A contractor in Doha sent me a recording last spring. Twelve hundred by six hundred duct, 1.0 mm galvanized, running from an AHU plenum to a VAV box about nine metres away. At around 60 percent fan speed the whole run hummed, and the building's acoustician put a meter on it and found a peak at 63 Hz. The duct was not leaking. The joints were tight. It was the flat panel between the transverse joints flexing like a drum skin, and the fix was not more sealant. It was four beads per panel that should have been there from the start.
I have this conversation maybe six times a year. Beading is the cheapest stiffening operation in a duct shop, it takes about four seconds per panel on a rotary beader, and it is still the first thing people cut when a job gets tight. Then they spend the money on site visits and extra hangers instead.
A bead is a shallow transverse corrugation rolled into the flat of the duct panel, running parallel to the duct axis on the two large sides and, on larger duct, on the small sides too. It does not make the sheet thicker. It changes the section.
Take a 1200 mm wide panel of 1.0 mm galvanized. Unbeaded, the panel resists bending with a section modulus proportional to thickness squared — about 1.0 mm² per millimetre of width. Roll four beads at 3 mm depth and 50 mm pitch along the edge region and you have effectively introduced four short stiffeners. In practice the improvement in resistance to the oil-canning mode — the one that drums — is 35 to 50 percent. Not a factor of ten. Half again. But half again is usually the difference between a duct that talks and one that does not.
Where beads do much more work is in resisting the diaphragm action of the panel under pressure. A rectangular duct at 800 Pa positive pressure tries to turn its flat sides into a barrel. The transverse joints hold the corners; the panel between them has to carry the load in bending. Beads break that span into shorter spans, and deflection goes with the fourth power of span. Halving the effective span does not halve the deflection, it cuts it by roughly sixteen times. That is why 50 mm pitch feels so much stiffer than 100 mm pitch even though the bead depth is the same.
Three parameters, and shops argue about all three. Here is what I specify, and why.
| Parameter | Light duty (0.5-0.7 mm) | Standard (0.8-1.0 mm) | Heavy (1.2-1.5 mm) |
|---|---|---|---|
| Bead depth | 2.0-2.5 mm | 3.0-3.5 mm | 4.0-5.0 mm |
| Bead pitch (centre to centre) | 60-75 mm | 45-60 mm | 40-50 mm |
| First bead from lock seam | 35 mm min | 40 mm min | 50 mm min |
| Beads per 1200 mm panel | 3 | 4 | 5 |
| Bead profile radius | 1.5 mm | 2.0 mm | 2.5 mm |
| Typical deflection reduction | 30% | 40% | 50% |
| Roll pressure setting (guide) | 2.5-3.0 bar | 3.5-4.0 bar | 4.5-5.5 bar |
| Feed speed through beader | 12 m/min | 10 m/min | 8 m/min |
Depth is limited by two things: the coating and the corner. Push a bead past about 5 mm on 1.0 mm galvanized and you will see the zinc crack along the bead crest within a few months, usually starting at the ends where the roll enters and exits. That is a corrosion site, and on an outdoor or coastal job it is the first place to rust. The other limit is the corner: on a 200 x 200 duct you cannot have a 5 mm bead 30 mm from the corner, because the bend radius eats the clearance and the bead deforms when the panel is folded.
Pitch is limited by appearance and by airflow. Tight pitch stiffens more but adds surface roughness inside the duct. On a low-velocity system nobody cares. On a 12 m/s supply run in a hospital, a bead every 40 mm across 1200 mm of panel adds measurable pressure drop and, worse, it gives the airstream something to sing against. I keep 50 mm as a floor for anything in an occupied space and go to 60 mm on acoustic jobs.
This is the mistake I see most often. The bead nearest the Pittsburgh lock needs clearance or the lock will not close cleanly and the corner will fight you at assembly. Forty millimetres from the seam is my rule on 0.8 to 1.0 mm. Some shops use 25 mm to get an extra bead in; those are the shops that call me about lock seams opening at the corners.
A rotary beader is a simple machine that rewards attention. Two or three rolls, a gap adjustment, a pressure regulator, and a feed motor. The bead quality comes almost entirely from three adjustments.
Roll gap. Set the gap to material thickness plus 0.05 mm. On 0.8 mm sheet that is 0.85 mm. Too tight and you thin the material at the bead root, which is where cracking starts. Too loose and the bead depth wanders along the length — you get 3 mm at the leading edge and 2.2 mm at the trailing edge, and the panel drums at one end.
Roll alignment. The upper and lower rolls must be parallel within 0.02 mm across the face. A 0.05 mm taper shows up as a bead that is deeper on the operator side. Check with a feeler gauge at both ends of the roll face, not just the middle.
Pressure and feed. Feed too fast and the material work-hardens at the bead entrance, then springs back less and the bead comes out shallower than the setting. On 1.2 mm I run 8 m/min and accept the slower cycle. On 0.6 mm, 12 m/min is fine.
Shops that run 0.6 and 1.0 mm on the same day need a setup card, not operator memory. Going from 0.6 to 1.0 mm means opening the gap by 0.4 mm and adding roughly 0.8 bar of pressure, and the bead depth will still come out 0.2 mm shallower because of the different springback. Without a card, the second job in the afternoon gets the first job's settings, and the panels come out with beads that are visually there but structurally half strength. Nobody notices until the duct drums on site.
Beading is not the only option and it is not always the right one.
Cross-breaking — a single large V-groove across the panel — is faster (one hit on the press brake) and stiffer per operation, but it only helps in one direction and it looks like a crease. Fine for plenums and plant rooms, not for exposed ductwork.
External stiffeners — angle or channel screwed or riveted across the panel — are what you do when the duct is already made and already drumming. Effective, ugly, and they add a thermal bridge and a cleaning problem.
Tighter hanger spacing is the answer people reach for first and it is usually the wrong one. Hangers support the duct; they do not stiffen the panel between transverse joints. Adding a hanger in the middle of a 1.5 m joint spacing does nothing for the 63 Hz mode.
Thicker material works, obviously, and costs more than beading by a wide margin. Going from 0.8 to 1.0 mm on a 1200 mm panel adds about 25 percent to the steel cost of that panel. Beading adds four seconds of machine time.
Zinc flaking along the bead crest. Depth too great for the coating weight, or rolls worn with a sharp entry radius. Replace rolls at 0.1 mm wear, or re-radius them. On Z275 coating you have more tolerance than on Z120.
Bead fading along the length. Roll gap opening under load, or a hydraulic/air pressure droop. Check the regulator and the roll locking screws. On a beader that has run five years without adjustment, the locking screws are usually the culprit.
Cracking at the bead ends. Shear burr on the sheet edge, or the bead running too close to the edge. Deburr, and keep 40 mm clearance.
Oil canning between beads. Pitch too wide for the panel width, or beads too shallow. It is a stiffness problem, not a flatness problem — no amount of levelling will fix it.
Marks on prepainted or stainless. Roll pickup. Clean the rolls with solvent every shift, and on stainless use a beader with polished rolls and a plastic film if the finish is visible.
Not every duct should be beaded. Skip it on:
Three checks, sixty seconds, at the end of the line.
Depth. A cheap digital caliper with a bead-depth attachment, or a purpose-made go/no-go gauge. Measure at both ends and the middle of three panels per batch. Spread should be under 0.3 mm.
Pitch. Tape measure across ten beads, divide by ten. If the nominal is 50 mm and you measure 53, the feed rolls are slipping.
Coating. Bend a scrap sample over a 25 mm mandrel at the bead. If the zinc or paint crazes, the depth is too high for the material and you should back off 0.3 mm before running the production batch.
The deflection check, if you want to be thorough, is a simple load test: support a 1200 mm panel on two edges, put 20 kg in the centre, and measure. Compare beaded against unbeaded from the same coil. If the difference is under 25 percent, your beads are too shallow to be doing anything, and you are paying for the operation without getting the stiffness.
The duct beading machine page lists roll configurations and the thickness ranges each one handles, and if the beader is going to live behind an automatic line, the rotary beader is the version that takes the sheet directly off the line without a second handling step. Both are worth a look before you decide to keep cross-breaking panels on the press brake instead. For shops running the full sequence, the beading station is normally specified as part of the Line 5 configuration rather than bought separately, because the transfer height has to match.
Four at 50 mm pitch for 0.8 to 1.0 mm material, positioned so the first bead clears the lock seam by at least 40 mm. On 0.6 mm go to five at 45 mm pitch, because the thinner sheet has less inherent stiffness and the bead is doing more of the work.
No. A rotary beader needs a flat panel with free edges. Once the duct is folded and locked, the only options are external stiffeners or replacing the section. This is why beading has to be a line decision, not a site fix.
Not measurably on a well-set machine. The bead is a forming operation, not a cut, and the material stays continuous. Leakage problems attributed to beading are almost always seam or joint problems that were there anyway.
Less than on galvanized. Stainless work-hardens fast, so a 3 mm bead on 304 at 0.8 mm will crack at the ends within a few hundred cycles if the roll radius is tight. Use 2.0 to 2.5 mm depth with a 2.5 mm roll radius, and slow the feed to 6 m/min.