Pool filter cross-reference methodology

Pool Filter Lab turns manufacturer catalog tables into searchable cartridge records. We preserve source disagreements because a clean-looking answer is not useful if it hides a fitment problem.

Source editions

The current build uses these documents, retrieved on 2026-07-29:

• 2026 Unicel Replacement Guide, including its series tables, OEM sections and replacement-measurement instructions.
• Pleatco Catalog & Cross Reference Guide, an undated catalog edition whose supplied PDF metadata records creation on 2008-03-21.
• OnCore OnPoint Cross Reference Guide, V2.0, updated January 2025.

We use manufacturer names and part numbers for identification. Their appearance does not imply sponsorship, endorsement or affiliation.

Parsing and joining

The build runs offline against text extracted from the supplied PDFs.

Unicel series rows provide the canonical part number, series, dimensions, end caps, pleats, media, area, notes and fitment text. Unicel OEM sections add brand, model and OEM-number relationships.

Pleatco product rows provide a Pleatco number, independent dimensions, end-cap fields, media, area, description, OEM references, and Unicel and Filbur cross-references when present. Its Unicel-to-Pleatco mapping section supplies additional alias edges.

OnCore’s chart supplies OnCore, Pleatco, Unicel and Filbur number relationships. Records are joined on the Unicel number. Arrays are deduplicated and sorted so the same inputs produce stable output.

The current build parsed 1,632 Unicel rows, 1,333 Pleatco rows and 262 OnCore rows, then emitted 528 cartridge records and 369 brand groupings.

Source priority

When both Unicel and Pleatco provide a dimension or scalar field, the main displayed value comes from Unicel. This is a source-priority rule, not a claim that Unicel is always correct.

Both source values remain attached to the field. We never average a disagreement.

This creates a deliberate Unicel-primary bias. It keeps display behavior consistent, but a user should inspect the source values and physical cartridge whenever a conflict appears.

Validation rules

The build applies these record checks:

• Outside diameter or length is flagged when Unicel and Pleatco differ by more than 1/16 inch.
• Square footage is flagged when the sources differ by 2 sq ft or more.
• Product-looking lines that fail the expected table shape are added to the rejection report.
• Cross-reference values pointing to a Unicel number absent from the parsed series are listed as orphans.
• Aliases that resolve to more than one cartridge are kept out of the direct alias map and listed as collisions.
• Known records are checked for expected dimensions, aliases, pleats and fitments before the build passes.

The current report passes its acceptance checks. It contains 489 cartridges with outside diameter and length, 471 with both pleat count and square footage, 388 with a Pleatco or Filbur alias, 11 parse rejects, 46 orphans and 51 alias collisions.

Verification tiers

Each cartridge carries a source-count tier:

• verified-3: the record is represented in at least three parsed sources.
• verified-2: the record is represented in two parsed sources.
• single-source: the record is represented in one parsed source.

In this build, 131 records are verified-3, 203 are verified-2 and 194 are single-source.

The label describes source presence, not laboratory testing, physical inspection or a guarantee of fit. An alias edge also carries its own evidence level: a Unicel-to-Pleatco pair independently present in both the Pleatco and OnCore inputs receives the higher alias evidence level.

Conflict flags

The validation report contains 236 field-level conflicts across 164 cartridges. They comprise 74 outside-diameter flags, 124 length flags and 38 square-foot flags.

A flag means the parsed source values crossed a stated threshold. It can reflect a catalog difference, a variant grouped under a cross-reference, or a source/table interpretation that needs review. The flag does not decide which value fits a particular housing.

Conflict pages show the values side by side. Users should confirm the printed part number, housing model, physical dimensions and end-cap connections.

Plausibility and rejection handling

The parser accepts rows only when the expected catalog fields align. Data-shaped failures are not silently discarded: the report records the source, line, text and rejection reason. This makes a suspect omission inspectable without forcing malformed text into a cartridge record.

Plausibility checking does not reconstruct filter-media geometry or prove performance. Pleat count, media weight, area and dimensions are kept as related evidence.

Limitations

• The canonical join and primary values are Unicel-based.
• Some cartridges are present in only one parsed source.
• Filbur numbers come from the Pleatco and OnCore cross-reference documents; this build has no separately parsed Filbur catalog.
• The Pleatco PDF does not state an edition date on its visible title page; we identify the supplied edition by title and PDF creation metadata.
• Source guides can change, and a newer catalog may revise a number, fitment or dimension.
• Text extraction can split or merge table cells. Rejects, conflicts, orphans and collisions expose known uncertainty but cannot prove that every parsed row is correct.
• Pool Filter Lab does not physically test cartridges or certify manufacturer claims.

Corrections

If a record looks wrong, use the correction form. Include the page, field, value you believe is wrong, the cartridge or housing number, and a source image or document reference when available.

We review the cited source row, reproduce the relevant parse, and update the data or editorial explanation when the evidence supports a change.