The quotation says “Class 150, WCB”. The shop cuts material against PN16 and tests at 1.5 × 16 = 24 bar. The customer’s inspector asks for 30 bar. Nobody miscalculated — two pressure languages were never converted. ASME B16.34 is a product standard (flanged, threaded and welding-end valves); it does not teach you how to run a pressure test, but it sets the starting point: the 38 °C (100 °F) rating, multiplied by 1.5 for the shell and 1.1 for the seat, with the result rounded up to the next 25 psi increment.
Those three steps stacked together produce the familiar “6 bar short” or “one class off” outcomes. This article runs: Class is not psi, derating with temperature, how the test pressure is computed, hold time by size, medium and gauges, the division of labour with API 598, the traps in Class-versus-PN, the Chinese mapping, and equipment.
1. Class is not psi — the 38 °C rating
The table below gives Group 1.1 (carbon steel, e.g. WCB) ratings at 38 °C (100 °F), the most-used block of B16.34 Table 2-1.1:
| Class | Rating at 38 °C (psig) | In bar | Shell test (1.5×, rounded) |
|---|---|---|---|
| Class 150 | 285 | ≈ 19.6 | ≈ 450 psi / 31 bar |
| Class 300 | 740 | ≈ 51 | ≈ 1 125 psi / 77.6 bar |
| Class 600 | 1 480 | ≈ 102 | ≈ 2 225 psi / 153 bar |
| Class 900 | 2 220 | ≈ 153 | ≈ 3 350 psi / 231 bar |
| Class 1500 | 3 705 | ≈ 255 | ≈ 5 575 psi / 384 bar |
| Class 2500 | 6 170 | ≈ 425 | ≈ 9 275 psi / 640 bar |
| Class 4500 | 11 100 | ≈ 765 | ≈ 16 650 psi / 1 148 bar |
Class 150’s 285 psig is only about 19.6 bar. It looks higher than PN16’s 16 bar, but the two are not on the same basis: PN is a nominal value at 20 °C, Class is a rating at 38 °C; and the flanges belong to ASME B16.5 versus EN 1092-1, so similar pressure does not mean they mate. Class 300’s 51 bar against PN40’s 40 bar is a wider gap still.
2. Ratings fall quickly with temperature
| Temperature | Class 150 (carbon steel, Group 1.1) | Note |
|---|---|---|
| 38 °C (100 °F) | 285 psig ≈ 19.6 bar | Rating basis; test pressure starts here |
| 93 °C (200 °F) | 260 psig ≈ 17.9 bar | Derating begins |
| 204 °C (400 °F) | 200 psig ≈ 13.8 bar | 70 % of the basis |
| 316 °C (600 °F) | 140 psig ≈ 9.7 bar | 49 % of the basis |
| 427 °C (800 °F) | 80 psig ≈ 5.5 bar | Carbon steel near its limit |
Two uses. First, selection: a carbon steel line at 300 °C and 12 bar cannot use Class 150 — at 316 °C it is down to 9.7 bar; go to Class 300. Second, testing: test pressure always comes from the 38 °C rating, never from the allowable at design temperature. Shops that mix the two end up testing at 9.7 × 1.5 = 14.6 bar and never reach the required pressure.
Material group matters as much: carbon steel (1.1), low-alloy steels (from 1.2), austenitic stainless (2.x) and nickel alloys (3.x) each have their own curve. The same Class 150 holds up far better in stainless at high temperature than in carbon steel. Quoting “Class 150” without the material group is an incomplete quote.
3. Test pressure — two multipliers, one rounding step
- Shell: ≥ 1.5 × the 38 °C rating;
- Seat: ≥ 1.1 × the 38 °C rating;
- Rounding: round up to the next 25 psi (about 1 bar) increment.
Take Class 2500 carbon steel: 425.5 bar (6 170 psi) at 38 °C → shell 1.5× = 9 255 psi → rounded up to 9 275 psi (about 640 bar). The rounding is not optional: an inspector checking against the standard’s own numbers will mark 9 255 as “not rounded per standard”. Likewise Class 150: 285 × 1.5 = 427.5 psi, reported as 450 psi (about 31 bar), not 29.5 bar.
One more point that gets missed: design pressure is not the multiplier base. A 16 bar design pressure with a Class 300 valve is still tested from the Class 300 38 °C rating, not from 16 bar. Valve class and line design pressure are two different things.
4. Hold time follows size, not pressure
| Size | Shell hold | Note |
|---|---|---|
| ≤ 2″ (DN 50) | 15 s | Fast cycle, unchanged criteria |
| 2½” – 6″ (DN 65 – 150) | 60 s | Most common band |
| 8″ – 12″ (DN 200 – 300) | 120 s | Needs pressure stability |
| ≥ 14″ (DN 350+) | 300 s | Five minutes; hardest on pressurisation control |
Pressure multiplier and hold time are independent variables: small valves see high pressure but short holds; large valves see relatively lower pressure but must hold for five minutes. On a bench, the 300 s hold above DN 350 is harder than the high pressure — trapped air in the valve and piping compresses slowly, so the curve keeps creeping and looks like decay when it is really gas coming into equilibrium.
5. Medium and gauges — three hard constraints
- Water at 5 – 50 °C: too cold adds brittle-fracture risk, too hot changes material behaviour and readings;
- Chloride below 100 ppm for stainless: testing an austenitic stainless valve straight off the mains can leave chloride-induced stress corrosion — it passes today and cracks in three months;
- Gauge range ≥ 1.5× and ≤ 4× the test pressure: for a 9 275 psi test, a 12 000 psi gauge (1.29×) fails the rule and a 15 000 psi gauge (1.6×) passes. This is the constraint most often broken at high class.
6. How B16.34 and API 598 divide the work
| Item | ASME B16.34 | API Std 598 |
|---|---|---|
| Nature | Product standard (design, materials, wall thickness, ends, marking, test requirements) | Inspection and test method standard |
| Test pressure | Specifies: 1.5× / 1.1× the 38 °C rating, rounded to 25 psi | Refers to B16.34 (steel) or its own table (cast iron) |
| Hold time | By size | By size; the values agree |
| Leakage criteria | Refers to product standards or MSS SP-61 etc. | Specifies: drops/min, bubbles/min by size and type |
| Special class | Special Class (SPL): additional RT/MT/PT/UT, higher rating at the same dimensions | No such concept |
The accurate statement is: B16.34 fixes how much pressure and for how long; API 598 fixes how to apply it and how much leakage passes. Both must be cited; one alone is incomplete. For the API 598 versus GB/T 13927 criteria see the comparison.
7. Class versus PN — three traps
- Close numbers, incompatible flanges: Class 150 and PN16 are close in pressure, but B16.5 and EN 1092-1 differ in bolt circle and facing, so they do not bolt together;
- Different reference temperature: PN at 20 °C, Class at 38 °C, and the high-temperature curves have different shapes — no linear conversion;
- Different rounding rules: the US system rounds up to 25 psi, the Chinese and European systems do not, so the same valve shows different test pressures in two reports. Neither is wrong; the rules differ.
8. The Chinese mapping
| Dimension | Chinese counterpart | Difference |
|---|---|---|
| General product requirements | GB/T 12224 (steel valves, general requirements) | Organised by PN and allowable stress; no Class or SPL system |
| Pressure test | GB/T 13927-2022 | 1.5× CWP shell, 1.1× CWP seat, no 25 psi rounding |
| Inspection and testing | GB/T 26480-2011 (MOD API 598) | Leakage criteria close; pressure basis still national |
| Flanges | GB/T 9113 series (aligns with EN 1092) / HG/T 20615 (aligns with B16.5) | Wrong series means holes do not line up |
For a US customer, the drawing and the report must both carry four pieces of information: Class, material group, 38 °C rating, and the rounded test pressure. Class alone gives the reviewer no way to check your numbers.
9. Configuring the equipment
- High-pressure tier: Class 2500 shell is about 640 bar and Class 4500 exceeds 1 100 bar — pump, piping, blinds and seals must be selected as one system at the top rating;
- Long holds at large size: 300 s above DN 350 needs pressure stabilisation and automatic top-up; closing a valve and hoping will not hold five minutes;
- Water temperature control: 5 – 50 °C, with chloride monitoring for stainless work (inline chloride/chlorine measurement, or demineralised water);
- Gauges: sized between 1.5× and 4× test pressure; dual gauges (low and high band) at high class;
- Data logging: pressure-time curves must export, and the rounding rule belongs in the report template so nobody hand-converts.
For high-pressure and large-size frame configuration see the valve comprehensive test bench; individual machines include the valve pressure test machine and shell hydrostatic test bench.
FAQ
Is Class 150 the same as PN16?
No. Class 150 is rated 285 psig (about 19.6 bar) at 38 °C; PN16 is 16 bar at 20 °C. The numbers are close, but the reference temperature differs and the flanges follow different standards (B16.5 versus EN 1092-1), so they neither convert nor bolt together.
Why round the test pressure up to 25 psi?
It is a B16.34 rule that puts results from different laboratories and benches onto one discrete series so acceptance can be checked. 427.5 psi is reported as 450 psi and 9 255 psi as 9 275 psi. Skip it and the report is marked “not rounded per standard”.
A line at 300 °C design temperature — which pressure for the test?
Still the 38 °C (100 °F) rating. The allowable at design temperature is for selection; test pressure recognises only the 38 °C rating. Mixing them is a common cause of under-pressurised tests on hot lines.
What does Special Class (SPL) actually change?
SPL requires additional non-destructive examination (radiographic, magnetic particle or penetrant, ultrasonic) and yields a higher rating at the same dimensions and material. Specifying SPL raises cost and lead time noticeably, and the NDE records must exist — stamping SPL on the nameplate is not enough.
Specific clauses follow the current ASME B16.34 edition; the tables above are for orientation in selection, quoting and test discussions. For PN versus CWP conversion see PN and CWP; for gas distribution valves see EN 13774; for US waterworks valves see AWWA C515; and the full map is in the valve test standards index.
Related test equipment
Each test above maps to a bench. Pick by valve type and by the item you need to prove:
- Valve pressure test machine — shell at 1.5x and seat at 1.1x, with a report per valve
- Valve testing equipment range — browse all models by valve type and test item