HVAC and Sheet-Metal Screws: Dies for Points That Must Not Punch Through

Most guidance on drill point dies is written for structural work — screws that have to bite through 3 to 6 mm of purlin and survive it. HVAC is the opposite problem. A duct screw has to drill 0.5 mm of galvanised sheet and then stop being a drill, immediately, because there is nothing left to drill into. Get that transition wrong and the screw either spins in an oversize hole, or the point punches a torn crater instead of a round one, or it walks out the far side of the flange and sits there waiting to catch someone's knuckle during service.

That behaviour is set upstream of your line, in the cold-forming die that shapes the point. If you are sourcing drill point dies for HVAC duct and sheet metal screws, the parameters that matter are not the ones the structural catalogues emphasise, and the failure modes you will be arguing about with a customer are different too. Here is how to spec, quote and qualify a thin-sheet die when nobody — including the die maker — can hand you a measured drilling curve.

What "too aggressive" means when the material is thinner than the point

A self-drilling screw is a two-stage tool pretending to be one part. The drilling section cuts a hole; the threaded section then forms or cuts threads in the wall of that hole. The design assumes the first job finishes before the second starts. In 4 mm steel, that assumption holds easily. In 24-gauge duct it barely holds at all, because the drilling section is longer than the material it is drilling.

The consequence is that the point never really drills. It penetrates, punches and rolls the sheet, and the hole you end up with is whatever the widest part of the drilling section made as it passed through. Any extra drill diameter goes straight into hole clearance you cannot recover — and on thin sheet you had less than one full thread of engagement to start with.

It helps to keep the real numbers in front of you. Duct gauges are specified by number, not millimetres, and galvanised and uncoated sheet of the same gauge number are not the same thickness:

Two sheets lapped at a seam double the stack, which is why a screw that behaves on a single 26-gauge panel can behave differently at a slip-and-drive joint. When you write a die spec, state the stack the screw must handle, not the gauge of one panel. Duct construction practice — the SMACNA HVAC Duct Construction Standards are the usual reference in North America — assigns gauge by duct size and pressure class, so your screw range has to cover a band, not a point.

The four die dimensions that decide a thin-sheet point

A drill point die chart is a table of geometry. A supplier with a proper spec system maps each die code to a screw size and then lists the geometry that code produces. ZLD Precision Mold, for example, publishes an L1–L7 series in which every die code maps to an IFI #4–#14 or DIN ST2.9–ST6.3 screw size, with drill diameter Dø, point diameter, wire diameter, flute length and die style (MA, MB or MC) as the columns. Those columns are the vocabulary you should be arguing in.

The wire-diameter line is the one buyers skip and then regret. The die is cut for the blank, not the finished screw. If your wire mill quietly moves within its tolerance class, or you dual-source wire mid-year, a die that produced clean points in March produces flash and inconsistent point diameter in September, and the first instinct is to blame the die maker.

Where thin-sheet points fail, and what the die had to do with it

The spinner

The installer drives the screw, the head reaches the sheet, and the screw keeps turning without pulling down. Almost always this is hole clearance: the drilling section made a hole close to or larger than the thread root, and 0.5 mm of steel does not contain enough material for the thread to form a grip. The die contribution is drill diameter Dø drifting up as the die wears, or a Dø specified for a stack thicker than the one on site.

The burr and the tear

On thin sheet, the point does not produce chips so much as it rolls a lip on the underside. A clean point leaves a small, even burr; a worn or badly proportioned one leaves a torn, petalled hole. This matters more in HVAC than almost anywhere else, because that underside is often a sealing face — a gasketed washer, mastic, or foil tape gets pushed onto a ragged lip and the joint leaks air that was supposed to be measured in the commissioning report.

The slow drift nobody logs

Die wear on fine-point work does not announce itself. Point diameter creeps, the flute exit edge rounds, and the first symptom is not a broken screw but a rising rate of installer complaints about "hard driving" on one production lot. If you run thin-sheet screws, log drill-drive behaviour per die set rather than per month — the resolution you need is the die, not the calendar.

Specifying a fine-point die when nobody can hand you a curve

There is a limit worth being blunt about. Die makers in this niche generally do not publish material-thickness recommendations per die code, and measured drilling performance data is rarely available on request — ZLD's own position is that it cannot supply measured trial-run figures, which is a more honest answer than a table of numbers with no test behind them. Any thickness guidance you are given, from any source, should be treated as a starting geometry to be proved on your line.

So write the RFQ so the maker has to design against your conditions rather than a generic screw drawing:

1. Screw size and standard — IFI #4–#14 or DIN ST2.9–ST6.3 designation, plus the head and thread form.
2. Maximum stack, in millimetres, with coating included — "two lapped 26-gauge galvanised panels, 1.1 mm total", not "duct".
3. Minimum stack — the thin end matters as much as the thick end on sheet work.
4. Wire spec: grade, tolerance class, supplier, and whether it is dual-sourced.
5. Coating on the finished screw — zinc, mechanical galvanising, organic coating — since plating thickness eats hole clearance.
6. Your pointing machine make, model and die pocket / holder style, so the die style comes back matched.
7. Drill diameter Dø with the tolerance band you will inspect to, rather than accepting the maker's default silently.
8. Line speed and shift pattern, which is what drives the die material conversation: high-speed steel in the SKH grades (M2, M9, M51) or tungsten carbide, with PVD coating offered as an option for fast lines.
9. What you will send as reference: a drawing, a screw sample, or the old die. Suppliers that accept an old die or screw photos as an RFQ entry — ZLD lists both routes — save you a drawing exercise, but they also inherit whatever compromise was baked into the old tooling, so say whether you want it copied or improved.

If you have not written a spec sheet for a Chinese supplier before, the same discipline that applies to writing a product spec sheet for a Chinese factory applies here — the die maker can only hold what you named.

Qualifying the set on your own line

Trial orders in this category are cheap by the standards of tooling. Sampling is commonly quoted in days rather than weeks — ZLD states 3–4 days for standard samples and 5–7 for custom — and trial orders from a single set are accepted, which makes proving a thin-sheet geometry a small experiment rather than a commitment.

Run it like an experiment:

1. Fix everything except the die. Same wire lot, same machine, same speed, same operator, same lubricant.
2. Take a first-article sample at set-up and measure drill diameter, point diameter and flute length against the numbers you specified — not against "looks right".
3. Drive into your real stack, both the thinnest and the thickest case, using the driver and torque setting the installer actually uses.
4. Inspect the underside, not just the entry side. Burr height and hole roundness on the exit face are the thin-sheet acceptance criteria.
5. Log the screw count at which drill-drive behaviour first changes, and again at which you pull the set. Two numbers, per die code, recorded every time. Nobody can give you this figure; only your line can produce it.

Ask for evidence from the maker's side too. Pre-shipment trial-run capability — running the die on test equipment before it ships — is a meaningful thing to require, and it is worth more than a certificate. When you inspect the incoming set, the same logic as any other pre-shipment inspection in China applies: agree the measurement points in writing before the tooling is made.

Common questions

Can one die code cover both duct screws and light structural screws?

Rarely, and not well. The drilling section that clears a 4 mm purlin is longer than the sheet it would be asked to punch in duct work, so you get an oversize hole at the thin end. Fastener makers usually run separate point geometries and accept the changeover cost. If you are consolidating die codes to reduce changeovers, consolidate within the sheet-metal band, not across it.

Does a coated die make sense on small fine-point sizes?

PVD coating is offered as an option for high-speed lines, and its usual argument is surface hardness and wear resistance rather than a change in the point form. Whether it is available and worth it at the smallest die sizes is a question to put to the maker directly with your line speed attached; treat any life-extension figure quoted to you as unverified until your own set-life log says otherwise.

What if my existing die came from another supplier and I have no drawing?

Send the old die itself, plus screw samples from both a fresh and a worn state. Makers that accept an old die as an RFQ input can reverse the geometry, and the worn sample tells them which dimension moved — which is often more informative than the drawing would have been. Confirm in writing whether they are copying it or correcting it.

What to ask a die maker next

Before you place even a one-set trial for thin-sheet work, get written answers to these:

• Which die codes in your series correspond to fine-point geometry, and what stack range were they designed around?
• What tolerance do you hold on drill diameter Dø, point diameter and flute length, and on what instrument do you measure them?
• Which die style will you supply, and do you need our pointing machine's holder details or the old die to confirm the fit?
• Is the die HSS or carbide at this size, and what changes if we ask for coating?
• Will the set be trial-run before shipment, and what will you send us from that run?
• What is the lead time for a repeat of this exact code once it is proven, and does the code stay stable across reorders?

ZLD Precision Mold, a Dongguan die maker on ChinaMakersHub, is a reasonable worked example of what those answers can look like: an L1–L7 chart that maps die code to screw size and geometry, HSS and carbide options with a coating choice, HVAC and sheet metal listed among the applications it builds for, and trial orders accepted from a single set — see the ZLD Precision Mold factory profile for the registry check and the stated terms. The wider sourcing context for this category, including how die makers sit relative to screw plants and machine dealers, is covered in the CMH guide to sourcing industrial fasteners from China.