[MARV]

I think you guys are getting close, but don't discount the notion of lateral slip. After all, a good deal of the cutting forces are in the XY plane. Anything you can do to minimize slip is to your advantage. Sometimes a reference edge on the work piece doesn't need to be cut, so a temp fence will work (think dbl stick tape). Other times sticky back abrasive is the answer. Here is a place that I used to buy from that may offer some interesting solutions (no affiliation);

All Star Adhesives - gasketing products for CNC routers and v-groove tape for v-grooving and coving applications

As a side note: my vac hold down system is powered by an eBay sourced, lubricated rotary vane pump made to pull down to 29"+, run all day long, was rebuilt/warranted and cost me $400 ($2400 list I think). They are out there. Its intended to service my 3' x 4' CNC gasketed and sealed MDF top as well as a vacuum press (not together). No spoil board. The key to making this system work is few leaks and some thrown together fixturing..

[lambodesigns]

Some more lessons learned from a few hours of cutting this morning.....

My tabs are really a combined onion skin and tab of .02" which worked great.... most of the time. The plywood is 3/4" prefinished maple and not what I would call stellar quality so some it is warped "a bit". If the area where I'm cutting the small parts is warped at all and not absolutely flat to the table it ends up trashing the parts because the dust brush pushes them into the bit. I'm going to up the thickness and see how the rest of the day goes but out of around 10 runs I've lost 2 sets of the parts which is more than double what I lost in over a 1000 runs before Trying to look on the bright side with the tabs.... it is overall quicker because I just lift the whole piece off the table and set it aside instead of picking each one out. Then after I get the next run going I pop them out while the cnc is running so overall at the end of the day it is faster with the tabs.

Canman77.... I got started on this because a friend gave me a small vacuum pump. It was one that required a lot of air to be moving to cool it down so a good seal meant it was overheating in no time. Without a good seal it didn't generate enough force to hold much. Worked fine but only for a minute or two at a time. Found out my shopvac worked the best of anything I had and used it for a little while. Really is amazing how much force they generate. Then when I got the large order in it justified the upgrade to 4 of the good ones and a dedicated box for them. Good luck on your testing and I think you will be happy with the results. If the mdf and a shopvac works for your test then ldf and a shopbot vacuum motor should be the ticket.

[MARV]

I should mention that when I first started out years ago, someone handed me a cast off refrigeration pump from a milk house operation. Two stage piston pump with separate motor drive. It didn't owe me anything, so I swapped output to intake, ran it all day at 24"HG and it never quit. Sometime, you make do with what you've got.

[ger21]

I think you guys are getting close, but don't discount the notion of lateral slip.

Yep. Parts will slide with far less force than it takes to lift them off the table.

[lambodesigns]

I use that same trick when I mount ice tires on my go karts. We cut the sidewalls off of old tires and then run screws through them. To mount the "skins" we use the intake from an air compressor to suck the tires down so that we can slip the skin over them. Funny to see a round tire get sucked into a square shape

I also have a regular vacuum pump that will pull a ton a vacuum as long as it is perfectly sealed that I picked up for $20. Uber quiet and will run forever. Once you lose any seal and it doesn't have enough cfm so vacuum goes to zero in a heartbeat. Someday I'm going to add a vacuum bag to my shop for gluing stuff up but I've just never found the justification for one because they aren't cheap.....and I am

Ding... fries are done... time to go load more plywood

[lambodesigns]

Yep. Parts will slide with far less force than it takes to lift them off the table.

I'm also using prefinished plywood and that stuff is slick as can be. Takes a bit more force to keep it in place over unfinished plywood.

[ger21]

Someday I'm going to add a vacuum bag to my shop for gluing stuff up but I've just never found the justification for one because they aren't cheap.....and I am

I got tired of veneering in a vacuum bag, and built a frame press on a 40x90 table.
I'm buying a new house, and was going to get rid of it to save room, but I used it a few more times, and can't part with it, as it's just so handy to have.

[canman77]

You guys beat me to the punch, my next question was going to be about lateral/cutting forces. From my reading online, it seems like that is a tougher number to figure out...not the cutting forces, but rather the friction coefficients of various materials and what they need to be in order to achieve the goal. Obviously, cutting faster and deeper is going to require more hold down power to overpower the lateral forces. I assume for most, it's just a trial and error thing like lambo has talked about with his latest run.

My weekend is starting to pile up with various jobs already so I don't know if I'll be able to get to testing anything until early next week. I'm expecting a slower work week so if I finish up early a few days I'll be able to test.

For what I'm doing with the prefinished ply, honestly...I'm not opposed to tabs, but I'm sick of the veneer chipping a little bit when the tool plunges again after cutting a tab when going against the grain. So even if I need to still use tabs, but less frequently and only with the grain, my workflow and end product will improve. Plus, not having to use screws for workholding would be super convenient.

[MARV]

Canman77,
This is where guessing and supposition comes in (read trial and error). If you were using a new surfaced spoilboard and could verify that your work piece was equally flat and pulled down well to it, then its reasonable to believe that your spoilboard would "backup" your part and prevent blowout. Much of this goes down the drain, however, as soon as you cut a different part. This why any well oiled cnc program within a commercial/industrial shop will have a fair understanding of the range of their workpiece requirements and try to fashion generic solutions to accommodate.

Vacuum work holding can be a maddening endeavor. Much of it can be overcome by knowing the fundamentals, thinking about the part in hand and applying a little and error. What is challenging for many guys on this forum is they want it to work for the one part they're going to cut today. Tomorrow, the part in hand is completely different.

Stay with it. I think what attracts most people here is learning and making stuff and vacuum work holding is a component of that.

[canman77]

Canman77,
This is where guessing and supposition comes in (read trial and error). If you were using a new surfaced spoilboard and could verify that your work piece was equally flat and pulled down well to it, then its reasonable to believe that your spoilboard would "backup" your part and prevent blowout. Much of this goes down the drain, however, as soon as you cut a different part. This why any well oiled cnc program within a commercial/industrial shop will have a fair understanding of the range of their workpiece requirements and try to fashion generic solutions to accommodate.

Vacuum work holding can be a maddening endeavor. Much of it can be overcome by knowing the fundamentals, thinking about the part in hand and applying a little and error. What is challenging for many guys on this forum is they want it to work for the one part they're going to cut today. Tomorrow, the part in hand is completely different.

Stay with it. I think what attracts most people here is learning and making stuff and vacuum work holding is a component of that.

I did not get around to building a little vacuum test this weekend, time ran away from me and I had way too much to do. The second half of the week and this weekend will be a lot more open. It actually works out because I can bounce my basic design off you guys and avoid some hassle before getting too deep in the project.

My expectations have changed for what the vacuum table can do and I basically only want to use it for sheet goods at this point, with bigger parts. Cabinet panels, shelving, etc. I'd still like to be able to remove the table when needed, so I still plan on having some flanges at the bottom of the table (well...more like box at this point I guess) with holes that line up with threaded inserts in the spoil board. I am going to draw up plans this evening for the box and will upload images later so you guys can tell me where I can improve. Should be easier than describing what I have in mind. Stay tuned!

[lambodesigns]

yeah... sometimes it is a matter of getting the recipe right for each part. I upped my "onion tabs" to .03" and didn't lose another part from it moving on me. Ended up adding tabs to the next size up part (around 5 square inches) so they would stay put. They cut fine but would get moved when the dust brush ran over them because they were close to another part and sometimes the waste piece would hit the bit. It didn't add any time to the run and the tabs were no problem to just break over by hand and the sander didn't even notice them. Also had zero chipping of the veneer.

Just finished drawing up a set of laundry room/garage cabinets for the neighbor and we are going to cut them tomorrow. I did his kitchen cabinets and while we got them cut there were a few times we had to pause the job and tape over the holes from the pegboard layout when I couldn't position the parts to eliminate cutting over the holes. Layout was much easier for this job and since the smallest piece is 12x18 I don't plan to use any tabs. Hopefully, it will go as smooth and I can declare victory on this project.

[canman77]

So here is what I'm looking at so far, based on everything I've seen online. Will definitely refine the model. The inside of the box is as seen here:



The inside measures 3" tall, which is more or less a guess based on what size hole I'll need to hook up any shopvac or vacuum pump. Not super important at this stage, I'd imagine. The inside cross pieces have holes for airflow which are really only necessary for gluing the top on, or adding another vacuum port somewhere else on the box. I don't see any other advantages to these holes, but they aren't hurting anything as far as I can tell.

The base is 0.75" bigger on all sides of the box, which will fit perfectly on the working area of the machine. I can use the holes to put the box in the same spot every time and not have to worry about resurfacing anything any time I need to use the vacuum box. I hope. Nothing inside the box needs to be sealed as long as the outside of the box has a solid seal, as I understand it.

Here is the top of the box:



Nothing special here, this design is all over the internet. The vacuum channels extend to 0.5" from the edge of the top. I believe all of these surfaces will need to be sealed? Or perhaps just the edges of this board? I am planning on using MDF or LDF for a spoilboard, so vacuum will be pulled through the entire board. As such, the top of the grid surface needs to be raw to accept wood glue. Then the edges of the MDF/LDF will be sealed.

Am I on the right track? I know it's a very basic system and it's all over the internet, but basic is good enough for me. I don't want to get into valves and turning vacuum on/off in different areas of the table, since this is already being designed for a handful of (repetitive) operations.

[peteeng]

Hello Canman - If you are serious about vacuum clamping you need to use a real vacuum pump. Using a blower type fan does not give you the vacuum you really need unless you are cutting large flat panels all the time. I use an oil vane vacuum pump that gives well over 99% vacuum. This means I get about 10 tonne per sq metre clamping. I use 12mm garden hose and make MDF bolsers that are consumable. I use a non slip cloth as a gasket for the first cut then it gets cut and becomes a custom gasket. You can also mill a groove and use foam rod or rubber extrusion as a seal. Small parts require a custom bolster but that's easy on the router. Making a 100mm thick vac table like you have there is a waste of height and volume. The ht of the vac table is driven by the tube size not whats required. The table could be 10mm thick and it will work, just need to get a tube to it somehow. The pump has to deal with that volume so becomes inefficient. Ideally in a vac system the reticulation side should have minimal volume and no leakage. You are not actually shifting air or pumping air so the size of the pipe can be small. Small vac pumps are not expensive either. Peter

[MARV]

Canman 77,
My experience backs up what Peter just posted. I'll admit, early on I did much of what you are doing. The results were not satisfactory.

The essence is this; large pumps on commercial systems are meant to overcome lousy setups and leaky surfaces. They are also VERY expensive to purchase and maintain. Large hoses on these systems are meant to accommodate quick draw down, but also are needed to provide cooling air for the pump (many commercial pumps to be used with CNC routers are throttled at around 20"HG for that reason). Breathable spoilboards are inherently inefficient as they restrict air flow (if you think this is wrong, meter the vacuum both before and after the spoil board). Most folks doing work on the side will find it difficult to justify the costs of similar systems.

I've taken an alternate route; I've powered my system with an oil lubed rotary vane pump (1HP, rebuilt, $400, made to pull 29"HG 24/7). I port that to my work holding surface via 3/8" ID (10mm) braided vinyl hose. My work surface is 1" MDF (no plenum), milled with 1/4" x 5/32" deep channels to receive 1/4" closed cell foam cord. All surfaces are sealed with epoxy (top, bottom, edges and grooves). The vacuum is supplied thru a valved distribution manifold to a number of zones. Each zone inlet is centered on a pad with a channel cut to vent to the main channel system. Gasketting is laid into the channels to approximate the shape of the work piece. If I need to cut the full depth of the work piece edge, I use a generic, breathable "riser" (usually 3/8" MDF and sealed where needed) to elevate the work. If I trash anything, its usually the disposable riser.

Admittedly, this isn't a quick and dirty system. Takes a bit of planning and execution, but is very serviceable. And allows me to pull a measurable 25"HG at the work piece. Hope this helps.

[canman77]

Thanks for the input guys. I feel like I need to clarify what I want to use the vacuum table for, and Peter touched on it. I just want to cut large flat panels with the vacuum (cabinet panels, etc.) and I'm not really even considering cutting small parts with it. My main objective is to get away with no tabs, or at least fewer tabs, which would dramatically increase productivity. Cutting tabs is a hassle, the veneer can chip, I have to flush trim the parts on the router table...I'd be thrilled to remove these steps with a vacuum system.

I realize a shopvac isn't a great option but people still report decent results so like I mentioned earlier, I just want to test with a shopvac for proof of concept. If it works, I plan on buying one of those shopbot vacuum pumps. They appear to give 5" HG which isn't amazing, but it should be enough to hold down a 12" square which is the smallest cabinet piece I'd cut.

For one off jobs that are pretty unique (custom guitar bodies, necks, etc.) I have no problems with tabs, especially because I don't generally have to worry about any veneers that can chip.

I still feel like I'm missing a critical piece of the puzzle that everyone else seems to know that I haven't figure out! For simple large flat pieces, won't my plan yield somewhat decent results? Ignore any small parts, I don't care about using vacuum for them.

[ger21]

What one considers "decent results" is relative. It also depends on a lot of factors, including tool size, and feedrate. Controlling leakage is critical to getting serviceable results

I just want to test with a shopvac for proof of concept. If it works, I plan on buying one of those shopbot vacuum pumps

I doubt you'd see much difference between them. They are both just vacuum cleaner motors and fans. Most guys using the Shopbot method use at least 2 of those motors, and up to 4 or them for 4x8 tables.

[MARV]

I looked for the vacuum unit you spoke of at Shopbot, but couldn't find it. Regarding shop vac blowers; many are single stage, meaning the air that is extracted thru the pick up hose also goes thru the motor to cool it. As the volume of air thru system is reduced the motor temperature goes up. In a 2 stage system, separate cooling air is sourced for the motor. I presume you already know this, but I thought I might restate the obvious. Keep us in the loop and best of luck with your vac system.

[canman77]

I've seen that number floating around quite a bit actually...4 of them on a 4x8. I have a 2x4 table, so I figure I'd only need one.

EDIT: MARV, here is a link to the shopbot vacuums. I don't know if they are actually affiliated with shopbot but apparently lots of shopbot owners use these:

https://www.centralvacuummotor.com/shopbot.htm

EDIT 2: I saw these pumps at Harbor Freight...I know, lots of people hate Harbor Freight but I've had good luck with their stuff. I was looking through the manual and didn't see any HG ratings... https://www.harborfreight.com/3-cfm-...ump-61176.html What do you guys think about this as a pump option?

[ger21]

EDIT 2: I saw these pumps at Harbor Freight...I know, lots of people hate Harbor Freight but I've had good luck with their stuff. I was looking through the manual and didn't see any HG ratings... https://www.harborfreight.com/3-cfm-...ump-61176.html What do you guys think about this as a pump option?

22 microns is about 29.92"Hg

These pump a lot of oil into the air.

have an oil less Gast pump like this, that pulls 26"Hg and 3CFM. https://www.ebay.com/itm/GAST-1-4HP-...cAAOSwKlxeVAZ4


With high vacuum low CFM pumps like these, you need to use pods or fixtures to hold your part, as there's not enough CFM to pull through MDF.

[lambodesigns]

I'm claiming victory on my 3'x4' vacuum table. Yesterday we cut 10 sheets of plywood into 14 cabinets in an easy 6 hours including breaks to solve all the world's problems as we were loading the next sheet Ended up the smallest stretcher was 6x18 and it was held down with no problem at all. The larger pieces were not going anywhere. No tabs at all but I did onion skin everything.... but that is more a reflection on my CNC than the vacuum table. The onion skin is more of a final pass to give a better edge than required to keep the part from moving. With the displays I build the exact dimensions aren't critical and I sand all the edges so I cut big parts without it.

We all use our CNCs for different reasons. Doing a lot of sheet goods mean you need cfm and with only pulling 5-6" of HG you can have very good results. I spent around $500 for the 4 vacuum motors I'm using and they continue to impress me as to how much holding power they have. They have more than paid for themselves many times over. Now one of the projects in my near future is cutting out chair legs from 8/4 stock using a long 1/2" bit. No way will the setup I have hold these parts down without a lot of help. A setup like Marv or Peteeng would be required to hold those parts down using vacuum alone. When I built my first chair it was before I had a vacuum table so I already have the clamps built to hold the parts down but I do have a good vacuum pump that I might use with some dedicated fixtures. Since I already have the parts in CAD it wouldn't take much to build a dedicated fixture for each part.

I will also say that every single time I tuned it on and the plywood was sucked down to the table it put a smile on my face. This was one of the worst pieces for warp but it still sucked down fine and all of the parts were held firmly in place.

https://www.youtube.com/watch?v=oO7dodYkPYI