Socket Head Cap Screw Dimensions Calculator
Socket Head Cap Screw Dimensions Calculator
Every ISO 4762 / DIN 912 dimension, with the head height made the headline because it equals the thread diameter exactly — plus the stress area computed from thread geometry, proof and clamp loads for four classes, and the one column four sources dispute.
Socket head cap screw dimensions
An M12 socket head cap screw in property class 12.9, tightened to 65 per cent of proof, in a normal-fit hole
k = d, and everything else computed from the thread’s own geometry
- k = d
- the head height equals the nominal thread diameter, exactly, at every size in ISO 4762. Five published tables agree at all eleven sizes, and Machinery’s Handbook’s counterbore chart independently gives the counterbore depth as the nominal diameter from M1.6 to M24. It is a property of the head form and not a rounding coincidence, and it is the single most useful thing to carry away: you can size a counterbore for any socket head cap screw from the M number alone
- b = 2d + 12
- the standard’s thread length for the partially threaded form, in millimetres, exact at all eleven sizes up to M24. A rule, not a table. It was checked only over those eleven sizes, which is as far as the table read for this page goes, so do not extrapolate it above M24 or to a length regime the standard treats differently
- d₂, d₃
- the basic pitch diameter and the minor diameter at the rounded root. With H = P√3/2 the fundamental triangle height, d₂ = d − 3H/4 and d₃ = d − 5H/4 − H/6, both diameters. These are geometry, not table entries, which is why this page can compute a stress area without reproducing a copyrighted column
- As
- the nominal tensile stress area, ISO 898-1’s own definition: the area of a circle whose diameter is the mean of d₂ and d₃. Computed here and checked against the standard’s published column at every size, worst deviation 0.4 per cent at M14 where the published figure is itself rounded. Note that the UNIFIED definition for an inch thread is different and gives 1 to 2 per cent less area, so inch and metric stress areas are not quite the same quantity
- Rp
- the class’s proof stress. For a steel class the designation gives it: 12.9 means a nominal tensile of 1200 MPa and a proof stress 0.9 of that, so 1080 MPa. A2-70 stainless to ISO 3506-1 is 450 MPa against a 700 MPa tensile, which is 41.7 per cent of 12.9’s proof stress
- t
- the key engagement, the depth of the hexagon socket. The one column four published tables will not agree on: three give about 0.5 d and one gives about 0.55 d. Both are printed on this page with their sources. It is NOT the counterbore depth, which is k
- 0.20
- the nut factor in the torque row, T = K·F·d. It is a lumped empirical coefficient for a plain, lightly oiled thread and it varies by a factor of two with plating, lubricant and surface finish, which is why the torque figure here is an order-of-magnitude check and not a specification
Worked example
An M12 socket head cap screw in property class 12.9, tightened to 65 per cent of proof, in a normal-fit hole
THE HEAD IS 12 mm TALL, because the screw is M12. That is the headline and it is not a coincidence of this size: k = d at every size in ISO 4762, M3 through M24, in all five published tables read for this page. The practical use is immediate — a counterbore to bring the head flush is 12.00 mm deep, and Machinery's Handbook's counterbore chart agrees, giving the depth as the nominal diameter at every size from M1.6 to M24
THE HEAD IS 18 mm ACROSS, which is 1.50 d. That ratio is not constant: it is 1.83 at M3 and falls through 1.75, 1.70, 1.67, 1.63 and 1.60 before settling at exactly 1.50 at M12 and staying there for every larger size. The hexagon socket is 10 mm, which is 0.833 d
THE STRESS AREA IS COMPUTED, NOT COPIED. With P = 1.75 mm and H = P√3/2 = 1.5155 mm: d₂ = 12 − 3H/4 = 10.8633, d₁ = 12 − 5H/4 = 10.1056, and d₃ = d₁ − H/6 = 9.8530 mm. The mean of d₂ and d₃ is 10.3582, so As = π/4 × 10.3582² = 84.27 mm². ISO 898-1's published table says 84.3, so the derivation is right to 0.04 per cent and no copyrighted column had to be reproduced
PROOF LOAD AND CLAMP LOAD. Class 12.9 has a proof stress of 1080 MPa, so the proof load is 84.27 × 1080 = 91.01 kN. At 65 per cent of that the clamp load is 59.16 kN, which is the number the joint is actually designed around. At a nut factor of 0.20 that needs about 142 N·m of torque — an order-of-magnitude figure, because the nut factor swings by two to one with plating and lubricant
AND IF SOMEBODY SUBSTITUTES STAINLESS. The same M12 in A2-70 has a proof stress of 450 MPa, so the proof load is 37.92 kN: 41.7 per cent of the 12.9 figure, and the same ratio at every size. Socket head cap screws are predominantly sold in 12.9 and A2-70 looks identical in the bin. Swapping one in for corrosion reasons removes 58 per cent of the available clamp load
THE HOLE, AND WHERE THE HEAD PRESSES. A normal-fit clearance hole is 12.80 mm and the counterbore is 19.25 mm, which leaves only 0.625 mm of radial gap around an 18 mm head — cut the hole and the counterbore from one setup. The head then bears on an annulus between 18 and 12.80 mm, which is 125.8 mm², so 59.16 kN of clamp load puts 470 MPa under the head. That is above the yield of structural steel and far above aluminium's, which is why a cap screw embosses a ring into a soft joint face and why a hardened washer is not optional there
THE ONE NUMBER NOT TO TRUST. Key engagement t. Three sources say 6.0 mm for an M12 and one says 6.6 — a 10 per cent spread, both printed on this page. It matters if you are facing a head down or checking whether a worn key still has engagement; it does NOT matter for the counterbore, which is set by k = d and is not in dispute
ISO 4762 / DIN 912: the columns every source agrees on
| Size | Pitch (mm) | Head dia d_k max (mm) | Head height k (mm) | k / d | d_k / d | Hex socket s (mm) | Wall under socket w (mm) | Thread length b (mm) | 2d + 12 (mm) | Tensile stress area (sq mm) |
|---|---|---|---|---|---|---|---|---|---|---|
| M3 | 0.50 | 5.50 | 3.00 | 1.0000 | 1.833 | 2.5 | 1.15 | 18 | 18 | 5.03 |
| M4 | 0.70 | 7.00 | 4.00 | 1.0000 | 1.750 | 3.0 | 1.40 | 20 | 20 | 8.78 |
| M5 | 0.80 | 8.50 | 5.00 | 1.0000 | 1.700 | 4.0 | 1.90 | 22 | 22 | 14.18 |
| M6 | 1.00 | 10.00 | 6.00 | 1.0000 | 1.667 | 5.0 | 2.30 | 24 | 24 | 20.12 |
| M8 | 1.25 | 13.00 | 8.00 | 1.0000 | 1.625 | 6.0 | 3.30 | 28 | 28 | 36.61 |
| M10 | 1.50 | 16.00 | 10.00 | 1.0000 | 1.600 | 8.0 | 4.00 | 32 | 32 | 57.99 |
| M12 | 1.75 | 18.00 | 12.00 | 1.0000 | 1.500 | 10.0 | 4.80 | 36 | 36 | 84.27 |
| M14 | 2.00 | 21.00 | 14.00 | 1.0000 | 1.500 | 12.0 | 5.80 | 40 | 40 | 115.44 |
| M16 | 2.00 | 24.00 | 16.00 | 1.0000 | 1.500 | 14.0 | 6.80 | 44 | 44 | 156.67 |
| M20 | 2.50 | 30.00 | 20.00 | 1.0000 | 1.500 | 17.0 | 8.60 | 52 | 52 | 244.79 |
| M24 | 3.00 | 36.00 | 24.00 | 1.0000 | 1.500 | 19.0 | 10.40 | 60 | 60 | 352.50 |
Key engagement t: four readings, and they do not agree
| Size | t — RoyMech | t — SSR Precision | t — fasten.it (DIN EN ISO 4762) | t — GlobalFastener (ISO 4762:2004) | Difference (mm) | Difference (%) | t / d, majority | t / d, RoyMech |
|---|---|---|---|---|---|---|---|---|
| M3 | 1.3 | 1.3 | 1.3 | 1.5 | 0.0 | 0.0 | 0.433 | 0.433 |
| M4 | 2.0 | 2.0 | 2.0 | 2.0 | 0.0 | 0.0 | 0.500 | 0.500 |
| M5 | 2.7 | 2.5 | 2.5 | 2.5 | 0.2 | 8.0 | 0.500 | 0.540 |
| M6 | 3.3 | 3.0 | 3.0 | 3.0 | 0.3 | 10.0 | 0.500 | 0.550 |
| M8 | 4.3 | 4.0 | 4.0 | 4.0 | 0.3 | 7.5 | 0.500 | 0.537 |
| M10 | 5.5 | 5.0 | 5.0 | 5.0 | 0.5 | 10.0 | 0.500 | 0.550 |
| M12 | 6.6 | 6.0 | 6.0 | 6.0 | 0.6 | 10.0 | 0.500 | 0.550 |
| M14 | — | — | 7.0 | — | — | — | 0.500 | — |
| M16 | 8.8 | 8.0 | 8.0 | 8.0 | 0.8 | 10.0 | 0.500 | 0.550 |
| M20 | 10.7 | 10.0 | 10.0 | 10.0 | 0.7 | 7.0 | 0.500 | 0.535 |
| M24 | 12.9 | 12.0 | 12.0 | 12.0 | 0.9 | 7.5 | 0.500 | 0.537 |
Proof load by property class, and what stainless costs you
| Size | Stress area (sq mm) | Proof load 8.8 (kN) | 10.9 (kN) | 12.9 (kN) | A2-70 (kN) | 12.9 clamped to 65% (kN) | A2-70 as % of 12.9 |
|---|---|---|---|---|---|---|---|
| M3 | 5.03 | 3.22 | 4.53 | 5.43 | 2.26 | 3.53 | 41.7 |
| M4 | 8.78 | 5.62 | 7.90 | 9.48 | 3.95 | 6.16 | 41.7 |
| M5 | 14.18 | 9.08 | 12.76 | 15.32 | 6.38 | 9.96 | 41.7 |
| M6 | 20.12 | 12.88 | 18.11 | 21.73 | 9.06 | 14.13 | 41.7 |
| M8 | 36.61 | 23.43 | 32.95 | 39.54 | 16.47 | 25.70 | 41.7 |
| M10 | 57.99 | 37.11 | 52.19 | 62.63 | 26.10 | 40.71 | 41.7 |
| M12 | 84.27 | 53.93 | 75.84 | 91.01 | 37.92 | 59.16 | 41.7 |
| M14 | 115.44 | 73.88 | 103.90 | 124.67 | 51.95 | 81.04 | 41.7 |
| M16 | 156.67 | 100.27 | 141.00 | 169.20 | 70.50 | 109.98 | 41.7 |
| M20 | 244.79 | 156.67 | 220.31 | 264.38 | 110.16 | 171.85 | 41.7 |
| M24 | 352.50 | 225.60 | 317.25 | 380.70 | 158.63 | 247.46 | 41.7 |
Counterbore, clearance hole, and the head bearing area left over
| Size | Close-fit drill (mm) | Normal-fit drill (mm) | Counterbore dia (mm) | Head dia d_k max (mm) | Radial gap (mm) | Counterbore depth (mm) | Depth / d | Head bearing area, normal fit (sq mm) |
|---|---|---|---|---|---|---|---|---|
| M3 | 3.40 | 3.70 | 6.50 | 5.50 | 0.500 | 3.00 | 1.0000 | 13.0 |
| M4 | 4.40 | 4.80 | 8.25 | 7.00 | 0.625 | 4.00 | 1.0000 | 20.4 |
| M5 | 5.40 | 5.80 | 9.75 | 8.50 | 0.625 | 5.00 | 1.0000 | 30.3 |
| M6 | 6.40 | 6.80 | 11.25 | 10.00 | 0.625 | 6.00 | 1.0000 | 42.2 |
| M8 | 8.40 | 8.80 | 14.25 | 13.00 | 0.625 | 8.00 | 1.0000 | 71.9 |
| M10 | 10.50 | 10.80 | 17.25 | 16.00 | 0.625 | 10.00 | 1.0000 | 109.5 |
| M12 | 12.50 | 12.80 | 19.25 | 18.00 | 0.625 | 12.00 | 1.0000 | 125.8 |
| M14 | 14.50 | 14.75 | 22.25 | 21.00 | 0.625 | 14.00 | 1.0000 | 175.5 |
| M16 | 16.50 | 16.75 | 25.50 | 24.00 | 0.750 | 16.00 | 1.0000 | 232.0 |
| M20 | 20.50 | 20.75 | 31.50 | 30.00 | 0.750 | 20.00 | 1.0000 | 368.7 |
| M24 | 24.50 | 24.75 | 37.50 | 36.00 | 0.750 | 24.00 | 1.0000 | 536.8 |
k equals d, b equals 2d plus 12, and one column four published tables will not agree on
The head of a socket head cap screw is exactly as tall as the screw is thick. k = d, at every size in ISO 4762: an M6 has a 6 mm head, an M16 has a 16 mm head, an M24 has a 24 mm head. Five independently published tables were read for this page and all five agree at all eleven sizes from M3 to M24, and a sixth confirmation comes from a different kind of document altogether — the metric counterbore chart in Machinery’s Handbook puts the counterbore depth at the nominal screw diameter from M1.6 to M24, which is the same fact stated by a machinist rather than a fastener catalogue. It is a property of the head form, not an artefact of rounding, and it is the one thing worth memorising about this screw: you can size a counterbore from the M number alone and you will never be wrong.
The head diameter is not a constant multiple, and that is the second useful fact. d_k / d starts at 1.83 at M3 and falls through 1.75, 1.70, 1.67, 1.63 and 1.60 before settling at exactly 1.50 at M12, where it stays for every larger size. So the small sizes have proportionally bigger heads. That is why an M3 cap screw can carry a 2.5 mm hexagon socket — 0.83 of its own thread diameter — and why head bearing area per unit of clamp load is actually better at the bottom of the range than at the top. The standard’s thread length is a rule too: b = 2d + 12 mm, exact at all eleven sizes.
On key engagement, four published tables give two different answers, and this page prints both. Three of them — SSR Precision, fasten.it’s DIN EN ISO 4762 table and GlobalFastener’s ISO 4762:2004 table — give a round column at about half the thread diameter: 1.3, 2.0, 2.5, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 10.0, 12.0 mm. RoyMech gives 1.3, 2.0, 2.7, 3.3, 4.3, 5.5, 6.6, 8.8, 10.7, 12.9 — identical at the first two sizes, then 7 to 10 per cent higher all the way up. Three sources against one makes the round column the majority reading, and it means the claim that RoyMech’s column is ISO 4762’s published t minimum is not supported by anything read here. Both are in the table, labelled by source. What t decides is how deep the hexagon socket goes: it bounds how much of a head you can face off before breaking into the socket, and it is the engagement a short or worn key has to live inside. It is not the counterbore depth, which is k, which is d, and which nobody disputes.
The stress area on this page is computed, not transcribed. ISO 898-1 defines the nominal stress area as the area of a circle whose diameter is the mean of the basic pitch diameter and the minor diameter at the rounded root, and both of those follow from the thread’s fundamental triangle: with H = P√3/2, the pitch diameter is d − 3H/4 and the root diameter is d − 5H/4 − H/6. Working it that way agrees with the standard’s published column to better than 0.4 per cent at every size, and the worst case is M14, where the published figure is itself rounded to three significant figures. One subtlety for anyone comparing an inch screw: the unified definition used for a UNC thread is a different expression and gives 1 to 2 per cent less area for the same geometry, so an inch stress area and a metric one are not strictly the same quantity.
Then the number a designer is actually after: clamp load. Proof load is the stress area times the class’s proof stress, and the clamp load is whatever fraction of proof the joint is tightened to — 60 to 70 per cent for a reusable joint, 75 to 90 for a yield-controlled one. An M12 in 12.9 at 65 per cent gives 59.2 kN. And A2-70 stainless gives 41.7 per cent of that, at every size. Socket head cap screws are predominantly sold in 12.9; A2-70 is the same shape with less than half the proof stress, and it is indistinguishable by eye. Substituting it for corrosion reasons without recalculating removes 58 per cent of the clamp load, and it brings two other things with it: austenitic stainless galls against itself, so a stainless screw in a stainless thread can seize while being tightened, and its modulus is about 10 per cent lower, which makes the bolt a softer spring and shifts the joint’s load sharing.
One figure on this page surprises people, and it should. The head bears on the annulus between its own diameter and the clearance hole, which for an M12 over a normal-fit hole is about 126 mm². At 59 kN of clamp load that is 470 MPa under the head — above the yield of structural steel and several times aluminium’s. It is not an unusual case; a 12.9 screw at 65 per cent of proof does this at every size. The head of a socket head cap screw is deliberately small, because that is what lets it live in a counterbore, so the pressure has nowhere to go. That is what hardened washers are for, and it is why a cap screw leaves a ring in anything soft.
Where to go next on this site. The bolt proof load and stress area page goes further into the property classes and the stress area definition; the bolt torque page does the nut factor properly rather than with the single 0.20 used here; the thread engagement length page sizes the tapped depth for the parent material; and the counterbore and countersink page covers the hole itself. For the hexagon key and the other drive systems, the drive size page has the cross-reference, and if the question is an inch screw rather than a metric one, the metric and imperial cross-reference computes how far apart the two really are.
Frequently asked questions
What is the head height of a socket head cap screw?
It is equal to the nominal thread diameter, exactly, at every size in ISO 4762. An M6 has a 6 mm head height, an M10 has 10 mm, an M24 has 24 mm. Five published tables were read for this page and all five agree at all eleven sizes, and Machinery’s Handbook’s metric counterbore chart independently gives the counterbore depth as the nominal diameter from M1.6 to M24. The practical value is that you can size a counterbore from the M number alone: an M8 cap screw needs an 8 mm deep counterbore to sit flush, and you do not need a table to know it.
What counterbore do I need for an M8 socket head cap screw?
14.25 mm diameter by 8.00 mm deep, with a normal-fit clearance hole of 8.80 mm or a close-fit one of 8.40 mm — the diameters from Machinery’s Handbook, the depth from the head height, which is the nominal diameter. Two cautions. The radial gap between a 14.25 mm counterbore and a 13.00 mm head is only 0.625 mm, so the counterbore and the clearance hole should be cut from the same setup or concentricity is not guaranteed. And a counterbore cut exactly to the head height leaves the head flush with nothing to spare; if the face is to be machined or painted afterwards, go a little deeper.
How much torque does an M12 12.9 socket head cap screw take?
About 142 N·m to reach 65 per cent of proof load, which is a clamp load of 59.2 kN — but read that as an order of magnitude rather than a specification. The figure comes from T = K·F·d with a nut factor K of 0.20, and the nut factor is a lumped empirical coefficient that varies by a factor of two between a dry thread, a lightly oiled one, a zinc-plated one and one with a thread locker on it. The clamp load is the real design quantity and the torque is only a way of getting there; where it matters, use the manufacturer’s own factor or control the tightening by angle or by bolt extension instead.
Is an A2-70 stainless cap screw as strong as a 12.9 one?
No, and the gap is much larger than most people expect: A2-70’s proof stress is 450 MPa against class 12.9’s 1080, so the same screw in the same hole gives 41.7 per cent of the clamp load. The ratio is the same at every size. Socket head cap screws are predominantly sold in 12.9, A2-70 looks identical, and the swap is usually made for corrosion reasons without anyone recalculating the joint. Three other differences come with it: austenitic stainless galls against itself and can seize while being tightened, its elastic modulus is about a tenth lower so the bolt is a softer spring, and in the annealed state it is not magnetic, which is the quickest way to spot that somebody has substituted one.
What is the key engagement t, and why do sources disagree?
It is the depth of the hexagon socket — how far the key goes in. Four published tables were read for this page and they give two different columns. SSR Precision, fasten.it and GlobalFastener all give a round column at about half the thread diameter (6.0 mm at M12); RoyMech gives about 0.55 of the thread diameter (6.6 mm at M12), identical at M3 and M4 and then 7 to 10 per cent higher. Three sources against one makes the round column the majority reading, and the claim that RoyMech’s is ISO 4762’s published minimum is not supported by anything read here. Both are printed on this page with their sources named. Where it matters — facing a head down, or checking that a worn key still has engagement — use the smaller figure, and if the consequence is real, measure the screw you have.
How long is the thread on an ISO 4762 cap screw?
b = 2d + 12 mm, exactly, at every size from M3 to M24: 18 mm on an M3, 28 on an M8, 36 on an M12, 60 on an M24. It is a rule rather than a table, which is worth knowing because it tells you immediately whether a given length will be fully threaded or will have a plain shank. A 25 mm long M8 is fully threaded; a 40 mm long M8 has 28 mm of thread and 12 mm of plain shank. The rule was checked here only across M3 to M24, which is as far as the published table read for this page goes, so do not extrapolate it beyond that.
Why does a socket head cap screw leave a ring in aluminium?
Because the head is small for the clamp load it delivers. The head bears on the annulus between its own diameter and the clearance hole, which for an M12 over a normal-fit hole is about 126 mm²; at 65 per cent of a 12.9 proof load that is 59 kN on 126 mm², or 470 MPa — above the yield of structural steel and several times that of aluminium alloy. The small head is the point of the screw, because it is what lets it live inside a counterbore, so the pressure has nowhere to spread. The fix is a hardened washer, or a flanged screw, or accepting the embossing and checking that the relaxation it causes does not lose the joint its clamp load.
Related calculators
References
- ISO 4762, Hexagon socket head cap screws, and DIN 912, its German predecessor and the designation the trade still uses. Both cited by number and neither reproduced. The dimensional columns used on this page were read in four independent published tables and are attributed to them individually below, because on one column — the key engagement t — those four tables do not agree, and this page prints the disagreement rather than choosing for the reader.
- RoyMech, Cap screws dimension table, stating its dimensions as “generally to BS EN ISO 4762, BS 3643-2 and BS 4168”. One of the three sources behind the socket head cap screw columns, and the MINORITY source on key engagement: it prints t as 1.3, 2.0, 2.7, 3.3, 4.3, 5.5, 6.6, 8.8, 10.7 and 12.9 mm for M3 to M24 (no M14), roughly 0.55 d, where the three other tables read for this page print a round column at roughly 0.5 d. Its dk, k and s agree exactly with all of them. It is also the source of the 35.31 mm dk minimum at M24 that one other table contradicts.
- SSR Precision, DIN 912 / ISO 4762 socket head cap screw dimension tables. The second of the three sources behind this page’s cap screw columns, and the one whose key engagement column — 1.3, 2.0, 2.5, 3.0, 4.0, 5.0, 6.0, 8.0, 10.0, 12.0 mm — turned out to be the majority reading once two further tables were checked against it. Its dk, k and s match RoyMech and whole-spec exactly at every size.
- whole-spec.com, ISO 4762 specification table. The third source behind this page’s cap screw columns. It confirms dk, k and s at every size, including the M14 that most catalogues skip, and it carries no key engagement column at all, so it cannot settle the argument over t — which is itself worth knowing, because a table that omits a column is not a table that agrees with you about it.
- fasten.it, DIN EN ISO 4762 Standard Dimensions for Hexagon socket head cap screws (read 6 October 2026). The most complete of the four tables, transcribed in full for this batch. It confirms dk max, dk min, k max and s nominal against the three earlier sources, and adds three columns none of them carried: k min, w min — the wall left between the bottom of the hexagon socket and the underside of the head, which is what actually ruptures when a head is machined down or a socket is over-torqued — and b, the standard’s thread length, which turns out to be exactly 2d + 12 mm at all eleven sizes. It also prints the M14 key engagement (7.0 mm) that the other three omit, and it gives dk min at M24 as 35.61 against RoyMech’s 35.31.
- globalfastener.com, ISO 4762:2004 standard detail page (read 6 October 2026). Read specifically as a fourth, independent reading of the key engagement column, and it sides with the round one: 1.5, 2.0, 2.5, 3.0, 4.0, 5.0, 6.0, 8.0, 10.0, 12.0 mm. Note the M3: it prints 1.5 where the other three print 1.3, which is the only size at which it departs from them, and it does not print M14 at all.
- EngineersEdge, ISO Socket Head Screw Size Data Table Chart ISO 4762 (read 6 October 2026). Read as a fifth check on the two columns this page treats as settled. It confirms head diameter and head height at every size from M3 to M24 — so k = d is now five sources deep — and it prints the hexagon socket as its MAXIMUM across flats (2.58 at M3, 19.275 at M24) rather than the nominal, which is a useful reminder that an “s” column can be either.
- ISO 898-1, Mechanical properties of fasteners made of carbon steel and alloy steel — Part 1: Bolts, screws and studs with specified property classes. Cited by number and not reproduced. Two things are taken from it. The first is the DEFINITION of the nominal stress area, As = π/4 · ((d₂+d₃)/2)² — a definition, not a table — which lets this page compute every stress area from basic thread geometry instead of transcribing the standard’s printed column. The second is the property class system, where the first number is a hundredth of the nominal tensile strength and the second is ten times the ratio of proof stress to it, so 12.9 means 1200 MPa and 0.9 × 1200 = 1080 MPa.
- Berger Tools, Property Classes — Screws and Nuts (PDF, read 6 October 2026). The first source on property classes and the one this page computes from: 800/640, 1000/900 and 1200/1080 MPa for 8.8, 10.9 and 12.9, with the designation rule spelled out — the first figure is a hundredth of the tensile strength and the second gives the yield ratio. It prints no split by diameter, which the other source does.
- TR Fastenings, Mechanical properties — steel screws and bolts, ISO 898-1 / EN 20898-1 (read 6 October 2026). Read as the second source on property classes, and it DISAGREES with the first in a way worth printing: it gives ISO 898-1’s tabulated minima rather than the figures the class designation implies — 800 MPa tensile for 8.8 up to M16 and 830 above it, 1040 for 10.9, 1220 for 12.9, with proof stresses of 640, 940 and 1100 MPa. This page uses the lower, designation-derived set, which is the conservative choice; a joint designed on the lower proof load will not be over-tightened by a screw that actually meets the higher one.
- ISO 3506-1, Mechanical properties of corrosion-resistant stainless steel fasteners. Cited by number, not reproduced. The A2-70 figures used here — 700 MPa minimum tensile, 450 MPa 0.2 per cent proof stress, for diameters up to M24 — were read in amesweb’s published summary of it (below), which also states the stress area definition in the same form ISO 898-1 uses. A2-70 is the reason this page carries a stainless column at all: a socket head cap screw looks the same in A2 as in 12.9 and has a proof stress 42 per cent of it.
- amesweb, Stainless Steel Bolt Grades Chart (read 6 October 2026), summarising ISO 3506-1. The source of the A2-70 and A4-80 figures: 700 and 800 MPa minimum tensile, 450 and 600 MPa 0.2 per cent proof stress, both for diameters up to M24. It also prints the stress area definition As,nom = π/4 · ((d₂+d₃)/2)² in the same form ISO 898-1 uses, which is the expression this page computes with.
- amesweb, Counterbore Hole Dimensions Chart (Metric) (read 6 October 2026), citing Machinery’s Handbook 29th and 30th editions. The source of the close-fit and normal-fit clearance drills and the counterbore diameters and depths on this page. It carries one figure worth the whole chart: the counterbore DEPTH is equal to the nominal screw diameter at every size from M1.6 to M24. That is k = d, arrived at from a machining chart rather than from a screw table, and it is the most independent confirmation of the finding on this page that could be asked for.
- Machinery’s Handbook, 29th and 30th editions (Oberg, Jones, Horton and Ryffel), cited as the authority amesweb’s counterbore and clearance-hole chart gives for its figures. Not read directly for this batch: the chart was, and the attribution is the chart’s. Recorded this way round deliberately, because “Machinery’s Handbook” is the easiest citation in engineering to put on a number nobody looked up.
- ISO 273, Fasteners — Clearance holes for bolts and screws. Cited by number and NOT used numerically, which is worth being explicit about: its three fit classes (fine, medium, coarse) were not obtainable in a readable published table during this batch, so the clearance holes printed on this page are the close-fit and normal-fit columns from Machinery’s Handbook by way of amesweb, attributed there and not to ISO 273. If you are working to ISO 273 specifically, check its own table: the two systems are close but they are not the same document.
CalcEngines mechanical calculators are for education and design estimates. Results depend on the values you enter and the assumptions stated on the page, and they do not replace a qualified engineer's check. Standards, material grades and manufacturer data change — verify against the current standard and the supplier's own catalogue before you specify, machine or build anything.
