Self-Clinching Nut Pushout and Torque-Out Figures: Minimum, Average or Unstated? What PEM, PSM and NP Fasteners Say

Why no published figure can be read as a floor

A pushout or torque-out figure printed in a catalogue looks like a limit, but nothing in this set of catalogues makes it one. PSM states its values are averages when correct installation is performed, and warns that hole size, material and installation will affect the results ¹. NP Fasteners goes further: all performance figures are averages obtained over a range of installations and should be used as an indicator only ⁶. If you read either table as a guaranteed minimum, you are claiming a floor the publisher has not offered. The stated basis is an average, so a part that installs slightly off the tabulated conditions can sit below the printed number. PEM's excerpt shows performance tables and detailed pushout and torque-out test methods but states no average, minimum or design-limit basis ¹⁰. That silence is not a guarantee in either direction — it leaves the basis unknown, not high.

What each publisher actually says about its own numbers

Read the qualifying note before the table, because it decides what the table is worth to you. PSM's note ties its averages to correct installation and points buyers to its technology centre for specific advice ¹. NP's note names the variables — panel material, hole preparation, installation tooling and method — and recommends testing in the actual application ⁶. PEM's excerpt is the outlier. It gives the test method in detail, including pushout from the grip side and torque-out from the shoulder side with a hardened test screw. But the excerpt carries no statement of what the resulting figures represent ¹⁰. A Captive-fastener page describes a single stud value as "approximately" 105 lbs pushout and 15.8 in-lbs torque-out, again with no stated basis ². So the classification is simple: PSM and NP are average/indicator publishers by their own words; PEM and Captive-fastener are unstated in these excerpts.

Handling a publisher that states no basis

When a catalogue gives you a number and no basis, the safe default is the status PSM and NP assign explicitly: indicator only. Until the publisher confirms otherwise in writing, use its figures to screen candidate sizes, not to size a joint.

• Ask the publisher in writing whether the table is average, minimum or design limit, and keep the reply with the drawing.
• Treat an unstated-basis figure as screening-only until that written answer arrives ¹⁰.
• Ask for the test report behind the table: sheet material, thickness, hole size and installation force.
• Request sample parts and run your own pushout and torque-out tests in your panel, as NP recommends ⁶.
• Do not carry an unstated-basis number into a design calculation as a floor ^{10, 2}.

Why figures from different publishers are not like-for-like

Overlapping thread sizes make the tables look comparable, but the test conditions behind them differ. PSM tabulates M3 to M6 on 304 stainless with separate values per shank code ¹. NP tabulates M3 to M10 on 304 stainless and M3 to M4 on aluminium ⁶. PEM's SMPS and SMPP nuts are tabulated on cold-rolled steel, with a 304 stainless condition at 0.7 mm and HRB 89 ¹⁰. Set an M3 against an M3 and the mismatch shows. PSM lists 600 N pushout and 1.7 Nm torque-out for shank code 0 on 304 stainless ¹; NP lists a 720–1300 N and 1.9–7.0 Nm range for M3 on 304 stainless ⁶. Different part series, shank codes, sheet thicknesses and material conditions sit behind those numbers, so the gap cannot be read as one supplier outperforming another. Before you compare any two values, match thread size, shank code, sheet material and thickness, and installation method.

The installation conditions a published figure depends on

Every publisher in this set makes performance conditional on installation, so a figure only holds where those conditions are met. PEM's installation guidance asks for the specified mounting hole size, a sharp hole punch, squeezing force between parallel surfaces, and full embedding of the clinching ring around the circumference ⁵. It also warns against over-squeezing, which crushes the head and distorts threads, and advises determining optimum installation force by test before production runs ⁵. NP sets panel hardness limits by fastener material: 80 HRB for steel, 70 HRB for 300-series stainless, 90 HRB for 400-series stainless and 50 HRB for aluminium ⁷. Its installation steps call for a correctly sized hole with no deburring or chamfering, the fastener installed from the punch side, and a parallel squeezing force until the head seats ³. Check your own panel against these conditions before you trust any tabulated value. If your sheet is harder, your hole preparation differs, or your tooling is not the recommended type, the published number describes someone else's joint.

Where the publishers diverge, and what resolves it

The dispute is not over the numbers themselves but over what they represent. PSM and NP state an average basis; PEM and Captive-fastener state none in these excerpts. The table sets the two positions side by side so you can see which catalogues need a written answer before use.

FAQ

Related pages

• Nuts product page
• Proof Load vs Yield and Tensile Strength in Bolt Selection: What the Evidence Settles
• Reading a Fastener Datasheet: Which Fields You Can Verify, and Which You Must Demand
• Blind and Self-Locking Fasteners for Joints With No Back Access: Rivet Nuts, Clinch Nuts and Self-Clinching Hardware Compared
• Bolt Head Types and Drive Styles: What the Evidence Can Settle Before You Specify