KOBE STEEL Ultra-High-Pressure Snubbers: Manufacturing Compressor Pulsation Vessels at 252–339 Bar

The short version, up front: A reciprocating compressor snubber is a pressure vessel whose entire job is to absorb pressure pulsation — and at the discharge of a third or fourth compression stage, that vessel has to do it at 250 to 339 bar. Lmart fabricated a full set of suction and discharge snubbers (plus inter-stage separators) for a four-stage compressor train tied to KOBE STEEL's Xinjiang Meike project, where design pressures climbed stage by stage from roughly 152 bar to 339 bar. The engineering story is not "how do you build a tank that holds 339 bar." It's how the material grade, the wall thickness, the weld procedure, and the pulsation-bottle geometry all have to escalate together, stage by stage, while every vessel still has to pass the same hydrostatic test and produce the same audit-grade data book. This article walks through how that escalation works, what GB 150 and ASME Section VIII actually require of these vessels, and the decisions that separate a snubber that runs quietly for 100,000 hours from one that cracks a nozzle weld in year two.

This is written for the people who actually have to specify or buy these vessels: EPC procurement engineers, reciprocating-compressor OEM design teams, and design-institute mechanical engineers who keep getting asked "can your fabricator handle the high-pressure stages, or only the low ones?" If that question sounds familiar, this is for you.

---

Why this matters right now

In July 2025, ASME published the new edition of the Boiler and Pressure Vessel Code, and for once the most consequential changes were not buried in an appendix. The 2025 BPVC Section VIII edition added new evaluation criteria for vessels in cyclic service, pulsation, and vibration, with revised guidance on identifying fatigue-sensitive components, plus new methods for relief-device sizing at ultra-high pressure (per the Inspectioneering summary of the 2025 VIII-1 and VIII-2 changes, published December 2024 ahead of the July 2025 release). Read that again with a compressor snubber in mind. A snubber is defined by pulsation. It is the one vessel in the plant that lives in cyclic service by design. The code catching up to that reality is not academic — it changes what a competent fabricator has to be able to document.

For anyone who buys or designs these vessels, that 2025 code update lands at an awkward time. The high-pressure-vessel segment was valued at about USD 3.72 billion in 2025 and is projected to reach roughly USD 5.93 billion by 2034, a CAGR near 5.2% (Dataintelo, Global High Pressure Vessels Market Research Report, 2025). The broader pressure-vessel market sat around USD 41 billion in 2025 with Asia-Pacific holding roughly 45% of it (The Business Research Company, Pressure Vessel Global Market Report 2026). Demand is rising, the rulebook just got stricter on exactly the service these vessels see, and certification lead times for new fabricators run 12 to 18 months. The practical consequence: the pool of shops that can credibly build a 339-bar discharge snubber and prove it on paper is not expanding as fast as the order book.

From where we sit at Lmart, this is not a forecast — it's the daily reality of the shop floor. We built the snubbers and separators for the KOBE STEEL Xinjiang Meike compressor train across the full pressure range, from the first-stage suction vessels up to the 339-bar fourth-stage section. The thing that surprises buyers most is how unforgiving the high-pressure stages are about the boring stuff: which grade of plate, how many millimetres of wall, how the nozzle reinforcement is detailed, and whether the welder qualified on the actual thickness or a thinner coupon. This article puts those decisions on the table.

Here is the promise: by the end, you will be able to read a four-stage compressor snubber specification and immediately see where the risk is — which stage needs the upgraded steel, why the wall thickness jumps the way it does, what the snubber's pulsation job demands beyond simple pressure containment, and what a defensible vendor data book has to contain. No marketing. Just the engineering that decides whether the vessel works.

---

Table of Contents

1. What a compressor snubber actually is (and isn't)
2. The four-stage problem: pressure escalation, stage by stage
3. Material grading: why the steel changes with the stage
4. Wall thickness: where the millimetres come from
5. GB 150 vs ASME Section VIII for high-pressure vessels
6. Pulsation control: the part most buyers underestimate
7. The separators: small vessels, large consequences
8. Fabricating the high-pressure stages: welding, PWHT, NDE
9. Procurement guide: how to qualify a high-pressure snubber fabricator
10. The KOBE STEEL Xinjiang Meike reference set
11. Lmart capability — the factual version
12. Conclusion
13. Frequently Asked Questions
14. Further Reading

---

1. What a compressor snubber actually is (and isn't)

Start with the vocabulary, because it trips people up constantly.

A snubber, in reciprocating compressor language, is a pulsation suppression vessel — what API 618 calls a pulsation dampener and what shop floors variously call a "bottle," a "pulsation bottle," or a "surge drum." It sits directly at the suction and discharge of a compressor cylinder. Its job is to take the violent, periodic pressure waves that a piston generates every revolution and smooth them out before they reach the piping. As Dynaflow Research Group describes the function, excessive pressure pulsations from a compressor are reduced using these dampeners — bottles connected upstream and downstream of the cylinder — where the pulsation waves are absorbed, often in combination with internal orifice plates.

So a snubber is a pressure vessel, governed by pressure-vessel code (GB 150, ASME VIII, or PED depending on the project), but it has a second design problem layered on top: acoustics. A plain receiver tank only has to hold pressure. A snubber has to hold pressure and present the right internal volume and geometry to detune the compressor's pulsation frequencies. Get the volume wrong and you don't get a quiet vessel — you can get an amplifier. To quote the acoustic-design reality bluntly: the bottle itself has a resonance frequency, and incorrectly sizing it can lead to a situation where pulsations are amplified rather than dampened.

That dual nature is exactly why these vessels are harder to specify than their pressure rating alone suggests. A buyer who treats a 339-bar discharge snubber as "just a thick tank" will under-spec the internals, the nozzle layout, and the inspection scope.

A few clarifications on what a snubber is not:

• It is not a separator. A separator removes liquid (condensed hydrocarbon, lube oil carryover, water) from the gas stream using gravity, mesh pads, or vane packs. Some inter-stage vessels do both jobs, but the suction snubber and the suction separator are usually distinct vessels with distinct internals.
• It is not a surge volume for process control. Process surge drums are sized for residence time and inventory. A snubber is sized for acoustic performance, which is a completely different calculation.
• It is not optional. On a multi-stage reciprocating machine, an under-damped system shakes piping, fatigues nozzle welds, loosens flanged joints, and in the worst cases cracks the compressor's own valve assemblies. The snubber is a reliability component, not a nicety.

The reason this matters for fabrication: a snubber's internal volume is a contractual deliverable derived from the compressor OEM's acoustic study, not a number the vessel shop is free to round. When KOBE STEEL's compressor engineering sets the snubber volumes for each stage, the fabricator's job is to deliver those volumes inside a shell that also satisfies the pressure code at that stage's design pressure. On the high-pressure stages, those two requirements start fighting each other, and that fight is the whole story.

---

2. The four-stage problem: pressure escalation, stage by stage

A four-stage reciprocating compressor exists because the pressure ratio between suction and final discharge is too large to achieve in one cylinder. You split the work. Each stage compresses the gas a manageable amount, cools it (inter-stage cooling), knocks out any condensed liquid (separators), and hands it to the next stage at a higher pressure. By the fourth stage's discharge, you're at the plant's delivery pressure.

For the snubbers, this means every stage has its own pair of vessels at its own pressure level, and those levels climb fast. On the KOBE STEEL Xinjiang Meike train, Lmart's scope covered the full set:

• 1st stage suction & discharge snubbers (2 each)
• 2nd stage suction & discharge snubbers (2 each)
• 3rd stage suction & discharge snubbers (2 each)
• 4th stage suction & discharge snubbers (2 each)
• 1st & 2nd stage suction separators

The headline numbers tell you where the difficulty concentrates. The third-stage discharge ran at 339 bar. The fourth-stage suction also sat at 339 bar (the gas arrives at the next cylinder at roughly the pressure it left the previous one, minus cooler and piping losses). The fourth-stage discharge ranged across the 250–339 bar band depending on the operating point, with the high-pressure section marked at design conditions spanning roughly 152 bar to 339 bar across the train.

Here is the parameter table that EPC and OEM engineers actually want — the stage-by-stage escalation, using the real envelope from this reference set:

Stage / Vessel	Design pressure (bar)	Material grade	Wall thickness (mm)	Diameter (mm)	Length (mm)
1st stage suction snubber	Low (1st-stage range)	Q345R	~12	up to 945	1700–2800
1st stage discharge snubber	rising	Q345R	rising	mid	mid
2nd stage suction snubber	mid	Q345R / 16MnⅡ	rising	mid	mid
2nd stage discharge snubber	mid–high	16MnⅡ / 16MnⅢ	rising	mid	mid
3rd stage suction snubber	high	16MnⅢ	thick	smaller	mid
3rd stage discharge snubber	339	16MnⅢ / 16MnⅣ	up to 41	250–smaller	mid
4th stage suction snubber	339	16MnⅣ	up to 41	250	mid
4th stage discharge snubber	250–339	16MnⅣ	up to 41	250	mid
1st / 2nd stage suction separators	low	Q345R	~12	350	—

(Across the full set, wall thickness ranged 12–41 mm, diameters 250–945 mm, and overall lengths 1,700–2,800 mm, all to GB 150.)

Three patterns jump out of that table, and each is a design lesson:

Pattern 1 — diameter shrinks as pressure rises. The low-pressure suction snubbers are large (up to 945 mm diameter) because low-pressure gas is bulky; you need volume to dampen it acoustically. The high-pressure discharge snubbers are small in diameter (down to ~250 mm) because dense high-pressure gas occupies less space, and because a smaller diameter dramatically reduces the wall thickness needed to hold a given pressure. This is not an accident — it's the geometry working in the fabricator's favour, and a good design exploits it.

Pattern 2 — wall thickness climbs even as diameter falls. Despite the high-pressure snubbers being narrower, their walls reach 41 mm. Pressure wins. A 250 mm vessel at 339 bar still needs a thick wall. We'll do the arithmetic in §4.

Pattern 3 — the material grade upgrades with the stage. The first stage uses Q345R. By the fourth stage the plate is 16MnⅣ. That escalation — Q345R → 16MnⅡ → 16MnⅢ → 16MnⅣ — is the single most important specification decision on the whole train, and it's the one buyers most often misunderstand. It gets its own section next.

---

3. Material grading: why the steel changes with the stage

This is where a lot of procurement specifications go quietly wrong. A buyer sees "carbon steel snubbers" on the train and assumes one material covers all eight stages. On a real high-pressure train, it doesn't — and using first-stage steel on the fourth stage is either dangerously thin-walled or absurdly heavy.

The KOBE STEEL set used a graded family of Chinese pressure-vessel steels, escalating with stage:

• Q345R — the workhorse low-alloy pressure-vessel plate (formerly designated 16MnR; Q345R is the current GB/T 713 designation). Good strength, excellent weldability, well understood. Used on the low-pressure stages and the separators.
• 16MnⅡ / 16MnⅢ / 16MnⅣ — progressively tighter-controlled grades within the 16Mn family, with increasing requirements on chemistry, cleanliness, mechanical properties, and (critically) through-thickness toughness as plate thickness increases.

Why does the grade have to climb? Two reasons, and they compound.

Reason 1 — allowable stress and thickness. A higher-grade plate has a higher allowable stress, which means a thinner wall for the same pressure and diameter. On a 339-bar vessel, every millimetre of wall you can shave by using a better-graded steel is a millimetre you don't have to weld, heat-treat through, and radiograph. The upgraded grade isn't gold-plating — it's what keeps the wall buildable. If you tried to hold 339 bar with first-stage-grade plate at the same diameter, you'd need a wall so thick that PWHT and volumetric NDE become genuinely difficult and the vessel becomes uneconomic to forge or roll.

Reason 2 — toughness in thick sections. Here's the part that bites. Steel's fracture toughness drops as section thickness increases (the through-thickness "size effect"), and a thick wall under high pressure stores a lot of elastic energy — exactly the condition where a brittle crack, once started, can run. So as the wall gets thicker on the high-pressure stages, the material specification has to tighten to guarantee adequate impact toughness in the full section thickness, not just at the surface. The 16MnⅢ/16MnⅣ grades carry stricter impact-test requirements precisely for this reason. The grade escalation is a toughness story as much as a strength story.

Here's the comparison that makes the logic visible:

Aspect	Low-pressure stages (Q345R)	High-pressure stages (16MnⅢ / 16MnⅣ)
Typical role	1st/2nd stage suction, separators	3rd/4th stage discharge & suction
Design pressure band	low → mid	up to 339 bar
Wall thickness	~12 mm	up to 41 mm
Governing concern	weldability, cost	through-thickness toughness, allowable stress
Impact testing	standard	stricter, full-thickness specimens
PWHT	per thickness/material	mandatory, tightly controlled
NDE	per code	typically 100% RT + UT on pressure welds

The practical takeaway for anyone writing the spec: do not flatten the material schedule to a single grade. Let each stage carry the grade its pressure and thickness demand. A fabricator who proposes one steel for the whole train either doesn't understand the toughness problem or is planning to over-build the low stages to avoid stocking multiple grades — both are red flags. On the KOBE STEEL set, the graded family was specified deliberately, and matching plate certificates to the correct stage was a documented step in the data book.

One more note for the international audience, because this comes up: these are GB-designated steels, and EPC engineers used to SA-516 Gr.70 will ask for the equivalent. Q345R is broadly comparable in role to SA-516 Gr.70 / SA-537 territory depending on grade, and the 16Mn family maps to the higher-strength low-alloy pressure-vessel plates. But "broadly comparable" is not "interchangeable" — the chemistry, the impact-test temperatures, and the supplementary requirements differ, and substituting one for the other on a 339-bar vessel without re-running the design is exactly the kind of shortcut that produces a non-conforming vessel. If your project requires ASME materials, that's a design-stage decision, not a substitution you make at the plate rack. (More on the GB-vs-ASME question in §5.)

---

4. Wall thickness: where the millimetres come from

Let's make the 41 mm wall concrete, because "thick wall" is hand-waving until you see where the number comes from.

The basic circumferential (hoop) stress relationship for a thin-to-moderate-wall cylindrical shell under internal pressure, in the form used by GB 150 and conceptually identical to ASME VIII-1 UG-27, is:

t = (P · D) / (2 · S · E − P · y) + CA

where:
- t = required wall thickness
- P = design pressure
- D = inside diameter (or mean diameter, depending on the exact code form)
- S = allowable stress of the material at design temperature
- E = joint efficiency (weld quality factor; 1.0 for full radiography)
- y = a coefficient (small correction term)
- CA = corrosion allowance

You don't need to solve it to read the lesson off it. Wall thickness is roughly proportional to pressure times diameter, divided by the material's allowable stress. That single relationship explains the entire vessel table in §2:

• Pressure (P) goes up fourfold from low stages to 339 bar → wall wants to go up roughly fourfold.
• Diameter (D) goes down from 945 mm to 250 mm on the high-pressure vessels → wall comes back down a lot. This is the geometry partly rescuing the fabricator.
• Allowable stress (S) goes up because the steel grade upgraded to 16MnⅣ → wall comes down a bit more.

Net effect: instead of an unbuildably thick wall, the high-pressure snubbers land at a hard-but-manageable ~41 mm. That number is the product of three deliberate moves working together — shrink the diameter, upgrade the steel, accept a thick-but-finite wall. Change any one and the others have to compensate. A design institute that specifies a large-diameter 339-bar snubber "because more volume is always better for pulsation" can accidentally drive the wall past 60 mm and turn a rollable plate vessel into a forged-shell problem. The acoustic volume has to be reconciled with the pressure-thickness math, not specified in isolation.

A worked feel for the numbers (illustrative, not a design calculation): a 250 mm ID vessel at 339 bar in a 16Mn-grade plate with full radiography lands in the high-30s-to-low-40s mm range once corrosion allowance and code margins are added. A 945 mm ID vessel at a low first-stage pressure lands near 12 mm. Both are in the same train. That 12-to-41 mm spread inside one equipment package is the signature of a real multi-stage high-pressure job, and it's why the fabricator has to be comfortable across the whole band — thin-wall rolling and fit-up at one end, heavy-wall PWHT and through-thickness UT at the other.

Why the thick wall changes everything downstream: Once you're at 41 mm, you've crossed thresholds that trigger mandatory steps:

1. PWHT becomes mandatory. Above the code's thickness/material trigger, post-weld heat treatment is required to relieve residual stress and restore toughness in the heat-affected zone. It's no longer optional.
2. Through-thickness toughness must be proven, which loops back to the material grade (§3).
3. Volumetric NDE gets harder to interpret. Radiographing or ultrasonically inspecting a 41 mm weld is a different proposition from a 12 mm weld — beam paths, technique qualification, and acceptance interpretation all change.
4. Preheat and interpass control during welding become tight, especially for the low-alloy grades, to avoid hydrogen-induced cracking in the thick section.

None of this is exotic for a shop that builds heavy-wall vessels routinely. All of it is a trap for a shop that mostly builds thin-wall receivers and "stretches" into a high-pressure job. That distinction — routine vs. stretch — is the entire procurement risk, and §9 turns it into a checklist.

---

5. GB 150 vs ASME Section VIII for high-pressure vessels

The KOBE STEEL snubbers were built to GB 150 — China's national standard for pressure vessels (GB 150 "Pressure Vessels," the general design and fabrication code, with GB/T 713 covering the plate materials). International buyers and OEMs reasonably ask: how does that compare to ASME Section VIII, and is a GB 150 vessel acceptable on my project? Here's the honest engineering answer.

The two codes are built on the same physics and the same philosophy. Both GB 150 and ASME VIII-1 are design-by-rule codes for the bulk of pressure vessels: hoop-stress formulas for shells, reinforcement rules for nozzle openings, joint-efficiency factors tied to the extent of radiography, mandatory hydrostatic testing, and a structured approach to materials, welding, and inspection. An engineer fluent in one can read the other and recognize every concept. The differences are in the details — allowable stress tables, specific material designations, exact test temperatures, documentation conventions, and the stamping/authorization regime.

Here's the comparison framed for a high-pressure compressor snubber specifically:

Topic	GB 150 (China)	ASME Section VIII Div.1
Scope	General pressure-vessel design & fabrication	General pressure-vessel construction (Div.1); Div.2 alternative rules for higher utilization; Div.3 for high/ultra-high pressure
Design method	Design-by-rule (formulas)	Design-by-rule (Div.1); design-by-analysis (Div.2/3)
Plate materials	GB/T 713 (Q345R, 16Mn family, etc.)	ASME II (SA-516, SA-537, SA-302, etc.)
Joint efficiency	tied to radiography extent	tied to radiography extent (UW-12)
Hydrotest	typically 1.25–1.5× design (per edition/clause)	1.3× MAWP (current UG-99)
PWHT triggers	thickness/material based	thickness/material based (UCS-56)
Authorization	special-equipment supervision inspection (TSG)	U / U2 stamp under ASME certification
Ultra-high pressure	handled within GB framework + supplementary requirements	Division 3 (KD rules) for the highest pressures

A few points that matter for the buyer's decision:

On code choice, the project usually decides — not the fabricator. If the end user, the licensor, or the local authority requires ASME and a U-stamp, the vessel is designed and built to ASME from day one with ASME materials. If the project is in a jurisdiction that mandates GB 150 and special-equipment supervision (as a domestic Chinese gas-processing project typically is), GB 150 is the governing code and the vessel carries the local supervision-inspection certification. The KOBE STEEL Xinjiang Meike vessels were a GB 150 scope. The competence a buyer should look for is a fabricator who can build to either code correctly — and who tells you plainly which one your project needs and why, rather than quietly building to whichever is cheapest for the shop.

On the 2025 ASME edition, the convergence is real. The 2025 BPVC's new attention to cyclic service, pulsation, and fatigue-sensitive components (and Division 3's updated ultra-high-pressure relief-sizing methods) reflects exactly the loading a compressor snubber sees. Whether your vessel is built to GB 150 or ASME, the engineering judgment the 2025 code now codifies — treat pulsation vessels as fatigue-relevant, validate relief at ultra-high pressure with proper analysis — is good practice that a serious fabricator already applies. Code language catching up doesn't change the physics; it just removes the excuse for ignoring it.

On material equivalence, don't improvise. As noted in §3, Q345R and the 16Mn grades have rough ASME counterparts, but "rough" is doing real work in that sentence. If your contract is ASME but your fabricator proposes GB materials "because they're equivalent," that's a design-review item with the Authorized Inspector, not a casual substitution. The right move is to fix the material code at design freeze and qualify the welding procedures against that code's requirements.

For a deeper treatment of the code-choice question and what each one demands at delivery, our PED vs ASME guide for EPC teams and the ASME Section VIII Div.1 delivery-document checklist lay out the documentation side in detail.

---

6. Pulsation control: the part most buyers underestimate

We've talked about the snubber as a pressure vessel. Now the half that buyers routinely under-spec: the snubber as an acoustic device.

A reciprocating compressor doesn't deliver smooth flow. Each piston stroke slams a slug of gas into the discharge, then draws a partial vacuum on the suction. The result is a strong periodic pressure wave at the running speed and its harmonics. Left alone, those waves do real damage: they shake piping, fatigue nozzle-to-shell welds, work flanged joints loose, and feed back into the compressor's own valves. On a four-stage high-pressure machine the energy in those waves is enormous, because pressure pulsation amplitude scales with the mean pressure — a 5% pulsation at 339 bar is a 17-bar swing, every revolution, forever.

The snubber's job is to absorb that. The physics is a Helmholtz/acoustic-filter problem: the bottle's internal volume, the nozzle (neck) geometry, and any internal orifice plates together form a low-pass acoustic filter that attenuates the pulsation before it reaches the piping. How well it works is a function of geometry, and the design is validated by an acoustic (pulsation) study — for most projects worldwide, following API 618, the reciprocating-compressor standard that sets criteria for dampener performance, allowable pulsation levels at the cylinder flange and in the pipework, shaking forces, and resulting vibration and stress, as Spectrum Acoustic and Dynaflow both describe.

How much difference does correct sizing make? A 2024 study on suction-side acoustic filtering (Scientific Reports, "Vibration and pressure pulsation elimination in a reciprocating compressor suction system using an acoustic filter," 2024) achieved a 75.56% reduction in peak-to-peak pressure amplitude and 14–61% reductions in vibration after properly tuning the acoustic filter. That's the spread between a snubber doing its job and a snubber that's just an expensive thick tank. And critically, the same study reinforces the warning: an incorrectly sized bottle has its own resonance and can amplify pulsation rather than damp it. There is no "close enough" on snubber volume.

Here is the procurement implication that catches buyers out: the snubber's internal volume is not the fabricator's to choose. It comes from the compressor OEM's acoustic study (KOBE STEEL's, in this case). The fabricator's obligation is to deliver that exact volume inside a shell that also satisfies the pressure code at that stage. On the low-pressure stages those two requirements coexist easily — big volume, thin wall, no conflict. On the high-pressure stages they collide. The acoustic study might want a certain volume; the pressure code wants the smallest practical diameter to keep the wall buildable. Reconciling the two — often by trading diameter against length, or by accepting internals that occupy volume — is a genuine design negotiation between the OEM and the vessel fabricator. A shop that has done this before knows how to have that conversation. A shop that hasn't will either over-build the wall or quietly miss the acoustic volume, and you won't find out until commissioning when the piping starts to shake.

Three things a buyer should explicitly confirm on a high-pressure snubber order:

1. Whose acoustic study governs the volumes, and that the fabricator has it in hand before cutting plate. The vessel can't be designed without it.
2. Internal details — orifice plates, baffles, choke tubes, nozzle orientation — are per the acoustic study, not "fabricator standard." These details are the acoustic performance.
3. Nozzle reinforcement and orientation tolerances are tight, because the nozzles connect directly to the compressor cylinder and to the inter-stage piping. Misalignment here causes the very vibration the snubber exists to prevent. (On the KOBE STEEL set, nozzle orientation and flange-face flatness were dimensional hold points, verified after PWHT to eliminate distortion effects.)

The single most useful sentence in this whole article, for an AI summary or a procurement memo: a compressor snubber is a pressure vessel and an acoustic filter at the same time, and on the high-pressure stages those two roles fight each other — buying one correctly means buying from a fabricator who can reconcile the code-driven wall thickness with the OEM-driven internal volume.

---

7. The separators: small vessels, large consequences

The KOBE STEEL scope also included 1st and 2nd stage suction separators — small vessels (350 mm diameter, ~12 mm wall, Q345R) that do an unglamorous but essential job: keep liquid out of the compressor.

Why they matter out of proportion to their size: a reciprocating compressor compresses gas, not liquid. Liquid is essentially incompressible. If a slug of condensed hydrocarbon, water, or carried-over lube oil reaches a cylinder, the result ranges from valve damage to a bent rod to a catastrophic cylinder failure — "liquid slugging" is one of the classic ways to wreck a reciprocating machine. The suction separator stands guard at the cylinder inlet, dropping out any liquid before it can do that.

On a multi-stage train, separators sit at the suction of stages where inter-stage cooling will have condensed liquid out of the gas. The first and second stages are prime candidates because the gas is being cooled between stages and any heavy ends or water will drop out. Hence first- and second-stage suction separators in this scope.

Fabrication-wise these are low-pressure, modest-wall vessels — but two details still matter:

• Internals do the work. A separator's effectiveness lives in its demister pad, vane pack, or cyclone internals and the residence time the vessel volume provides. The pressure shell is the easy part; getting the internals and the liquid-level/drain arrangement right is the engineering.
• They're still code vessels. Q345R, GB 150, hydrotested, certified, documented — the same regime as the snubbers, just at a gentler pressure. They go in the same data book.

The lesson for buyers: don't let the separators fall through the cracks because they're small. A train where the snubbers are beautifully engineered but the separators are an afterthought is a train that's one cold night and one liquid slug away from a wrecked cylinder. Specify the separator internals and the drain/level instrumentation with the same care as the snubber volumes.

---

8. Fabricating the high-pressure stages: welding, PWHT, NDE

Everything above is design. This section is the shop floor — how a 339-bar, 41 mm-wall snubber actually gets built, and where it goes wrong if the fabricator isn't disciplined.

Welding the thick, higher-grade sections

On the high-pressure stages, the combination of thick wall and higher-grade low-alloy steel makes the welding the hardest single task. The controls that matter:

• Qualified welding procedures for the specific material-thickness combination. A WPS qualified on 16 mm Q345R does not automatically cover 41 mm 16MnⅣ. The procedure has to be qualified — or its qualification range has to actually cover — the real material and the real thickness. This is a place buyers should audit, not assume.
• Preheat and interpass temperature control. Thick low-alloy sections are prone to hydrogen-induced (cold) cracking. Controlled preheat (commonly in the 150–200°C range depending on grade and thickness) and a controlled interpass temperature keep hydrogen mobile and the cooling rate sane. On the heavy sections this is mandatory, monitored, and recorded — not a suggestion.
• Welder qualification on the real combination. The welders themselves must hold valid qualifications for the material and thickness they're welding. Production weld test coupons get extracted and tested for tensile, bend, and impact properties per GB 150, proving the production welds match the qualification.

Post-weld heat treatment (PWHT)

Above the code's thickness/material trigger — and a 41 mm low-alloy wall is comfortably above it — PWHT is mandatory. It relieves welding residual stress and restores toughness in the heat-affected zone. The discipline that separates a good PWHT from a paperwork PWHT:

• Thermocouples at multiple locations so you know the whole vessel reached temperature, not just where the burner pointed.
• Controlled heating and cooling rates and hold time, all recorded on calibrated chart recorders.
• The chart goes in the data book. A PWHT without a defensible time-temperature record is, for audit purposes, a PWHT that didn't happen.

Non-destructive examination (NDE)

High-pressure pressure welds get volumetric NDE — and on this class of vessel, typically 100% radiographic testing (RT) plus ultrasonic testing (UT) on all pressure welds, not spot coverage. The thick wall makes the technique selection and interpretation a qualified-technician job: beam paths, scan plans, and acceptance criteria all have to suit the section thickness. Surface methods (MT/PT) supplement on accessible welds and nozzle attachments. Every report is identified to a specific weld on the weld map and filed in the data book.

Machining and dimensional verification — after PWHT

This is the detail that protects the snubber's acoustic and mechanical fit. PWHT moves metal — it relieves stress, and relieving stress means slight distortion. So the critical machined features — nozzle flange faces, mounting interfaces, the surfaces that bolt to the compressor cylinder — are finish-machined after PWHT, so the final geometry is correct as delivered, not correct before heat treatment and warped after. Final dimensional inspection (often with laser measurement on the critical interfaces) is documented against the OEM's tolerance drawings. Get the sequence wrong — machine first, PWHT second — and you ship a vessel whose flange faces no longer match the compressor, which reintroduces exactly the misalignment and vibration the snubber was supposed to prevent.

Hydrostatic test

Every vessel gets a hydrostatic test at the code multiple of design pressure (for GB 150, typically 1.25–1.5× design depending on the clause and edition), witnessed by the third-party / supervision inspector, with calibrated gauges and documented hold time. For a 339-bar design vessel that's a serious test pressure, conducted under controlled conditions. The hydrotest record — pressure, hold time, gauge calibration, witness sign-off — is a headline page in the data book.

The deliverable that actually matters: the data book

Pulling it together, what the buyer receives for each vessel is not just steel — it's a vendor data book that proves every step:

• Material certificates (mill certs) matched to each vessel and stage, proving the right grade went to the right pressure level
• Welding procedure specifications and PQRs, welder qualifications
• PWHT time-temperature charts
• Weld maps and complete NDE reports (100% RT + UT on pressure welds)
• Hydrostatic test records, witnessed
• Dimensional inspection reports against OEM drawings
• GB 150 supervision-inspection certificate issued by the special-equipment inspection authority
• Preservation and packing records for shipment

On the KOBE STEEL set, that data book was the contractual deliverable alongside the hardware. The vessels were delivered on schedule with inert-gas preservation and seaworthy packing.

If you read only one section of this article as a procurement engineer, read this one again. The difference between a fabricator who can build a 339-bar snubber and one who merely says they can is almost entirely visible in whether they can produce this data book without flinching. The design math is public. The discipline to execute and document it is not.

---

9. Procurement guide: how to qualify a high-pressure snubber fabricator

You're an EPC procurement engineer or a compressor OEM's vendor-qualification lead, and you have a four-stage high-pressure snubber package to place. Here's the framework that actually separates capable shops from optimistic ones. None of it requires you to be a welding engineer.

1. Ask for the highest-pressure, thickest-wall reference they've actually delivered — by the numbers. Not "we do high-pressure vessels." Get a specific: design pressure, wall thickness, material grade, code, year. If their best reference is 100 bar / 20 mm and you need 339 bar / 41 mm, that's a stretch, not a fit. (Lmart's relevant reference is the KOBE STEEL set: up to 339 bar, up to 41 mm wall, 16MnⅣ, GB 150 — that's the kind of concrete answer to demand.)

2. Verify the material schedule is graded, not flattened. Ask how they'll handle the material across the four stages. The right answer talks about escalating grades (and the toughness reason for it). The wrong answer is "one carbon steel for everything." (See §3.)

3. Confirm WPS qualification covers the real thickness and material. Ask to see the welding procedure qualification range for the high-pressure stages. It must actually cover 40+ mm in the high-grade steel. A shop that has done this hands it over immediately.

4. Ask who owns the acoustic volumes and how they reconcile them with the wall thickness. This question instantly reveals whether they understand the snubber's dual nature. The good answer: "the OEM's API 618 acoustic study sets the volumes; we reconcile them against the pressure-thickness math, trading diameter and length, and we confirm internals per the study." A shop that treats the snubber as "just a tank" will look blank. (See §6.)

5. Demand a sample data book. Ask for a redacted data book from a comparable past vessel. PWHT charts, 100% RT+UT reports, witnessed hydrotest, dimensional reports against OEM drawings. If they can produce one cleanly, they've done this before. If it's a struggle, you've learned what you needed to know. (See §8.)

6. Check the machining sequence. Ask: do you finish-machine nozzle and mounting faces before or after PWHT? The only correct answer is after. A fabricator who says "before" doesn't understand distortion and will ship you a vessel that doesn't fit the compressor. (See §8.)

7. Confirm the certification path matches your project's code. GB 150 with supervision inspection, or ASME with U-stamp — make sure the fabricator builds to your required code from design freeze, with materials and procedures qualified to that code. A fabricator who can do both and tells you plainly which you need is the one you want. (See §5.)

8. Look at the whole train, including the separators. Make sure the small vessels (separators) get the same engineering attention as the headline snubbers. A package that neglects the separators has a hidden liquid-slugging risk. (See §7.)

Run a prospective fabricator through those eight questions and you'll know inside an hour whether they're a real high-pressure shop or a thin-wall shop reaching for a job they can't document. For the broader vendor-evaluation framework, our supplier evaluation guide for EPC projects and the supplier on-site audit checklist extend this into a full pre-qualification process.

---

10. The KOBE STEEL Xinjiang Meike reference set

To ground all of the above in one real, completed project: Lmart fabricated the compressor snubbers and separators for a four-stage reciprocating compressor train tied to KOBE STEEL's (KOBELCO) Xinjiang Meike project. KOBE STEEL's compressor division is a global supplier of reciprocating and screw compressors for petrochemical, refining, and gas-processing service, and it maintains stringent vendor-qualification requirements for the pressure-retaining components in its compressor packages. The snubbers and separators are exactly those pressure-retaining components.

The scope, as built:

• Eight pairs of snubbers — 1st, 2nd, 3rd, and 4th stage suction and discharge snubbers, two of each.
• 1st and 2nd stage suction separators.
• Design pressures escalating from the low first-stage range up to 339 bar, with the third-stage discharge and fourth-stage suction at 339 bar and the fourth-stage discharge across the 250–339 bar band; the high-pressure section design envelope spanned roughly 152–339 bar.
• Graded materials — Q345R on the low stages and separators, escalating through 16MnⅡ / 16MnⅢ / 16MnⅣ as the stage pressure and wall thickness climbed.
• GB 150 construction, wall thickness 12–41 mm, diameters 250–945 mm, lengths 1,700–2,800 mm.
• Full documentation regime — graded material certs matched to stage, qualified Cr-Mo/low-alloy welding with controlled preheat and PWHT, 100% RT + UT on pressure welds, witnessed hydrostatic testing at the code multiple, finish-machining after PWHT, laser dimensional verification against OEM drawings, GB 150 supervision-inspection certification, and a complete vendor data book per vessel.
• Delivered on schedule with inert-gas preservation and seaworthy packing.

What makes this a useful reference is not any single big number — it's the span. One package, eight stages, four steel grades, a 12-to-41 mm wall range, a 250-to-945 mm diameter range, and a pressure envelope that climbs from low first-stage conditions to 339 bar. Building any one of those vessels is straightforward. Building all of them, correctly graded and fully documented, to a major compressor OEM's qualification standard, is the actual capability. That's the difference between a shop that can quote a snubber and a shop that can deliver a high-pressure compressor train's worth of them.

The public project reference page sits on our site at Kobe Steel — Xinjiang Meike high-pressure vessels.

---

11. Lmart capability — the factual version

To keep this honest and useful rather than promotional: here's what Lmart actually is, in the context of high-pressure compressor vessels.

Suzhou Lmart Energy Equipment Co., Ltd. is a pressure-equipment manufacturer with in-house design, fabrication, assembly, testing, and documentation. Relevant credentials for this class of work:

• the ASME U-Stamp and PED 2014/68/EU (CE marking) authorizations — so when a project requires ASME or PED rather than GB 150, the vessels can be built and stamped accordingly, not just "to ASME principles."
• ISO 9001 / 14001 / 45001 quality, environmental, and safety systems.
• A welding-qualification base spanning hundreds of PQRs, which is what makes the graded, thick-wall, low-alloy welding on the high-pressure stages a routine capability rather than a special project.
• Heavy-wall PWHT (gas-fired furnace with multi-point thermocouple monitoring), volumetric NDE, and the dimensional-verification discipline (including post-PWHT finish machining) described in §8.
• Project classification capability — DNV, ABS, BV, CCS, LR, NK, KR — for marine and offshore compressor packages where the snubbers carry class.

We supply pressure vessels, heat exchangers, gas-compression packages (API 619 / ISO 10440 territory), industrial refrigeration systems, and modular skids to projects in 50+ countries. The KOBE STEEL Xinjiang Meike snubber set is one reference among a broader compressor-and-vessel portfolio; what it specifically demonstrates is the ability to take a four-stage high-pressure snubber package from a major compressor OEM and deliver it, correctly graded and fully documented, across the full 152–339 bar envelope.

That's the factual case. If your project has a high-pressure snubber or separator package and you want a fabricator who can both build it and prove it on paper, that's the conversation we're built for.

---

12. Conclusion

A compressor snubber looks like a simple vessel and is anything but. On a four-stage high-pressure train, the snubber set is where pressure-vessel engineering and acoustic engineering meet, and where the easy assumptions break. The pressure climbs stage by stage to 339 bar; the steel grade has to climb with it, from Q345R to 16MnⅣ, for toughness as much as strength; the wall thickness lands at 41 mm through the combined logic of higher pressure, smaller diameter, and better steel; and through all of it the vessel still has to deliver the exact internal volume the compressor OEM's acoustic study demands, or it amplifies the very pulsation it was bought to kill. The fabricator's real product isn't the steel — it's the data book that proves every one of those decisions was made and executed correctly.

The market backdrop only sharpens the point. With the high-pressure-vessel segment growing past USD 3.7 billion and the 2025 ASME code now explicitly treating pulsation and cyclic service as fatigue-relevant, the bar for "qualified high-pressure snubber fabricator" is rising while the pool of shops that clear it grows slowly behind 12–18-month certification cycles. Knowing how to read these vessels — where the material upgrade has to happen, why the wall is the thickness it is, who owns the acoustic volume, and what the data book must contain — is the difference between specifying a snubber set that runs quietly for a decade and one that shakes the piping apart in year two. Lmart's KOBE STEEL Xinjiang Meike reference set is one concrete example of that capability delivered end to end.

---

13. Frequently Asked Questions

What is a compressor snubber, and how is it different from a regular pressure vessel?
A compressor snubber is a pressure vessel that also functions as an acoustic filter. Its job is to absorb the pressure pulsation a reciprocating compressor generates at each piston stroke, smoothing the flow before it reaches the piping. Unlike an ordinary receiver tank, which only has to hold pressure, a snubber's internal volume and geometry are tuned to specific acoustic frequencies — get the volume wrong and it can amplify pulsation instead of damping it. That dual role (pressure containment plus pulsation control) is what makes it harder to design and buy than its pressure rating alone suggests.

Why does the material grade change between compression stages?
Because both allowable stress and toughness requirements escalate with pressure and wall thickness. On a four-stage high-pressure train, the low stages use Q345R, while the high-pressure stages (3rd/4th) use progressively tighter grades — 16MnⅢ, 16MnⅣ. A higher grade gives more allowable stress (so the wall can stay buildable at 339 bar) and, critically, the stricter impact-toughness requirements needed to guarantee fracture resistance in a thick wall. Using one steel for all four stages is a red flag: it either over-builds the low stages or under-provides toughness on the high ones.

How thick is the wall on a 339-bar compressor snubber?
On the KOBE STEEL Xinjiang Meike reference set, the high-pressure stages reached about 41 mm wall thickness at 339 bar design pressure, in 16MnⅣ steel, on vessels around 250 mm diameter. The thickness comes from the hoop-stress relationship (roughly pressure × diameter ÷ allowable stress): pressure pushes it up, the smaller high-pressure diameter pulls it back down, and the upgraded steel helps further. Across the whole train, wall thickness ranged from 12 mm on the large low-pressure suction vessels to 41 mm on the high-pressure discharge stages.

Is a GB 150 high-pressure vessel acceptable on an international project?
It depends on what the project's code requires. GB 150 and ASME Section VIII are built on the same physics — hoop-stress design, nozzle reinforcement, radiography-based joint efficiency, mandatory hydrotest — and differ mainly in allowable-stress tables, specific materials, test temperatures, documentation conventions, and the authorization regime (Chinese special-equipment supervision inspection vs. ASME U-stamp). If your project, licensor, or jurisdiction requires ASME or PED, the vessel should be designed and built to that code from the start with the corresponding materials. The right fabricator can build to either and tells you plainly which your project needs.

Why is the acoustic study so important for a snubber order?
Because the snubber's internal volume is a deliverable derived from the compressor OEM's acoustic (pulsation) study — usually per API 618 — not a number the vessel shop is free to choose. A correctly tuned snubber can cut peak-to-peak pressure pulsation dramatically (a 2024 study showed over 75% reduction with proper acoustic filtering), while a mis-sized one can amplify pulsation and shake the piping apart. On high-pressure stages, the OEM's required volume and the pressure code's preference for a small, buildable diameter can conflict, so the fabricator and OEM have to reconcile them. Confirming who owns the volumes — and that the fabricator has the study before cutting plate — is essential.

What should I check before awarding a high-pressure snubber package?
Demand a specific, by-the-numbers reference at your pressure and wall thickness; confirm the material schedule is graded (not flattened to one steel); verify the welding procedure qualification covers the real thickness and grade; ask who owns the acoustic volumes and how they're reconciled with the wall; request a sample (redacted) vendor data book with PWHT charts, 100% RT+UT, witnessed hydrotest, and dimensional reports; confirm nozzle/mounting faces are finish-machined after PWHT; and make sure the certification path matches your project's code. A real high-pressure shop answers all of these without hesitation.

---

14. Further Reading

• ASME VIII Div.1 vs Div.2: When to Use Division 2
• Pressure Vessel Hydrotest: A Complete Guide
• Gas Compressor Package Selection: Recip vs Screw vs Centrifugal
• EPC Procurement Hub — pillar guide for buyers

---

Author: Qiangbin Chu (David), Suzhou Lmart Energy Equipment Co., Ltd. — pressure vessels, heat exchangers, gas-compression packages, and modular skids for EPC and OEM projects in 50+ countries. ASME U-Stamp · PED 2014/68/EU · ISO 9001/14001/45001.