-did the local
butcher shop....so the price was real sharp!!

Craig

[Joe Jones]

Hey! Now there is an idea! I would have to find a way to be sure to filter out any particles from the routed material so they are not fed INTO the compressor, but ....

[machinedude]

Hey! Now there is an idea! I would have to find a way to be sure to filter out any particles from the routed material so they are not fed INTO the compressor, but ....

just out of curiosity if you have a screw compressor what kind of volume is the compressor able to do? i came across a manufacture in Ohio that made an adjustable venturi that would pull 25hg and had several sizes at a reasonable price. the problem i seen was the large ones used a lot of air. far more than what i have but a screw compressor might be different?

i think any air compressor can be modified to pull vacuum. i came across several video's of people doing it. i'm not sure how much a reservoir would help to keep up with demands? it's an interesting concept at the very least.

[joeavaerage]

Hi,
as I posted earlier I use a small piston compressor as a vacuum pump for vacuum bagging. It works fine.

Firstly, if you use low density polystyrene foam as a core excess vacuum will crush it. In fact about 1 psi vacuum, or about 2 in.Hg is about as much as it will
handle without being deformed. I make model aircraft wings this way, very obviously if the foam is deformed by excess vacuum then the shape of my airfoil is ruined.

I made a vacuum operated switch to turn the pump on and off. What I found was that the pump would cycle often every 20-30 seconds but for very short durations,
a few seconds at a time. This gave the electric motor a hard time. I made a vacuum ballast out of a disused automotive LPG tank. Now the pump will run
for several minutes until the ballast and attached vacuum bag 'pump down' to 2 in.Hg or so. Thereafter the pump might run for 20 seconds every five minutes or so
if there are no leaks in your vacuum bag.

If I have a hard core that can tolerate more pressure the I let the pump run until it achieves its ultimate vacuum, on a good day about 26 in.Hg.
Being a refrigerant compressor means that it has a reed valve. They work under vacuum, not hugely well but they do work. Sometimes a wee
squirt of oil can help the reed valve seal if you want just a little more vacuum. If I were being particular I would replace the reed valve with
a rubber 'duck bill' valve....but it works well enough so I've never bothered.

The other question is 'how much vacuum do you really need?' If you have a part 10 inches square, ie 100 in² then only 2/3 atmosphere
vacuum (about 10 psi or 20in.Hg) will 'clamp' the part to the table with 1000lbforce. Is that not enough? Even if you had a pump that could pull
a perfect vacuum (the unobtainable ideal) of 1 atmosphere (14.7 psi, 29.9 in.Hg) you'd get 1470lbforce, more clamping sure, but you pay
tens of thousands for high vacuum pumps....IS IT REQUIRED??

If you require high vacuum for resin degassing, vacuum desiccation, vacuum distillation then you require a near perfect vacuum of 0.001 Torr
( 1 Torr is near as dammit 1mm.Hg) then a recycled compressor just wont even get close, for those applications you need a vane pump and probably
dual stage at that.

I would guess that your screw compressor will get to about 20in.Hg, maybe a bit better, maybe a bit worse. That would result in about 10 lbforce
clamping per square inch of part area. Will that work?

Craig

[routalot]

If you are running a 10X5 table with a 1/2 inch MDF spoilboard you are going to need a serious high flow vacuum pump.I have spent a good bit of time running a 10X6 table and the pump has a 20 hp motor.Trying to get away with insufficient flow will lead to disappointment and not having a zoned table won't help much.You can help matters by laying a piece of impermeable plastic on the unused portion of the spoilboard.The challenge to be overcome is holding the smaller items in place as the clamping force is proportional to size and while holding tabs can help,having to clean them off each part soon loses it's novelty.

[ger21]

Yes, pulling vacuum through an MDF spoilboard is at the opposite spectrum of vacuum bagging, and the vacuum fixtures you see on mills.
When pulling through a spoilboard, you need a very high CFM pump, preferably in the 200-300 (or more) CFM range. A small fixture can get away with 2-3 CFM.
When cutting wood, you're always going to have some amount of leakage, which the CFM needs to overcome.

Cutting force also dictates how much actual vacuum you'll need. I big spindle, large tool, and fast deep cuts require a LOT more vacuum than a 1/4" bit taking shallow cuts.

[Joe Jones]

Wow! You guys sure know a lot more about this than I do!

Using MDF as a spoil board, doesn't the surface suffer vacuum loss when you begin carving into it? Should I start with a solid frame and a steel (or aluminum) perforated grate, and place a spoil board on top of that with the edges sealed by ... paint? or Glue?

I was hoping a vac table would not be so complicated. Apparently there is a lot more ot it than just building a hollow box with a porous top surface!

[ger21]

The spoilboard needs an air tight surface underneath it, to hold the vacuum.
You also need to remove about 0.01" from each side of the MDF, to allow more flow through it Yes, as you expose more of the spoilboard, you get vacuum leaks, and holding force decreases.This is part of why you need very high flow rates.

[Joe Jones]

It seems easier to create a board for pegs around the outside of the material, and drive wedges between pegs placed in holes and the edges of the material to be routed.

I was using strips of plastic with two screws. I would place the plastic strips against the material and run the two screws down into the spoil board on all four sides. This worked okay, but being the lazy S.O.B. that I am, I was looking for something more "sexy" and easier to do.

Has anyone experimented with 3M spray adhesive on the spoil board, laying the material onto the adhesive and allowing a bit of cure time to hold the material in place? I believe it could be pried up after the routing was done, and then cleaned off with mineral spirits or ??? Maybe that would create a monstrous gooey mess after several pieces were cut. :devious: What about a sheet of spongy rubbery non-slip material like they use in the bottom of tool drawers? Hmmm... If a small patch of it can hold a cell phone to a slanted surface in a moving car ... would a LARGE sheet of it hold material being routed?

I found 16"x16" metal pegboard tiles at Lowes. They are nice, but of course they would not serve as a spoil board.

http://www.lowes.com/pd/Blue-Hawk-1-Piece-Steel-Pegboard-Actual-16-in-x-16-in/50220687