Mesh Density, Thickness and HG/T 21618 Basics

Knitted demister pads are specified by more than “stainless mesh.” A purchase order that names only an alloy and an outside diameter leaves key construction parameters undefined. Lock wire diameter, pad density and pad thickness early, together with material, crimp, layering, grids and support details relevant to the route.

Wire diameter

Worth knowing before you write a number: HG/T 21618 tabulates density, specific surface area and voidage for its mesh types, but it does not fix a wire diameter. The sourced reference points come from the related drawer-type standard, HG/T 21586-1998 Annex A, which specifies round wire of Ø0.19 mm for its high-efficiency style and Ø0.23 mm for both its standard style and its 高穿透型 style (described here as the high-permeability style), alongside flat wire of 0.1 × 0.28 mm and 0.1 × 0.4 mm. Retaining the Chinese style name avoids confusing it with proprietary “high-capacity” constructions. Finer and coarser wire is knitted outside those styles, so treat these as orientation rather than limits — and note that flat-wire cross-section figures are sometimes misquoted as if they were round-wire diameters. For otherwise comparable material and construction, finer wire provides more surface area per unit mass; coarser wire can improve handling robustness, but fouling and hydraulic behaviour still depend on the complete mesh construction.

Two consequences are worth stating plainly:

• Wire diameter and corrosion life are directly linked. There is no separate corrosion allowance on a 0.19 mm wire. Even a small uniform radial metal-loss rate can consume a large fraction of its radius; see the explicitly illustrative arithmetic in demister pad materials selection.
• Knitted pads are not specified by mesh count. Mesh count describes woven cloth with a defined aperture, and a knitted structure has no equivalent aperture. The mesh / micron converter on this site applies to square-weave woven cloth only — ASTM E2016 and ISO 9044 cover industrial woven wire cloth within their stated scopes, and ASTM E2814 covers Dutch-weave filter cloth. Applying any of them to a knitted demister pad is a category error, and asking a fabricator for “60 mesh demister” leaves the knitted construction undefined.

Always pair wire diameter with density. The same visual “mesh” can hide very different structures.

Density (kg/m³)

Density here means mesh mass per unit of pad volume as installed — after the knitted material has been crimped, layered and compressed into its grids. For sourced reference points, the four mesh styles tabulated in HG/T 21618-1998 clause 4.0.2 run from 128 to 186 kg/m³, and the three styles in HG/T 21586-1998 Annex A run from 98 to 182 kg/m³. These figures apply to the listed constructions and are orientation, not universal limits.

One note under that table changes the comparison: HG/T 21618 states its tabulated values were derived for a material density of 7930 kg/m³ and requires them to be corrected for any other material. A polymer or nickel-alloy pad of the same construction does not carry the same kg/m³ figure, so a density target copied across materials is not the same specification.

For otherwise comparable wire, crimp, layering, thickness and material, raising density generally increases surface area and pressure drop and can reduce fouling tolerance. A more open construction can favour drainage and resistance to plugging, but density by itself does not establish capacity or efficiency. In fouling service, increasing density without checking solids and liquid loading can accelerate plugging and raise pressure drop. See demister fouling and clogging.

Two practical cautions on density figures:

• Installed versus free state. A quoted density only means something against a stated installed thickness. The same knitted material compressed harder into a thinner grid gives a higher density and a different pressure drop.
• Measured density on a used pad is not the original specification. Deposits add mass, compression reduces volume and corrosion removes metal. Weigh a cleaned sample, state whether it was cleaned, and treat the result as an estimate. The method and its traps are covered in mist eliminator replacement without drawings.

Thickness

HG/T 21618-1998 clause 5.0.1 standardises the mesh layer at 100 mm and 150 mm, and its selection guidance treats 150 mm as the usual choice, with 100 mm where the separation requirement is less demanding. Thicker builds exist for difficult mist duty and staged designs, outside the standard’s two sizes. Thickness interacts with velocity: a thick pad operated above its flooding margin still re-entrains, because the limit is the velocity at which collected liquid can no longer drain against the gas flow. Adding thickness to fix carryover that is caused by excessive velocity does not work.

Available height above the support ring is a basic constraint, and it must accommodate the mesh layer plus the top and bottom grids. Confirm it before selecting a thickness.

Where pressure drop has little margin to give — vacuum service being the clearest case — thickness and density have to be chosen against that limit rather than maximised for capture. See vacuum tower demisters.

How the three parameters interact

This is the part that a single “density” figure conceals.

Density is mass per volume, so a given density can be produced from fine wire or from coarse wire. For otherwise comparable material, crimp, layering and thickness, finer wire at a given density provides more wire surface and changes the passage geometry; this can improve fine-mist interception while increasing pressure drop and fouling sensitivity. Coarser wire changes those trade-offs in the other direction, but neither result should be treated as a performance guarantee without the complete construction and operating point.

So density alone does not define performance, and neither does wire diameter alone. The specification is the combination — wire diameter, density, installed thickness, plus how the material is crimped and layered. Suppliers quoting against only one or two of these are not quoting the same product as each other.

Crimp and layer orientation belong in the same discussion. Knitted mesh is crimped and stacked in layers, and the orientation of that stack relative to gas flow and to gravity affects drainage and re-entrainment. For a replacement, record the orientation of the old pad and its sections before removal; for a new design, agree it rather than leaving it to the shop floor.

HG/T 21618 context

China’s chemical industry standard HG/T 21618-1998, Mesh Demister, provides structure, marking and common dimensional practice for the constructions within its scope. The published volume carries an effective date of 1 March 1999. Its stated scope is separating entrained droplets larger than 3–5 μm in vertical cylindrical vessels, over DN300–DN5200 overall — but the range splits by installation direction: units installed from above run DN300 to DN5200, units installed from below run DN700 to DN4600. The 1998 revision also added a process-calculation annex for selection. The related standard for drawer-type (withdrawable) units is HG/T 21586-1998, which covers DN300–DN5000 at atmospheric pressure, with stainless steel limited to 400 °C and carbon steel to 300 °C. China’s Ministry of Industry and Information Technology listed HG/T 21586-1998 for revision in its 2024 industry-standard review; the 1998 text remains the cited edition until a replacement is formally published.

Used properly, these standards provide a structural, fabrication and preliminary-selection reference. HG/T 21618 includes process-calculation guidance, but it does not replace the complete project design basis, applicable vessel code, validated duty data or project specification. Its use does not, by itself, guarantee separation efficiency for a particular service; final sizing and any performance warranty must be tied to the frozen operating envelope and acceptance criteria.

Verifying what you received

For a delivered pad, two checks need no laboratory:

1. Wire diameter with a micrometer, on several strands from different areas.
2. Apparent density, by weighing a sample of known installed dimensions and dividing mass by volume.

Both should be reconcilable with the agreed specification. Where these measurements are contractual, they belong in the quotation or PI as defined measurement points with acceptance limits, so that the dimensional inspection record delivered with the shipment is checked against an agreed basis rather than against an assumption.

What to put on the RFQ

• Target density and thickness, or the duty and a request to propose them
• Wire diameter and material — the alloy grade for metal, or the polymer type with raw material producer, commercial designation and temperature limits for monofilament products
• Vessel inside diameter, manway clear opening and segmentation needs
• Support grid, tie-wire and hold-down requirements, including material
• Layer orientation and section arrangement where it matters
• Documents to be included in the quoted supply and listed in the PI: the ordinary material or batch traceability record, a dimensional inspection record against agreed measurement points and acceptance limits, pre-shipment product and packing photographs, and the packing list

A fuller field list, including the commercial and confidentiality items, is in how to specify a demister RFQ.

Common mistakes

• Ordering by alloy and diameter only, and leaving density, wire diameter and thickness to the fabricator.
• Asking for a “mesh count” on a knitted pad.
• Comparing quotations on density alone when wire diameters differ.
• Increasing density to cure carryover that is caused by velocity, distribution or level control.
• Quoting a density without the installed thickness it refers to.
• Treating a measured density from a fouled or compressed used pad as the original specification.
• Assuming HG/T 21618 conformity is a performance guarantee for your duty.

Reference status note: why you may not find HG/T 21618 on the usual standards platform

This absence can create a document-control question, so the evidence boundary should be explicit.

Searches performed on 26 August 2026 did not return HG/T 21618 on China’s national standards public service platform or the industry standards service platform. That absence is not evidence that the standard is fake or withdrawn, but neither does it establish why those databases omit it. The volume itself identifies the approving body, drafting units, effective date and the National Editorial Centre for Chemical Engineering Construction Standards; those facts should not be extended into an unsupported claim about another database’s inclusion rules.

For text verification, use the published volume. For current status, check Ministry of Industry and Information Technology publication, review and withdrawal announcements and use the National Digital Standards Library’s validity-tracking service where a record is available. Commercial third-party standards databases can be a starting point but are not status authority.

There is a concrete example of why that matters. Several commercial database entries list an implementation date of 1 June 2000 for HG/T 21618-1998. The published volume states 1 March 1999. For the text and effective date printed in that edition, use the volume; treat any database disagreement as an item to resolve rather than as proof of the standard’s current status.

One specific piece of misinformation circulates and should be dismissed: GB/T 36803-2018 cannot supersede HG/T 21618 because it concerns the determination of volatile organic compound emissions from adhesives by a bag method and has nothing to do with demisters. In the Ministry of Industry and Information Technology announcements checked through 26 August 2026, no formally published replacement for HG/T 21618-1998 was identified.

The 1998 date raises a fair status question. HG/T 21618 did not appear in the continue-valid, revise or withdraw attachments to the Ministry’s 2024 review, and no formally published replacement was identified in the searches recorded above. The controlled position is therefore to keep the 1998 text as the current cited edition while monitoring official notices, not to infer either validity or withdrawal from a missing database record. Use the standard for its structural, dimensional and preliminary-calculation content, not as a substitute for a complete project design basis or as a performance guarantee.