Vacuum Meter Required for CNC Milling
A vacuum meter is required for CNC milling whenever you hold a part with differential pressure instead of clamps. The gauge is your only read on how much holding force you actually have. This page explains the mechanism, the pressure thresholds, and the geometries where vacuum workholding stops making sense.
Key takeaways
How a Vacuum Meter Sits Inside the CNC Milling Process
A vacuum meter required for CNC milling does one job: it reports the pressure difference between the sealed cavity under the workpiece and the shop atmosphere. That number is the only direct evidence of holding force. Everything else, pump model, hose diameter, table groove pattern, is just a path to that reading.
The mechanism is simple. Atmospheric pressure pushes down on the exposed top face of the part. Inside the sealed pocket the pump removes air, so the pressure there drops. The net force is the difference between the two, multiplied by the area the seal encloses. At sea level the atmosphere delivers roughly 101 kPa. Pull a gauge to −80 kPa and you recover about 80 percent of that.
So the meter is not monitoring the pump. It is monitoring the interface between part, seal, and table. A healthy pump with a torn seal strip reads the same as a weak pump with a perfect one. The gauge cannot tell you which. That is why the reading has to be paired with a leak test before the spindle starts.
In our shop the pressure gauge lives on the vacuum line, downstream of any filter or trap, as close to the table plenum as practical. Upstream placement hides pressure losses in the plumbing. When a reading drifts during a cut, we want to know whether the leak is at the part or back at the pump.
- 1Read downstreamMount the gauge after filters and traps so it sees true table pressure.
- 2Log the numberRecord the stable reading for each setup. Drift between runs is your early leak warning.
- 3One gauge per zoneMulti-zone tables need per-zone readings, not one shared line gauge.
Holding Force From Pressure and Sealed Area
Force equals pressure difference times sealed area, F = ΔP × A. The arithmetic is unforgiving on small parts. A 100 mm × 100 mm pocket at −80 kPa yields about 800 N, roughly 82 kgf, spread evenly. That is plenty for a light face-milling pass on aluminium. Shrink the pocket to 25 mm × 25 mm and the same −80 kPa gives only about 50 N. A 12 mm end mill at moderate feed will walk that part.
This is the first practical boundary. Vacuum workholding favours parts with a large continuous footprint. Thin plates, housings, and covers are natural candidates. Small brackets, pins, and narrow bosses are not, unless you can gang several of them under one seal and count the combined area.
The distribution matters as much as the total. Clamps load a part at a few points, which can bow thin stock between them. Vacuum loads the whole seated face. For a 3 mm aluminium cover, that difference decides whether the finished part is flat or sprung. On a 6061 plate at −80 kPa we routinely hold flatness inside ±0.005 mm after a light finishing pass.
Cutting force pushes back against all of this. A 16 mm carbide cutter in 7075 at 0.1 mm tooth feed can generate several hundred newtons of tangential load. The vacuum force needs a comfortable margin above that, and the margin should be checked at the worst point in the toolpath, not the average.
- 1Large footprint, easy callPlates and covers above 150 mm square hold well at −60 to −80 kPa.
- 2Small footprint, rethinkUnder 40 mm square, mechanical clamping is usually the safer route.
- 3Gang small partsGrouping parts under one seal recovers the area you lost.
When Vacuum Workholding Stops Working
Through-holes are the most common deal breaker. A hole that passes through the seated face connects the sealed pocket to open air. The pump then fights an infinite leak. You can still run the part if the hole lands outside the seal loop, or if you plug it with a sacrificial plug, or if the hole is drilled late in the process. Drilling it first and expecting vacuum to hold is a setup that will fail.
Ductile materials that smear are the second boundary. Aluminium and copper can push material into the seal groove and onto the seating face during heavy roughing. Once a chip sits under the seal, the reading drops and the part creeps. We keep the seated face clean and inspect it between operations. A 0.2 mm chip under a seal strip is enough to lose 20 kPa.
Thermal drift matters on long cycles. The part, the table, and the fixture all grow at different rates. A reading that starts at −80 kPa and settles at −72 kPa after an hour of roughing is telling you something. Often it is just seal compression, but it can also be a slow leak from a fitting. Either way, the number is the symptom and you should chase it.
Finally, vacuum workholding cannot replace rigidity. It holds the part against the table, but it does not stiffen the part itself. A tall, thin wall in a deep pocket will still chatter, regardless of how good the gauge reading looks. In that case the answer is a support structure, not more vacuum.
- 1Plan hole sequenceKeep through-holes outside the seal, or drill them in a later op.
- 2Watch the driftA slow drop over an hour points to seal compression or a fitting leak.
- 3Vacuum is not stiffnessThin walls need support, not a higher gauge reading.
Setting a Vacuum Table Before the First Cut
Five checks, in order. Do not skip the leak test.
- 1Clean the seating faceWipe the table and the part face. Any chip above 0.1 mm will lift the part and open a leak path.
- 2Lay the seal and check the loopSeal cord should sit in the groove without gaps at corners. Keep the enclosed area as large as the part allows.
- 3Pull vacuum with no partCap the zones and run the pump. A healthy table reaches −85 kPa or better and holds it with the valve closed.
- 4Seat the part and read the gaugeTarget −60 to −80 kPa for aluminium plates. Below −50 kPa, stop and find the leak before cutting.
- 5Leak-test for 60 secondsClose the valve and watch the gauge. A drop under 5 kPa per minute is acceptable for most milling.
- 6Re-check after the first passRoughing shifts the part slightly. Confirm the reading is still in range before the finishing pass.
Vacuum Workholding vs Mechanical Clamping
Match the method to the part, not to habit.
| Criterion | Vacuum workholding | Mechanical clamping |
|---|---|---|
| Part footprint | Above 150 mm square works well | Any size, down to a few millimetres |
| Seating face | Must be flat within 0.05 mm | Face can be rough or irregular |
| Thin plate risk | Low, load is spread evenly | High, point loads bow the part |
| Setup time | Slow on first run, fast on repeats | Predictable, similar every run |
| Tool access | Top face fully open | Clamps and straps block paths |
| Through-holes | Break the seal, needs planning | No effect on holding |
| Best fit | Plates, covers, housings | Brackets, shafts, small prismatic parts |
The Call We Make in the Shop
For flat plates, covers, and housings above 150 mm square, vacuum workholding with a live gauge is the better choice. For small brackets, parts with through-holes in the seating face, or anything under 40 mm square, clamp it mechanically and skip the vacuum setup entirely.
Common questions
What pressure should the gauge read before I start milling?
For aluminium plates on a sealed vacuum table, aim for −60 to −80 kPa. That range gives a comfortable margin over typical cutting forces on a large footprint.
Below −50 kPa the margin gets thin. Stop and find the leak rather than pushing the feed rate down to compensate.
Can I hold a part with a through-hole on a vacuum table?
Yes, if the hole stays outside the seal loop or if you plug it. A hole that crosses the sealed pocket turns the pocket into an open leak and the pump cannot recover.
A common workaround is to machine the through-hole in a second operation after the vacuum-held face work is complete.
Does a higher vacuum always mean better holding?
No. Holding force depends on pressure difference times sealed area. A small part at a perfect −85 kPa can still be weaker than a large part at −60 kPa.
Check the force number against the cutting load at the worst point in the toolpath, not the average load.
Why does my gauge reading drop during a long cut?
The usual causes are seal compression, a fitting leak, or thermal movement between the part and the table. Chips under the seal do the same thing.
Log the reading at the start and end of each operation. A repeatable drop points to a specific fitting or seal that needs attention.
Is vacuum workholding suitable for titanium or Inconel?
It can be, but the cutting forces in titanium and Inconel are much higher than in aluminium. The sealed area needs to be generous, and the pressure should sit at the top of the usable range.
For small titanium parts we usually clamp mechanically and reserve vacuum for large, flat sections.
How flat does the part face need to be for vacuum to seal?
A seating face flat within 0.05 mm will seal reliably with standard cord. Rougher faces need a softer seal material or a gasket sheet.
If the part face is bowed, the seal will touch on the high spots only and the gauge will show a weak, drifting reading.
Send Us Your Vacuum-Held Parts
Upload a STEP file and our engineers will confirm whether vacuum workholding fits your geometry before we quote.
12-hour quote100% inspectionNDA on request
More from GreatLight
We publish setup notes, tooling trials and inspection data from the factory floor.
