devices(ubiquiti): three 150 W PoE rows sourced, and US-16-150W was 28 W over budget #2265

Merged
jaylouisw merged 1 commit from catalogue/1120-0522 into main 2026-11-20 05:43:23 +00:00
Owner

Refs #130 only.

That request is still blocked and this branch does not answer it: udw-pro is not a Ubiquiti product page (307 to /), not a card in the eight-grid census (408 cards / 398 SKUs re-counted today, one Dream Wall and no Pro tier beside the 73 existing -Pro SKUs), and not in the archived storefront. No UDW-PRO field is modelled here, and no family-sibling number is reused for it.

What this branch does is the catalogue underneath the queue: three Ubiquiti 150 W PoE rows that served a power-supply rating as a PSE allocation, which is the defect show power inline reports as free watts.

row was now vendor statement
us-16-150w 150 W (name) 122 W storefront row Total available PoE 122W; datasheet p.7 Including PoE Output 150W / Excluding PoE Output 28W
es-16-150w 150 W (name) 150 W, traced storefront key feature 150W total PoE availability
es-8-150w 150 W (name) 150 W, traced storefront key feature 150W total PoE availability

150 - 28 = 122, and Ubiquiti prints 122 itself, so nothing here is arithmetic of ours. This is the third instance of the class the census file already names (US-8-60W 60 -> 48 W, US-8-150W 150 -> 130 W).

Two census corrections the same tables forced. ES-8-150W and ES-16-150W both print an out-of-band UART console that neither spec declared - added (11 and 19 jacks). US-16-150W keeps its console: its datasheet model bullets list (1) Serial Console Port while its own spec table's Management row reads Ethernet In-Band alone, and that conflict is recorded rather than resolved in either direction.

Sources (all Ubiquiti-hosted, measured 2026-11-20):

  • https://dl.ui.com/datasheets/unifi/UniFi_PoE_Switch.pdf - 200, 3,962,155 B, 12 pp, text layer, one table per model. This is the link the product's own page serves (www.ui.com/downloads/datasheets/... 301s here).
  • https://web.archive.org/web/20231113045611id_/https://store.ui.com/us/en/products/us-16-150w - 200, 21,753 B gzip / 131,196 B decoded; Total available PoE 122W, 122W total PoE availability, tag poe:122w.
  • https://web.archive.org/web/20240409153711id_/https://store.ui.com/us/en/products/es-16-150w - 200, 95,590 B decoded; 150W total PoE availability.
  • https://web.archive.org/web/20240906195421id_/https://store.ui.com/us/en/products/es-8-150w - 200, 20,742 B gzip / 124,166 B decoded; 150W total PoE availability, Management interface: Ethernet In-Band, UART console Out-of-Band.
  • live controls: techspecs.ui.com/uisp/wired/es-16-150w 200 / 574,980 B, .../es-8-150w 200 / 583,247 B (prints no Management row), techspecs.ui.com/unifi/switching/us-16-150w 307 to /, dl.ui.com/ds/es-16-150w_ds.pdf and us-16-150w_ds.pdf 302 to the techspecs root.

Research basis: research/ubiquiti-unifi-gen1-poe-census.md said this family has no datasheet text layer. That was true only of the dl.ui.com/ds/ path; the datasheet lives at dl.ui.com/datasheets/unifi/, and the archived-storefront route for retired SKUs is now recorded with it.

Gates: node scripts/syntax-check.js all OK; node scripts/audit-port-types.js no port contradicts its type; node test/hardware-invariants.test.js 26/26; npm test 401 suites / 13,351 assertions, 0 failed; audit-poe-budget "inferred from name" 11 -> 8. docs/CONNECTORS.md regenerated (29,731 -> 29,733 sockets).

Refs #130 only. That request is still blocked and this branch does not answer it: `udw-pro` is not a Ubiquiti product page (307 to `/`), not a card in the eight-grid census (408 cards / 398 SKUs re-counted today, one Dream Wall and no Pro tier beside the 73 existing `-Pro` SKUs), and not in the archived storefront. No UDW-PRO field is modelled here, and no family-sibling number is reused for it. What this branch does is the catalogue underneath the queue: three Ubiquiti 150 W PoE rows that served a **power-supply rating** as a PSE allocation, which is the defect `show power inline` reports as free watts. | row | was | now | vendor statement | |---|---|---|---| | `us-16-150w` | 150 W (name) | **122 W** | storefront row `Total available PoE 122W`; datasheet p.7 `Including PoE Output 150W / Excluding PoE Output 28W` | | `es-16-150w` | 150 W (name) | 150 W, **traced** | storefront key feature `150W total PoE availability` | | `es-8-150w` | 150 W (name) | 150 W, **traced** | storefront key feature `150W total PoE availability` | 150 - 28 = 122, and Ubiquiti prints 122 itself, so nothing here is arithmetic of ours. This is the third instance of the class the census file already names (US-8-60W 60 -> 48 W, US-8-150W 150 -> 130 W). **Two census corrections the same tables forced.** `ES-8-150W` and `ES-16-150W` both print an out-of-band UART console that neither spec declared - added (11 and 19 jacks). `US-16-150W` keeps its console: its datasheet model bullets list `(1) Serial Console Port` while its own spec table's Management row reads `Ethernet In-Band` alone, and that conflict is recorded rather than resolved in either direction. **Sources** (all Ubiquiti-hosted, measured 2026-11-20): - `https://dl.ui.com/datasheets/unifi/UniFi_PoE_Switch.pdf` - 200, 3,962,155 B, 12 pp, text layer, one table per model. This is the link the product's own page serves (`www.ui.com/downloads/datasheets/...` 301s here). - `https://web.archive.org/web/20231113045611id_/https://store.ui.com/us/en/products/us-16-150w` - 200, 21,753 B gzip / 131,196 B decoded; `Total available PoE 122W`, `122W total PoE availability`, tag `poe:122w`. - `https://web.archive.org/web/20240409153711id_/https://store.ui.com/us/en/products/es-16-150w` - 200, 95,590 B decoded; `150W total PoE availability`. - `https://web.archive.org/web/20240906195421id_/https://store.ui.com/us/en/products/es-8-150w` - 200, 20,742 B gzip / 124,166 B decoded; `150W total PoE availability`, `Management interface: Ethernet In-Band, UART console Out-of-Band`. - live controls: `techspecs.ui.com/uisp/wired/es-16-150w` 200 / 574,980 B, `.../es-8-150w` 200 / 583,247 B (prints no Management row), `techspecs.ui.com/unifi/switching/us-16-150w` 307 to `/`, `dl.ui.com/ds/es-16-150w_ds.pdf` and `us-16-150w_ds.pdf` 302 to the techspecs root. **Research basis:** `research/ubiquiti-unifi-gen1-poe-census.md` said this family has no datasheet text layer. That was true only of the `dl.ui.com/ds/` path; the datasheet lives at `dl.ui.com/datasheets/unifi/`, and the archived-storefront route for retired SKUs is now recorded with it. **Gates:** `node scripts/syntax-check.js` all OK; `node scripts/audit-port-types.js` no port contradicts its type; `node test/hardware-invariants.test.js` 26/26; `npm test` 401 suites / 13,351 assertions, 0 failed; `audit-poe-budget` "inferred from name" 11 -> 8. `docs/CONNECTORS.md` regenerated (29,731 -> 29,733 sockets).
devices(ubiquiti): three 150 W rows sourced, and US-16-150W was 28 W over its budget
All checks were successful
CI / Classify the change (pull_request) Successful in 23s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 12s
CI / Test (Node 18) (pull_request) Successful in 1m19s
CI / Test (Node 20) (pull_request) Successful in 1m5s
CI / Test (Node 22) (pull_request) Successful in 1m3s
CI / Server smoke and Docker build (pull_request) Successful in 20s
CI / Arm auto-merge (pull_request) Successful in 4s
ed01318291
The 150 W in the SKU string is a power-supply rating in this generation, not
what a powered device may draw. Three rows served it as a budget:

- us-16-150w   150 W -> 122 W. Ubiquiti prints "Total available PoE 122W" on
  its own storefront row and "Including PoE Output 150W / Excluding PoE Output
  28W" on datasheet p.7; 150 - 28 = 122. The old number allocated 28 W the
  supply cannot give, and show power inline handed it out.
- es-16-150w   150 W, now traced to "150W total PoE availability" (archived
  storefront) instead of to the model name.
- es-8-150w    150 W, same trace.

Two census corrections the vendor tables forced: ES-8-150W and ES-16-150W both
print an out-of-band UART console that neither spec declared (11 and 19 jacks
now). US-16-150W keeps its console: the datasheet's model bullets list
"(1) Serial Console Port" while its own spec table omits the row, and that
conflict is recorded rather than resolved.

Research basis: the family datasheet is at dl.ui.com/datasheets/unifi/, not
under the dl.ui.com/ds/ convention the census file had ruled out, so
research/ubiquiti-unifi-gen1-poe-census.md is corrected; the archived
storefront route (web.archive.org/web/<ts>id_/store.ui.com/...), which still
carries the retired SKUs' tables, is recorded with it.

audit-poe-budget: "inferred from name" 11 -> 8. No closing keyword in this
message or the PR body; the device request that prompted the sweep is untouched.
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-20 05:41:07 +00:00
jaylouisw deleted branch catalogue/1120-0522 2026-11-20 05:43:26 +00:00
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
jaylouisw/packetforge!2265
No description provided.