The pump is not holding the part
This is the thing worth getting straight before anything else. A vacuum fixture works because the atmosphere is already pressing on everything at about 14.7 pounds per square inch, and removing some of the air underneath the part means that pressure is no longer balanced. The pump does not pull. It only takes air out, and the atmosphere does the work.
The consequence is a hard ceiling. Twenty inches of mercury is 9.82 psi, which on 288 square inches of sealed area at 85 percent effectiveness gives 2,405 pounds. A perfect vacuum on the same area would give 3,598 pounds and no pump reaches it. Doubling the pump size cannot get you past that number, because there is no more atmosphere to work with. What a bigger pump does buy is the ability to hold a given vacuum against more leakage, which on porous material is exactly the problem worth solving — but it is a flow problem, not a pressure one.
Two numbers, and one of them is usually guessed
The area is easy: measure inside the gasket. The vacuum level is where estimates go astray, because there are three different readings you might take and only one of them is relevant. At the pump with the port blocked, you see the best the pump can do. At the table with nothing on it, you see the leakage of the table itself. At the table with the real part sitting on the real gasket, with the machine running, you see the number that belongs on this page. On a warped sheet or a hard old gasket the third number can be half the first.
| Gauge reading | psi of differential | On 288 sq in at 85 percent sealing |
|---|---|---|
| 10 inHg | 4.91 psi | 1,202 lb |
| 15 inHg | 7.37 psi | 1,804 lb |
| 20 inHg | 9.82 psi | 2,405 lb |
| 25 inHg | 12.28 psi | 3,006 lb |
| 29.92 inHg | 14.70 psi | 3,597 lb, and unreachable |
Force down is not the same as force sideways
Almost nothing in routing tries to lift a part straight off the table. The cutter pushes it sideways, and what resists that is friction, which is the hold-down force multiplied by a coefficient that depends on the two surfaces. That coefficient is not something anyone can hand you: it depends on the table material, the part material, the surface finish of both, and whether there is dust between them. The way to get it is to tilt a scrap on the same surface until it slides and take the tangent of the angle, and to repeat it a few times because the spread is wide.
Even with a good coefficient, sideways resistance is not the whole story. A tall narrow part can tip about its leading edge before it slides. A cut that opens along one side of a part destroys the seal on that side while the cutter is still working on it. Small parts lose their grip disproportionately, because the seal length around them is short relative to the force the tool applies. This is exactly why tabs and vacuum are usually used together rather than as alternatives.
What this page will not tell you
It will not tell you whether your setup is adequate. It converts a gauge reading and an area into pounds, and if you enter a requirement of your own it prints that beside the result without comment. The reason is that adequacy depends on the cutting forces at your feed, depth and tool engagement, on the part geometry, on where the seal fails first as the cut progresses, and on what the machine and pump manufacturers say about the equipment. None of that is arithmetic.
Where this sits
Tabs are the other half of holding a part down, and the tab holding and cleanup calculator works out how many and what they cost afterwards. The cut that is trying to move the part is timed on the CNC router job time calculator, and how hard it pushes has a lot to do with the depth per pass there. Shop air for pneumatic clamping instead is on the air compressor CFM calculator, and the extraction that runs alongside a router is on the dust collection calculator.
Questions people ask
Why does the calculation use gauge vacuum rather than absolute pressure?
Because the force comes from the difference between the two sides of the part, and the top side is at atmospheric pressure. A reading of 20 inHg of vacuum means the pressure underneath is 20 inches of mercury below atmospheric, and that difference is what presses the part down. If your gauge reads absolute pressure instead, subtract it from atmospheric before entering it: a gauge reading 9.92 inHg absolute at sea level is 20 inHg of vacuum. Getting this backwards is a common way to end up with a number three times too large or too small.
Does altitude change the answer?
Yes, and by more than people expect. The ceiling on hold-down is local atmospheric pressure, which falls with elevation. At around 5,000 feet the atmosphere is near 12.2 psi rather than 14.7, so the maximum possible force on any area drops by about 17 percent, and a pump that pulls 25 inHg at sea level cannot pull it at all up there because 25 inHg is above the local atmosphere. This page uses the sea level figure. If you work at elevation, the vacuum you can actually measure at your table already reflects the real limit, which is another argument for entering a measured number rather than a nominal one.
How do I estimate the share of the area that is actually sealing?
Start from what you can see. A gasket groove that runs off the edge of the material is not sealing past that point, so subtract that region. A warped sheet lifts in the middle and the seal only closes where it touches. A part smaller than the pod leaves gasket doing nothing. Beyond the visible, the honest check is the gauge itself: put the real part down and see how much vacuum you keep. A setup that holds most of its no-load reading is sealing well, and one that collapses is not, whatever the area suggests. If you have no basis at all, running the figure at 100 percent and treating the result as an upper bound is more useful than guessing a number in the middle.
Why is a small part so much harder to hold than a large one?
Because the force scales with area while the cutting force scales with how much tool is engaged, which does not shrink with the part. A part four inches square has sixteen square inches of hold-down. The same cutter at the same depth pushes just as hard on it as it does on a part twenty times the size. On top of that, the seal around a small part is short, so a cut that opens along one edge removes a large fraction of it at once. This is the region where vacuum alone stops being the answer and tabs, a sacrificial skin or an adhesive method take over.
Can I add the pods together like this if they are on separate zones?
Only if they are all actually pulling the vacuum you entered at the same time, which is the assumption in the multiplication. Separate zones on a shared pump do not necessarily hold the same level, because a leaking zone drags the whole manifold down. If one zone is under a warped part and another is under a flat one, they are at different vacuums and the total is not the pod force times the pod count. In that case work each zone out separately with its own measured reading and add the results by hand, which will be lower than the single-figure calculation and will be closer to the truth.