Wet-and-Dry Paper (Waterproof Coated Abrasive)

Identity

• Names: wet-and-dry paper / wet-or-dry paper (common UK and trade usage); "Wetordry" is 3M's own trademark. That the 3M brand is the origin of every generic regional "wet-and-dry" phrase was not established by the sources consulted and is not asserted here. Also: waterproof sandpaper.
• Family/category: a coated abrasive (abrasive grains attached to a backing by one or more bond coats, not loose grit) whose backing and bond are rated for wet use with an approved liquid. Sits in the same COAT root as Emery Cloth and the Sanding Block. Wet-use capability depends on the complete construction (backing treatment and bond), not on a waterproof sheet under ordinary glue. The family is defined by that wet rating, not by mineral, color, grit range or sheet size.
• One-line: a flexible coated abrasive built so that it can be used with an approved liquid such as water. The liquid can reduce loading, heat and airborne dust; it does not eliminate exposure or guarantee compatibility with a finish, substrate or tool. Some products are also rated for dry use. Pick the exact sheet, grade system, lubricant, backing method and sequence from the abrasive maker and the finish or substrate maker.
• Explicitly NOT: Emery Cloth (natural emery on a cloth backing, a different construction, though the two are colloquially conflated in the UK market, see Lineage-misuse); a rigid sanding block or foam sanding sponge; any dry-only sandpaper, which is not waterproof and can disintegrate if used wet (see Damaging).

✦ How NOT to use it

• Not-for: routine dry stock removal where equivalent dry-only paper does the same for less; flattening or truing a reference surface.
• Dangerous:
  • Respiratory, skin and eye exposure: wet sanding reduces airborne dust; it does not eliminate exposure. Dust or slurry can contain hazardous material from the substrate or coating, including lead or chromium in older paint, wood dust, or formulation-specific contaminants. NIOSH Health Hazard Evaluation 1998-0285-2989, read in this pass, found that wet scraping and manual sanding produced much lower lead-in-air levels than the dry equivalents (task-based TWAs of 4.7 to 53 µg/m³ wet versus 29 to 160 µg/m³ dry), recommends wet methods "whenever feasible," and says wet exposures "could exceed the OSHA PEL of 50 µg/m³ if this task was performed throughout an entire 8-hour work day" and "could be a health hazard if the activity is performed 8 hours or more" [WD14]. The 3M Wetordry SDS tells users to consider the material being abraded and warns that substrate dust may be hazardous or combustible [manual, relayed] [WD10]. NIOSH's auto-body hazard-control sheet supports hazardous lead / chromium / abrasive dust and local-exhaust sanding controls [manual, relayed] [WD13]. Controls:
    • Read the coating or substrate SDS, or do a hazard assessment, before starting. Do not assume the paper's "wet" rating is a safety rating.
    • Ventilation or local capture where needed.
    • Eye protection and suitable gloves.
    • Respirator selected through the hazard assessment when it requires one, not a generic "dust mask." (3M's 314 datasheet lists a dust mask among its precautions [manual, relayed] [WD9]; that is that product's line, not a substitute for the assessment.)
    • Controlled slurry cleanup and disposal; no washing it into an uncontrolled drain.
    • Where old coatings may contain lead or chromium: lead-safe hygiene, no eating or drinking in the work area, washing hands and changing clothes before leaving (take-home contamination), and proper waste handling.
    • Dry use is not made safe by the "wet-or-dry" name; used dry, the paper makes ordinary abrasive dust like any other coated abrasive.
  • Power tools with liquid: the controlling rule is that electric tools must not be used in damp or wet locations unless approved for that purpose (OSHA electrical power-tool guidance [manual, relayed] [WD16]). The earlier line calling a pneumatic sander "the standard, sourced safe alternative" is withdrawn: pneumatic power removes one electrical source but does not make the operation safe.
    • Default: hand sanding, or a complete manufacturer-approved damp / wet system. Some paint-finishing systems use specified powered equipment damp, not flooded, and "damp" and "wet" are not interchangeable: keep the maker's exact word [manual, relayed] [WD12].
    • Electric tools: never expose an electric tool to liquid unless it is approved for that environment and used with all the electrical protection required. Keep liquid away from outlets, cords, batteries and unrelated equipment.
    • Pneumatic tools: follow the rated air pressure, the specified pad and abrasive, the PPE (eye protection, plus hearing and respiratory protection as the assessment requires), the air-supply and hose-condition checks and the slurry-containment instructions for that tool. This entry does not cite a tool maker's manual and names no specific model.
    • Shocks and tool failure are not the only hazards: slurry on floors, wet hands on controls and flying fragments from a loaded or torn disc are process hazards that the maker's instructions cover.
• Damaging:
  • Sanding through a finish coat to the substrate. Practitioner threads show how often it happens (titles such as "Did I wet sand through my clear coat?"), and trade guidance warns that fine grits can still remove a lot of clear with pressure. A gauge cannot reliably prove how much clear is left (see Matching). Test on a panel, mask edges and body lines, back with a suitable pad, and stop early.
  • Lubricant and cure mistakes. Water in defects or exposed wood, oil on a surface meant for coating, and sanding a coat that is not fully cured can all ruin the work.
  • Using a non-waterproof, dry-only paper as if it were this tool. A cellulose backing swells and tears when wet and a hide-glue-class bond releases grit (Empire Abrasives, EngineerFix, Riken MT; trade sources, URLs not recorded). This is a materials-mechanism claim, not an anecdote.
  • Assuming any paper labeled "wet/dry" is interchangeable. Backing weight, bond, mineral and rating differ by product; read the datasheet.
• Beginner-trap:
  • Applying a universal rule. "Always keep it wet," "never skip a grit," or "use this sequence" are product- and process-specific. Use wet or dry as the product specifies, and use the named manufacturer's sequence. Automotive sources observe that scratches left by a skipped step can reappear after buffing, so check scratch removal between steps rather than trusting the surface mid-progression.
  • Hand-pressure edge-rounding (inference by analogy, not directly sourced for this tool): unbacked flexible abrasive under bare fingers rounds edges.
• Lineage-misuse:
  • Treating it as identical to Emery Cloth. The UK colloquial conflation: different backing (paper versus cloth) and typically a different mineral. Cloth's fold-and-wrap durability is why Emery Cloth stays the default for hand-wrapping tubular metal stock.
  • Treating it exactly like dry-only sandpaper (not managing loading or slurry, running it until it clogs). The loading mechanism is documented in wet-sanding guides; the "dry mentality" explanation is this entry's inference.

Correct use

Start with: use wet or dry as the product and process specify. Some of these sheets are rated for dry use as well, and the manufacturer says which liquid and how much.

• Read before sanding. The abrasive product information and the coating or substrate SDS. Confirm the finish is fully cured, test on an inconspicuous area or a test panel, and test the lubricant for compatibility with the finish you are going to put on next. Oil can contaminate a surface meant for coating.
• Lubricant. If wet, use only the liquid the maker approves, in the amount it specifies. A current 3M consumer sheet permits water or cutting oil; automotive systems specify water and sometimes "damp, not wet" [manual, relayed] [WD8][WD12]. Water can enter defects or exposed wood.
• Sequence. Follow the named manufacturer's progression. A current 3M consumer instruction says to skip no more than one grit, not that a grade may never be skipped; other product systems set their own sequences [manual, relayed] [WD8]. Verify scratch removal between steps under clean, suitable lighting. Scratches left by a skipped step can reappear after buffing (an automotive-trade observation).
• Backing and edges. Use a suitable backing block or interface pad where flatness matters; keep the paper clean and rinse or replace it as the maker directs rather than letting loaded grit cut random deep scratches. Mask or avoid edges and body lines. The edge-rounding risk of unbacked flexible abrasive is an inference by analogy from the Sanding Sponge entry, not a sourced finding for this tool.
• Thickness. See the gauge correction in Matching. Stop before exposing substrate or a lower coating layer.
• Slurry. Wet methods produce contaminated slurry that needs controlled cleanup and disposal (see Dangerous). Deep operation belongs to Reed.

✦ Matching

• Does: controlled abrasion with a product rated for wet use. It reduces loading and airborne dust when used wet, and can level or refine compatible coatings and materials. Suitability depends on the exact abrasive product, mineral, backing weight, grade system, lubricant, finish cure, coating chemistry and thickness, substrate, surface geometry and downstream contamination limits.
• Does not guarantee: dust-free work, safe exposure, no breakthrough, a glass-smooth result, or compatibility with every finish, metal, wood, crystal, plating, lubricant or power tool.
• Match by the product datasheet and the finish or substrate maker's process, not by the family label. Examples are product-and-process pairs, not blanket approvals:
  • Automotive clearcoat: a named manufacturer process, for example 3M's collision-repair procedure for P1500 / P2000 and finer steps with specified tools and pads [manual, relayed] [WD12]. Practitioner forums are texture only.
  • Lutherie and wood coatings: the finish maker's guidance on cure time and leveling, with a test panel first.
  • Watch pivots, plating, blades: material-specific procedures only.
• Explicit exclusions: unknown coatings; delicate plating; optical coatings; coated or plastic watch crystals and unsupported crystal materials (water, oil or abrasion can damage them); and catalyzed coatings, whose chemistry-specific hazards must come from that coating's SDS. Where the coating or its cure is unknown, do not sand wet on the strength of this entry.
• Selection fields needed: abrasive mineral, backing weight / material, bond, grade system, grit / particle size, wet / dry rating, approved lubricant, hand / machine rating, substrate / finish compatibility, and the intended step (stock removal, defect leveling, scratch refinement, polishing preparation).
• Paint-thickness gauge, corrected. A conventional gauge can identify unusually thin or repainted areas and establish relative total film thickness, but it does not prove remaining clearcoat. Treat unknown history, edges, body lines and previously corrected areas conservatively; follow the coating or vehicle repair procedure or refer the work to a qualified refinisher. Common magnetic and eddy-current gauges measure total coating thickness above metal and do not tell primer, base coat, clearcoat, repaint, filler or contamination apart. DeFelsko says the user must estimate individual layers unless before-and-after measurements exist, and specialized ultrasonic or destructive instruments may separate some layers within geometry and thickness limits [manual, relayed] [WD11]. A total-thickness reading can give false confidence and lead to irreversible clearcoat breakthrough. The earlier "check remaining clearcoat depth with a paint-depth gauge" prerequisite is withdrawn.
• Kit-fit, pairs with: the approved lubricant; a suitable backing block or interface pad; buffing and polishing compounds as a next stage per the finish maker; Crocus Cloth for a final ultra-fine metal polish (an internal cross-reference).
• Kit-fit, substitutes for: dry sandpaper where heat, dust or clogging limits the job, only when the sheet's rating, the lubricant and the finish all allow. Does NOT substitute for Emery Cloth on metal deburring around round stock, or for a rigid flattening tool: like every coated abrasive it conditions and polishes, it does not true a surface.
• Not-for (project-keyed): routine dry stock removal where a cheaper dry-only paper does the same; any finished surface where thickness, cure or coating chemistry is unknown; use on an electric sander not approved for wet use (see Dangerous).
• Access: physically simple for hand use, but consequential finish work is not automatically beginner-safe. The risk is often irreversible damage (breakthrough, contamination, wrong lubricant) rather than tool complexity, and the exposure controls in Dangerous apply to every user.
• Revival: a maker reaches for it when a cured finish has to look genuinely smooth before a final buff, the difference between "sanded" and "polished." The result comes from the right grade, system, lubricant and scratch-removal checks, not from rinsing slurry alone.

Anatomy

• Backing: paper (or other material) treated to withstand an approved liquid. Paper backing weight matters: the flexible A-weight and stronger C-weight behave differently in conformability and suitability for leveling, so do not treat all sheets as equal.
• Bond: a make and size coat system for wet use. Hide-glue-class bonds in dry-only papers release grit or break down if wetted.
• Abrasive mineral: silicon carbide and aluminum oxide are both common, and a given product can be either. Per the review of the manufacturer pages [manual, relayed] [WD8][WD9]: 3M Wetordry 431Q is silicon carbide and is sold as coarse as 60 grit; 3M 314 is aluminum oxide on latex-impregnated waterproof paper, P60 to P1200. Silicon carbide is sharp and friable, aluminum oxide tougher and longer wearing; those differences affect selection and do not define wet versus dry capability. "Never natural emery" is not established by the sources consulted and is withdrawn.
• Color, sheet size, grit range, backing weight: product-specific. Black or dark grey, 9"×11" sheets and 400 to 3000-grit ranges are market cues for some automotive products, not what makes a sheet wet-and-dry; buyers should not reject a valid product by one of them.
• Grading: keep the systems distinct. FEPA P-grades, ANSI / CAMI grades, micron grades and proprietary structured-abrasive grades (such as 3M's) are different systems, and same-looking numbers are not always equivalent. Norton's technical data compares them rather than equating them [manual, relayed] [WD15]. Write the prefix and system exactly (P1500 versus 1500), do not translate by number alone, and use the abrasive and finish maker's progression.

Variations (living)

• Automotive-oriented sheets: sold in fine-finishing grades for clearcoat wet-sanding; sheet size varies by product.
• Hand-cut offcuts for fine work: small pieces or dowel-mounted strips for pivot-polishing sticks (see Emery Stick for an internal outline). Cut-up sheets carry the same wet and dry rating as their parent product.
• Later waterproof formats: waterproof foam sanding products (see Sanding Sponge) and film-backed micro-abrasives are later waterproof-abrasive formats, separate products with their own datasheets, not variants of this entry and not established descendants of it.

Trades

Each is a trade where wet sanding is used, and none is blanket approval of the family for that work.

• Automotive refinishing / bodywork: clearcoat wet-sanding and paint correction under a named manufacturer process. The richest practitioner record.
• Lutherie: finish leveling on cured nitrocellulose / varnish per the finish maker.
• Watchmaking / horology: pivot polishing and some crystal work with exclusions above.
• Knife-making / metalworking: blade polishing progressions.
• Woodworking finishing: leveling cured film finishes only where the finish maker's cure, lubricant and test-panel guidance allow. Catalyzed finishes carry coating-specific hazards.

Lineage

• Ancestors: dry-only coated abrasives, glass paper (commercialized by John Oakey & Sons, founded 1833, per Emery Cloth's own history) and garnet/flint paper, used strictly dry. Emery Cloth is a parallel, not ancestral, branch. A trade source (FreshScientific, secondary) says traditional glasspaper and garnet paper were "designed for dry use until the 1921 invention of waterproof wet-and-dry sandpaper"; this is consistent with, not independent confirmation of, the Okie story below.
• The mineral and the backing/bond are separate inventions. Silicon carbide was invented by Edward Goodrich Acheson in 1891 (Monongahela, PA) while attempting to synthesize diamond, patented as US Patent 492,767, granted February 28, 1893; the Carborundum Company was founded September 21, 1891. That is mineral history. It must not imply all waterproof abrasive paper uses silicon carbide: aluminum-oxide products exist (see Anatomy).
• The waterproof backing and bond. Francis G. Okie (1880–1975), a Philadelphia printing-ink manufacturer and not a 3M employee at the time, is the credited inventor. The Minnesota Inventors Hall of Fame page, read directly in this pass, supports his dates, the 1919 Saturday Evening Post article on finishing-trade workers dying of lead poisoning or silicosis as his motive, and first sales to automakers in May 1921 [WD1]. 3M's company history supports Okie selling patented waterproof sandpaper to 3M, joining the company in 1921, and early automotive use [manual, relayed] [WD2]. The Hall of Fame account has Okie hand-mixing a waterproof adhesive, coating paper, sprinkling on garnet and testing it, then writing to Minnesota Mining & Manufacturing for mineral supply; 3M contracted him and bought his patent rights. The commercial product was branded "Wetordry."
• Patent record, corrected. The earlier version of this entry listed US1484759A and US1581657A as two unread candidates. Per the review of the patent records [manual, relayed] (the pages could not be fetched in this pass): US1484759A is by Frank Joseph Crupi, assigned to Herman Behr & Co., and is not an Okie / 3M patent, so it is removed [WD6]. US1581657A names Francis Gurney Okie, is assigned to Minnesota Mining and Manufacturing, and claims a stable waterproof flexible abrasive article and processes for water abrading, with a 1922 priority claim, a 1923 filing and 1926 publication [WD5]. Related Okie patents the reviewer names: US1565027 / US1565028 and US1635800 / US1635801. The 1921 commercial introduction and the later patent filing and publication dates are different events, and this entry does not claim a single "original patent" because the priority chain and claims have not been fully traced.
• Descendants / later formats: wet-sanding technique across several trades (see Trades). Film-backed micro-abrasives and waterproof foam sanding products are later waterproof-abrasive formats; functional similarity does not establish direct design or patent descent from Okie's paper, and no source establishing descent was found. The earlier entry's "direct descendants that borrowed this invention" is withdrawn.
• Contemporaries: Emery Cloth (cloth-backed, natural emery, a genuinely different construction); Crocus Cloth (a non-bonded ultra-fine polish; the corpus's own entry describes it sequenced after wet/dry paper in a blade-polish progression, which is an internal cross-reference, not an independent source).

History (Petroski "form follows failure" lens)

• The failure this tool addresses, as its inventor's story tells it. Two problems drove the invention, both from the Hall of Fame's account of Okie: dry sandpaper's rapid clogging/loading, which cut its useful life, and the health hazard of airborne dust in finishing trades, with the Saturday Evening Post article on lead poisoning and silicosis as the trigger [WD1]. Mitigation attempts of the day (ventilators, cheesecloth over the face) are named as inadequate.
• Historical wording versus modern advice. The Hall of Fame page says that when waterproof sandpaper reached the market "it eliminated this dust hazard" [WD1]. That is historical promotional language, not exposure-control guidance. 3M's own history describes wet sanding as keeping lead-bearing paint dust "at bay," which supports reduction, not elimination [manual, relayed] [WD2]. Wet sanding reduces airborne dust; it does not make sanding exposure-free (see Dangerous).
• Automotive trade sources describe the opposite failure from the user's side. Several automotive trade guides (Dr. Beasley's, eQualle, Titan Coatings) say ordinary dry-only sandpaper will soften and fall apart if used wet. These are trade sources (URLs not recorded in the original run), and are practitioner texture, not primary evidence.
• 3M and corundum. Emery Cloth's own history documents 3M's 1902 founding as a failed corundum-mining venture that pivoted to abrasives by 1916. This is an internal cross-reference, five years before the 1921 launch.

Human side

• No single named, credentialed authority for this tool was located, unlike sibling entries with Flexner- or Sellers-tier voices; flagged honestly. The human-side material below is practitioner forum and trade-blog texture, and none of it is primary evidence or a safety authority.
• Automotive refinishing / bodywork. Practitioner forums (Auto Geek Online threads such as "Wet sanding with 3000 grit?," "Wet Sanding: What did I do wrong?," "wet sanding and reclearing") and trade guides (eQualle, Titan Coatings, Painted OEM Parts) describe clearcoat wet-sanding progressions. One commonly quoted sequence is 1500, 2000, (2500), 3000 before compounding, with 1500 called a "safe starting point unless orange peel is extreme." Treat these as forum anecdote, not a procedure: the grade system (P-grade, CAMI/ANSI, or proprietary) is not stated in those threads, and the actual progression is the abrasive and paint-finishing maker's. For comparison, 3M's collision-repair procedure specifies product-specific P1500 / P2000, 3000 and optional 8000 steps with specified tools and pads, and damp-not-wet use [manual, relayed] [WD12].
• Lutherie. MIMF and Lutherie Academy describe wet-sanding for finish leveling on cured nitrocellulose / varnish guitar finishes and mention "waterproof sandpaper (usually black)." The color is a trade cue, not a family property. Follow the finish maker's cure and leveling guidance.
• Watchmaking / horology. NAWCC forum threads describe wet-or-dry automotive-grade paper as a cheap substitute for horology-specific abrasive in pivot polishing. Crystals are a special case: coatings and some plastic or coated crystals can be damaged (see Matching exclusions). The corpus's Emery Stick entry notes DIY pivot-polishing sticks as an internal cross-reference.
• Knife-making / metalworking. Forum posts name "3M Wet or Dry" in full blade-polishing grit sequences, framing the wet benefit as lower heat and friction. This is forum texture; the corpus's Crocus Cloth entry is an internal cross-reference, not corroboration.
• Regional naming note. The UK conflation of "wet-and-dry" with Emery Cloth, flagged in that entry and seen again in a bit-tech.net thread and a UK Airgun Forums post (April 2024), is a market-listing observation. The bit-tech.net thread is the same source Emery Cloth already cites.

Proof of existence

• The founding invention (Okie, 3M, "Wetordry," first sales to automakers in May 1921) is supported by two external sources: the Minnesota Inventors Hall of Fame record, read in this pass [WD1], and 3M's company history [manual, relayed] [WD2]. The corpus's separately-researched Sanding Block entry gives the same story, but it overlaps in its 3M / FundingUniverse material and is a cross-reference, not independent corroboration.
• Silicon-carbide mineral patent: US Patent 492,767, Acheson, granted February 28, 1893 (cited in the original run to HDP Patents Blog, EBSCO Research Starters and encyclopedic sources; not re-fetched in this pass) [WD4].
• Waterproof-abrasive patent: US1581657A, Okie / Minnesota Mining and Manufacturing, a waterproof flexible abrasive article and processes for water abrading, 1922 priority, 1923 filing, 1926 publication [manual, relayed] [WD5]. This resolves the earlier "named gap" in the form it was stated. It does not establish a single "original patent."
• Museum record: the Smithsonian NMAH holds "3M Sandpaper" (object ID 2014.0063.01, nmah_1461587), a 1923 3M sandpaper package, gift of 3M Company; the object record does not confirm it as silicon-carbide or Wetordry-branded stock. A companion object (nmah_1461588) is similarly unconfirmed. "3M Scotch WetorDry Masking Tape" (nmah_1691712, a 1925 product that borrowed the name) is not the sandpaper. No Science Museum Group object for wet-and-dry paper was located; the Watt-workshop sandpaper pieces predate waterproof coated abrasive entirely.

References

One exact URL per source where a URL is known. "Reached" means read directly in this pass; "relayed" means the claim is carried from the reviewer's description of the page.

Practitioner and trade texture from the original run (not primary evidence; URL not recorded in the original run, none supplied here): Auto Geek Online threads (#111407, #110256, #25286); Sawmill Creek "Wet sanding questions" (#245108); Dr. Beasley's, eQualle, Titan Coatings and Painted OEM Parts clearcoat guides; MIMF and Lutherie Academy; NAWCC Message Board and WatchUSeek; BladeForums; Sail Magazine "Sandpapers and Sanding"; DT Online "Glasspaper"; FreshScientific; Empire Abrasives, EngineerFix and Riken MT on non-waterproof paper; bit-tech.net "Wet n Dry paper vs Emery Cloth??" (same source cited in Emery Cloth); UK Airgun Forums (April 2024). The corpus's earlier NIOSH citations (doc 99-113; HHE 2023-0028-3396) are not carried as support in this pass.

Internal cross-references, not evidence: Emery Cloth, Crocus Cloth, Emery Stick, Sanding Sponge, Sanding Block (Hand Sanding). Museum records: Smithsonian NMAH 2014.0063.01 (nmah_1461587), nmah_1461588 and nmah_1691712 (URLs not recorded).

The reviewer's corrections (10)

• BLOCKER — 1. Replace “eliminates dust” with exposure-control language — Exact claim: The one-line says the product was invented “to stop dry-sanding dust from being inhaled”; History says it “eliminated this dust hazard”; Revival says rinsing slurry is what makes the finish achievable.
• BLOCKER — 2. Correct the paint-thickness-gauge advice — Exact claim: “Check remaining clearcoat depth with a paint-depth gauge before starting,” repeated as a prerequisite for automotive clearcoat work.
• BLOCKER — 3. Narrow universal finish, material, and trade matching — Exact claim: “Any coated-abrasive task where dry sanding would either overheat the work or throw an unacceptable amount of airborne dust”; “any woodworking finish”; and broad approval for automotive clearcoat, nitrocellulose/varnish, catalyzed finishes, watch pivots/crystals, plating, and knife blades.
• BLOCKER — 4. Make wet-power-tool controls complete and model-specific — Exact claim: A pneumatic sander is presented as “the standard, sourced safe alternative” to an unrated corded sander.
• SHOULD-FIX — 5. Resolve the patent attribution correctly — Exact claim: “A genuine, honestly-flagged patent gap” names US1484759A and US1581657A as two plausible candidates for Okie’s original waterproof-binder invention, saying neither was read.
• SHOULD-FIX — 6. Define the family without universal mineral, color, or format claims — Exact claim: Wet-and-dry paper is “usually silicon carbide, sometimes aluminum oxide, never natural emery,” “commonly black,” generally 400–3000+, and automotive sheets are “typically 9×11.”
• SHOULD-FIX — 7. Separate CAMI/ANSI, FEPA P-grade, micron, and proprietary grades — Exact claim: “Standard CAMI/FEPA numeric grit,” followed by unqualified sequences such as 1500 → 2000 → 3000.
• SHOULD-FIX — 8. Remove universal technique rules — Exact claim: “Keep the surface lubricated with water or a light oil throughout”; “work a full grit progression in sequence without skipping steps”; and “rinsing the slurry away ... is what makes that level of finish achievable by hand.”
• SHOULD-FIX — 9. Remove unsupported causal descendants and internal corroboration claims — Exact claim: Film-backed micro-abrasives and waterproof sanding sponges are “direct” descendants that “borrowed” this invention; two separate corpus entries are called independent corroboration.
• SHOULD-FIX — 10. Repair references and provenance — Exact claim: The References section uses bare domains, grouped unnamed sources, thread titles without URLs, and “multiple independent trade/how-to sources”; Provenance exposes internal model/research passes workflow.