Containment self-weights: Legrand Swifts, Unitrunk Speedway, Cablofil catalogues. Cleats: Ellis Patents datasheets (Vulcan+ DS03E iss. 06, Atlas DS13E iss. 03, Emperor 07/2026). Cables: 142 in the shared store, from Prysmian and supplier datasheets. SWA to BS 5467 from the Eland Cables dimensions tables, with the 5-core 70 and 95mm² sizes from the Cleveland Cable BS 5467 datasheet (the Eland table stops at 50mm²). Singles: 6491X PVC and 6491B LSZH, 15 sizes each from 2.5 to 300mm², from the Cleveland Cable datasheets. Cat 6A from Brand-Rex and Leviton datasheets, AV and control from Belden technical data. Rod capacities: BS 8519:2020 Table E.2.

18/09/2026 (2.8.1): a sample was added to the list of typical arrangements, a dedicated life safety bracket for the primary supply with FP600 power cables on a heavy-duty ladder and FP500 control cables on a tray above it; the secondary supply is left for a separate route (BS 8519:2020 7.1). BS 8519 is applied when the sample is chosen and every check on it is a pass. The owner ruled the same day that a fire alarm or emergency lighting run beside a life safety run does not raise the 7.3 line, because those installations are subject to BS 5839-1 and BS 5266-1; the tool already worked that way and did not change. The printed report check now prints the sample as well, and once more with its second run changed to the secondary supply, eight cases in all, and measured the text size, page use, repeated sentences and page numbers with no finding. The automated check of the life safety services, the 2,553-bracket enumeration, the 27 random brackets, the 56 hand-calculated checks and the audit checks were re-run on the final build with no difference; the calculations, the wording shown in the tool, the printed report and the exports are unchanged; verified live after upload. 18/09/2026 (2.8.0): two services were added to the Service list on a run, Life safety, primary supply and Life safety, secondary supply, for the supplies to life safety and fire-fighting equipment; BS 8519:2020 6.1 recommends a secondary power supply independent of the primary supply. BS 8519 is applied to the bracket when either service is chosen and stays applied while a run serves life safety. The tool now states when the primary and the secondary supplies share one bracket (7.1, diverse cable routes), when life safety cables share a bracket with other services (7.3, a dedicated cable support system) and, for Engineering Toolkit users, when a cable on a life safety run is not a fire-resistant type (clause 11); each line reads the same in the tool, in the printed report and in the report data. Every clause quoted was read in BS 8519:2020, and a new automated check reads them again from the standard on every run. The same check built 144 brackets on 2.7.8 and on 2.8.0, for every earlier service with BS 8519 off and at each category, and found no difference in the wording shown in the tool, the drawing, the printed report or the report data; it compared the figures of a life safety run with those of an Other run on 162 brackets and found them equal, and it fails on the 2.7.8 build. The 2,553-bracket enumeration, the 27 random brackets, the 56 hand-calculated checks, the printed report check and the audit checks were re-run on the final build with no difference; no width, load, reaction or rod value changed for any earlier service; verified live after upload. 16/09/2026 (2.7.8): the wording in the tool was revised in plain UK English. The Engineering Toolkit lines and the cable weight tip now say in the tool, and the segregation checks are called checks. The calculations, the printed report and the exports are unchanged. The printed report check on all six presets, the audit checks, the 27 random brackets, the 56 hand-calculated checks and the 2,553-bracket enumeration were re-run on the final build with no difference; verified live after upload. 15/09/2026 (2.7.7): the printed report was rearranged to read as a technical report, after the owner's review of a printed 2.7.6 report. The bracket drawing now prints large enough to read: text on the report is 8 pt or larger and the drawing labels 6.8 pt or larger, where some labels printed at 3 to 4 pt on 2.7.6. One summary line names each run that needs action, the run schedule shows a size check and a load check for every run, the basis of the load check is stated once instead of on every run, and a findings table lists the runs that need action and the fire-resistant cable lines. Columns with no entries are left out, the sections after page 1 no longer start on a new page, and every page shows its page number. A new automated check prints the report to PDF for every preset and measures the text size, nearly empty pages, repeated sentences, empty columns and page numbers; it fails on the 2.7.6 build. A second check compared the wording and the drawing in the tool and the report data on 2.7.6 and 2.7.7 and found no difference; the 2,553-bracket enumeration, the 27 random brackets, the 56 hand-calculated checks and the audit checks were re-run on the final build with no difference; no width, load, reaction or rod value changed; verified live after upload. 15/09/2026 (2.7.6): the notes written inside the program code were rewritten in plain UK English. The calculations, the wording shown in the tool, the printed report and the exports are unchanged. The 2,553-bracket enumeration and the 27 random brackets were re-run on 2.7.6 with no difference; verified live after upload. 15/09/2026 (2.7.5): the emergency lighting lines were read back against BS 5266-1:2025, which replaced the 2016 edition. Its clause 8.2 applies to emergency lighting fed from a central power supply system, so the segregation note and the fire-resistant cable line now say so, the cable line cites 8.2.2, and supply cables to self-contained luminaires are named as needing no fire protection (8.1). The wording in the tool and the notes written inside the program code were revised in plain UK English. A new automated check reads every BS 5266-1 citation and the wording of every program file, in the tool, the printed report and the report data, and fails on the 2.7.4 build; the 2,553-bracket enumeration, the 27 random brackets, the 56 hand-calculated checks and the audit checks were re-run on the final build with no difference; no width, load, reaction or rod value changed; verified live after upload. 15/09/2026 (2.7.4): the wording of one line was revised. On a BS 8519 fire route the fire alarm and emergency lighting cable lines now state that the category applies to the supports only, with the drop rods and the containment load checked for the fire survival time; the previous sentence did not say what was checked. The automated check reads the sentence in the tool, the printed report and the report data and fails on the 2.7.3 build; the 2,553-bracket enumeration and the audit checks were re-run on the final build with no difference; no width, load, reaction or rod value changed; verified live after upload. 15/09/2026 (2.7.3): changes from the owner's review of a printed 2.7.2 report. On a BS 8519 fire route the fire-resistant cables on each run are now checked in the tool, in the printed report and in the exports alike; before, the only note sat in the design notes shown in the tool. The check was corrected to the standards at the same time: BS 8519:2020 clause 1 excludes the installations covered by BS 5839-1 and BS 5266-1, so fire alarm cables are referred to BS 5839-1:2025 clause 25 and emergency lighting cables to BS 5266-1, and FP200 Gold is treated as the Category 1 control cable its datasheet declares. The segregation reference now cites BS 5839-1:2025 35.5 (26.2 was the 2017 numbering). BS 8519 is switched on from the Standards applied strip, the tier spacing is asked only on a bracket with more than one tier, and the report headline is coloured by the worst verdict. A new automated check reads the fire cable lines in the tool, the printed report and the exports against the scope of the standard and fails on the 2.7.2 build; the 2,553-bracket enumeration, the 27 random brackets and the audit checks were re-run on the final build with no difference; no width, load, reaction or rod value changed; verified live after upload. 10/09/2026 (2.7.2): the tier spacing, the BS 8519 fire route and the cable category, read only by a fire-rated or multi-tier bracket, moved into a route options section on the add-a-run panel, closed until opened, so the common run needs only its cables, spare and support spacing. Three wording defects found by the owner on the live tool the same evening are corrected: the over-table bracket advice offered a third rod, which a two-rod trapeze cannot take, and now reads reduce the bracket spacing or split the route; the worst rod tile read over table and now names the rod exceeded; and the indicative load check offered a heavier duty to a heavy-duty tray and a shorter span where no tabulated span was rated for the load, and now names only a remedy that passes on the benchmark rows, or says that none does. An automated check tests all four against the benchmark rows and the live page and fails on the 2.7.1 build; the 2,553-bracket enumeration, the 27 random brackets and the audit checks were re-run on the final build with no difference; no width, load, reaction or rod value changed; verified live after upload. 04/09/2026: a full enumeration drove 2,553 brackets through the live engine, every containment type with single-core, multicore and mixed cable sets in flat and trefoil formation, spaced and touching, with and without spare, at two spans and on unrated and fire-rated routes of every category, plus bunched data runs at four depths, trunking by reference at fifteen sizes, three-run brackets across tier layouts and over-width splits. Every width, load, bracket reaction, rod size and tier count matched an independent calculation. 04/09/2026 (2.7.1): every numeric input was checked for the figure a reader could enter in its place. The custom cable diameter is now defined as the overall diameter over the sheath, not the conductor size; the weight as kilograms per kilometre, with a figure that reads like kilograms per metre flagged in the tool; the support spacing in metres, not the cleat spacing, limited to 12 m and flagged above 6 m; the tier spacing as the clear distance between tray surfaces that Tables 4C4 and 4C5 Note 3 tabulate at 300 mm. No width, load, factor or rod value changed: the 27 random brackets and the audit checks were re-run and all passed, and the script on the live site was confirmed identical to the tested build. 03/09/2026 (2.7.0): a check across every PD tool that every printed value is reached through the input the standard derives it from. The grouping factor was found to count every tier on the bracket as a tray of grouped circuits, including data and AV tiers; it now counts only the tiers with loaded circuits and says which, and asks the vertical spacing between tiers that Tables 4C4 and 4C5 Note 3 assume. The loading verdict now names the duty chosen and is an indicative check against the manufacturer's declared SWL, which can be entered. Single-core cables over 40 mm diameter, which Table D1 does not cover, take the tested cleat centres of the basis product or ask for the manufacturer's figure. A perforated or unperforated tray is now a choice, unperforated reading Reference Method C (Table 4A2). Blank spare, blank custom diameter and blank weight ask for the value. The 56 offline checks and the 27 random brackets were re-run and all passed; verified live after upload. 02/09/2026 (2.6.1): the revised wording was checked in the tool against the house style, using the multi-service bracket preset; no size, load or factor changed. Claims audit 14/07/2026: every cited clause was checked against the standard itself (BS 7671:2018+A4:2026 Tables 4A2, 4C1, 4C4, 4C5; OSG Appendix D Table D1; BS 8519:2020 Annex I and Table E.2; BS EN 61439-1 Table 7; BS EN 61537; BS 5839-1:2017 26.2, now BS 5839-1:2025 35.5; BS 5266-1:2025 8.2.6; BEAMA guide); one grouping factor was corrected to the standard as a result. The 56 hand-calculated checks are re-run offline before every release, and the 03/07/2026 audit figures were reproduced on the live v2 engine on 13/07/2026. Randomised cross-check 15/07/2026: 27 random brackets, including a 12-run, 4-tier case, matched independently calculated results exactly. Grouping Cg reproduces the Trimble ProDesign worked example. Cable data audit 11/08/2026: all 23 BS 5467 SWA rows were checked row by row against the Eland Cables dimensions tables (Issue 04.0913), and the 3-core 25 mm² was cross-checked against the Doncaster and Cleveland datasheets. Three figures were corrected: the 5-core 10 mm² diameter, 22.09 to 22.9 mm, and the 5-core 70 and 95 mm² rows, which had no published source and now follow the Cleveland Cable BS 5467 datasheet, the only one of the five checked that publishes 5-core above 50 mm². The offline checks were re-run on 12/08/2026 after the interface update and all passed.