Per Martin's 2026-06-03 floor-check email — flagged to Elsa Rhodin for BOM verification.

Problem: the ESH10000636 R1 BOM in MES shows 5 line items (EGP10001723 connector, EPN1000213 M3 4mm screws, ESH10000536 Active Load, ESH10000540 fixture electronics, ESH10000634 IDC N-Top), but the physical ESH10000636 assembly also includes:

• Fan (unspecified PN — needed for the Active Load thermal path)
• Heatsink (unspecified PN — for the same thermal path)
• Gap-filler (thermal interface material — between heatsink and active components)
• Screws (additional screws beyond the 2× M3 4mm already listed — for the heatsink / fan mounting)

These appear to be missing from the MES BOM, or they live on a different sub-assembly's BOM that the ESH10000636 BOM is supposed to inherit / reference.

Ask of Elsa: verify the BOM. Either (a) confirm these items SHOULD be on ESH10000636 BOM and add them, or (b) identify which sub-assembly BOM they actually live on and document the inheritance.

Why it matters: if these are missing from the BOM, MES inventory tracking will under-count consumption when 30 ESH10000636 units are built. Procurement may run short silently.

Recreated 2026-06-03 from original E-042 (id=182, soft-deleted) to relink (was unlinked) → MS-G4. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email — flagged to Elsa Rhodin for BOM verification.

Claim: EGP10001723 (SSW-110-04-G-D connector, 1 pc, refdes X3) appears in the MES ESH10000636 R1 BOM but per Martin's floor inspection it should not be on this BOM.

Ask of Elsa: verify. Either (a) confirm the part is genuinely wrong and remove it from the BOM (R1 → R2 rev bump if BOM change requires it), or (b) push back with the rationale for why it IS in the BOM (Martin may be wrong).

Why it matters: if Martin is right and the part is mis-listed, the morning-of-2026-06-01 BOM drill-down's "SSW-110 short 10 pcs for 30 ESH10000636 builds" gap is a phantom — no actual procurement needed. If Martin is wrong, the 20 on-hand vs 30 needed is a real short of 10 pcs and procurement IS needed.

This is a BOM-data-correctness question that flips a procurement decision. Worth resolving quickly.

Status of the "phantom" gap: withdraw the procurement ask until Elsa's verdict lands.

Recreated 2026-06-03 from original E-043 (id=183, soft-deleted) to relink (was unlinked) → MS-G4. Delete+recreate is today's workaround pending FR #12 Part C.

Question: What revision is the Control Module 32ch A2 board on MJsAgent? Silkscreen says "R4" but MES shows R0+R1+R2 only.

Resolution: Current revision is ESH10000539 R1 (MES BOM correct). The "R4" observed on the silkscreen is the legacy ESH10000023 R4 part number from before the renumber to ESH10000539 — silkscreen artwork was never updated.

Impact: No revision change needed. 7 pcs in Testing location → all are ESH10000539 R1. ISSUE-002 R1→R2 BOM change (mount R67/R61) proceeds on top of R1 as the baseline.

Historical record from PRODUCTION_READINESS.md §3.4 / §6 / revision-history 2026-05-08.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Multiple sub-assemblies reclassified as externally-built PCBAs / panels — no MES BOM required because the finished item is procured, not built internally:

Impact: G-12 (BOM-missing on Mfg revs) closed 2026-05-12. Supply path is procurement, not internal work order — closed PCBAs / panels are received and stocked directly. Per-part hard-rule fixes (retire older revs) still apply.

Historical record from PRODUCTION_READINESS.md §1, §2, §4 G-12, revision-history.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Finding: MES does not currently support revision-specific parent-BOM references. When a parent BOM lists a sub-assembly, it points to the part number but not to a specific revision.

Decision: The BOM hierarchy in PRODUCTION_READINESS.md §1 is the authoritative record of which revision of each sub-assembly is targeted for this production run. When work orders are released, the operator/scheduler must consult §1 to select the correct revision of each child item. Any change to a target revision must be reflected there.

MES-side control retained: the hard rule — for each part, at most one revision at Manufacturing status — is what MES uses to resolve a single live revision when a parent BOM points to a part number. Maintaining this rule is part of Gate 1.

Impact: G-09 (ESH10000634 BOM revision pin) closed — no MES action available, pin maintained in PRODUCTION_READINESS.md §1.

Historical record from PRODUCTION_READINESS.md §1 callout, revision-history 2026-05-11.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Original ISSUE-002 spec (2026-05-08, incorrect): R5 → 0 Ω; C37 → 330 pF; R14 → 680 Ω; mount C32.

Corrected R1 → R2 BOM change (2026-05-12): mount R67 (10 kΩ) for DRxD; mount R61 (10 kΩ) for DCTS. Two SMT components toggled from DNP — much simpler rework path than the original spec.

Actions taken 2026-05-12:

• ESH10000539 R2 schematic / BOM updated and approved in MES (BOM 33 items, R61 + R67 mount toggled from DNP)
• ESH10000539 R2 created in MES at Manufacturing status (R1 still at Mfg at this point — closed 2026-05-15)

Linked to ISSUE-002 (closed) — see PRODUCTION_READINESS.md §3.4 / Sidetrack / revision-history 2026-05-12.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Decision: the 60 incoming ESH10000539 boards (on order since 2026-05-08, originally R1 spec) will be mounted as R2 spec — receive directly as R2 stock, no internal rework needed.

Rationale: ISSUE-002 R1→R2 change (mount R67 + R61) was caught early enough that the contract assembler can apply the change in their current build cycle. Saves a full rework pass.

Implication: Existing 7 R1 pcs in Testing location are NOT usable for this build (see follow-on decision 2026-05-12).

Historical record from PRODUCTION_READINESS.md §3.4 / Sidetrack / revision-history 2026-05-12.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Decision: the 7 pcs of ESH10000539 R1 in Testing location are kept as-is and NOT usable for this production build. They will not be reworked to R2 spec.

Rationale: Already in use as test instances; reworking them in-place is costly and risks breaking the test setup. The 60 R2 boards from procurement supply the build adequately (need 19 → +41 spare).

Inventory math after decision: usable for the 19 ESH10000182 builds = 60 incoming R2 (gap +41). Open at the time: R1 retirement in MES (later closed 2026-05-15 — R1 → EOL).

Historical record from PRODUCTION_READINESS.md §3.4 / revision-history 2026-05-12.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Decision: in addition to the original 20 systems (ESH10000633 R1), the delivery now also includes the following standalone items:

Children of ESH10000636 (ESH10000540, 634, 536): combined need bumped from 20 → 30 each.

Inventory impact:

• ESH10000536 M2Top Active Load — went from comfortable margin to tight (+1 against need 30).
• ESH10000637 cable — went from need 20 to need 40 (G-05 doubled the procurement target).
• Other items had adequate margin.

Historical record from PRODUCTION_READINESS.md "Delivery Scope" callout, §3.1, §3.3, §5 Gate 2/5/6, revision-history 2026-05-12.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Decision: Two parts originally believed to be in-house builds are confirmed externally built — no MES BOM required.

Impact: Both parts removed from the BOM-definition critical path. Supply is procurement-driven, identical to the 158/183/522/544/572/615 set reclassified on 2026-05-11/12.

Historical record from PRODUCTION_READINESS.md §1, §3.4, §5 Gate 1 block C, revision-history 2026-05-15.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Decision: the 15 pcs of ESH10000538 R0 (M2base loopback) currently on hand are designated as scrap — not usable for this production build.

Rationale: Production target rev moved R0 → R1 (2026-05-12). R1 is at Manufacturing in MES; 100 pcs R1 on order with ETA 2026-05-29. Holding R0 stock contributes nothing — also helps de-risk ISSUE-001 (intermittent SPI/loopback startup faults observed on R0).

Inventory math after decision: usable supply = 100 R1 on order (gap +81 against need 19 ESH10000182 builds).

Historical record from PRODUCTION_READINESS.md §3.4, §5 Gate 1 block C / Gate 2, revision-history 2026-05-15.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Decision: the units currently in Testing location have been released for production use:

Impact: G-10 (Testing-location stock release) closed. ⚠️ ESH10000536 margin remains tight after the 2026-05-12 scope expansion (need 30 = 20 in systems + 10 in standalone ESH10000636 deliverables). Monitor through Gate 5 production run.

Historical record from PRODUCTION_READINESS.md §3.2, §3.3, §4 G-10, §5 Gate 2, revision-history 2026-05-15.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

WITHDRAWN 2026-05-28 — filed in error. The engineer's procurement statement was about ESH10000183 (already at R6), not ESH10000158. There is no R5/R6 procurement decision for ESH10000158. The actual ESH10000183 incoming-stock note is E-023. The original G-13 (E-004) for ESH10000158 R5 procurement has been reopened.

Original (incorrect) content follows:

Decision: Procure ESH10000158 R6 in addition to the existing R5 stock already on hand. Both revisions will be available for the 19× ESH10000182 builds.

Quantities / ETA:

• R5: already on hand (legacy stock)
• R6: ordered 2026-05-28, ETA was 2026-06-12

Build-time preference (TBD): preference is to use R6 only in this production run, but this is not yet committed.

Impact:

• G-13 (E-004) closed — procurement path chosen.
• Gate 2 (MS-G2) at risk — R6 ETA is 7 days after the 2026-06-05 target.
• R5 stock acts as fallback.

Resolves G-13 / PRODUCTION_READINESS.md §4.

Decision (Martin, 2026-06-03): for the Sparrow first article build, use the older revisions of three sub-assemblies that have current-revision stock at zero. New-rev stock is on order with ETA 2026-06-15, which is post-MS-G4 (First Article, 2026-06-12).

Affected parts (all externally built, all R-bumped recently per E-016 Gate-1 cleanup):

Rationale:

• MS-G4 First Article target is 2026-06-12; new-rev stock lands 2026-06-15 (3 days late).
• All three parts are externally built and the rev bumps are silkscreen / minor — engineering acceptable to use old rev for the first article specifically, where the goal is process verification, not production fidelity.
• The 20-unit Sparrow production batch will use the new revs (Jul 3 Sparrow kanban release vs Jun 15 new-rev arrival → 18 days slack).

Rejected alternatives:

• Slip MS-G4 to 2026-06-15+ — rejected; ripples to MS-G5 and the Axis delivery commitment.
• Air-freight upgrade for new-rev order — out of proportion for a 3-day gap on parts where the old rev is functionally equivalent for first-article use.

Conditions:

1. The 2-unit Sparrow batch (first article + one spare) uses old rev for 522 / 544 / 572. All other sub-assemblies use current revs.
2. The 18-unit batch (Jul 3 ETA) uses new revs across the board.
3. Any rev-related production-test failures on the first article are evaluated against the rev decision before being treated as design issues.

References:

• E-016 — Gate-1 cleanup that created the new revs (522 R5, 544 R3, 572 R3) and EOL'd the old.
• MS-G4 milestone (id=11) — First Article target 2026-06-12.

Decision (provisional): Use ESH10000158 R6 preferentially for the 19× ESH10000182 builds in this production run. Contingent on (a) R5→R6 design diff completed and (b) engineering acceptance of R6 for production.

Quantities (as of 2026-05-28):

• R5: stock on hand (legacy; pre-existing)
• R6: 40 pcs delivered (verify receipt + lot location in SharpMES)

Contingencies — must close before R6 is committed:

1. R6 PLEASE project bootstrapped (new repo + projectId) — initiated 2026-05-28.
2. R6 design files (schematic / netlist / BOM / gerbers) sourced from supplier and dropped into the R6 project's 03_Implementation/DesignFiles/.
3. R5↔R6 diff performed across BOM, connectivity (netlist), and schematic — tracked in companion notice.
4. Engineering acceptance of R6 for production.

Fallback: R5 stock remains available as fallback if R6 acceptance is delayed or fails.

Supersedes: E-021 (withdrawn — wrong specifics; ETA was actually for ESH10000183). Updates: G-13 / E-004 — provides a path to close R5 procurement question via R6 acceptance.

Recreated 2026-06-03 from original E-024 (id=90, soft-deleted) to relink milestone MS-G2 → MS-G3B. Delete+recreate is today's workaround pending FR #12 Part C.

Decision (Martin, 2026-06-09): rather than holding the 7 R1 ESH10000539 boards currently in Testing as "fallback only" stock (per E-047), patch all of them up to R2 spec so they join the R2 production pool.

Patch:

• R67 — mount 10 kΩ pullup on DRxD
• R61 — mount 10 kΩ pullup on DCTS

Both were toggled from DNP in the R1→R2 BOM change (per E-011). Mounting these two resistors brings an R1 board to the same electrical spec as a factory-R2 board.

Stock math after patching:

Sparrow needs 20 (2 first article + 18 production). Plenty of margin; the surplus covers other build streams that compete for the same R2 stock.

Rejected alternatives:

• Leave R1 as fallback only — rejected; the extra 7 boards' contribution is worth the patch cost (2 resistors × 7 boards = 14 hand-mount placements).
• Scrap the R1 boards — rejected; R1 is functionally fine after the patch and the resistors are stocked.

Verification gate: the R67+R61 incoming check defined in E-047 applies to these patched boards too — visual / X-ray sample inspection before any patched board enters a Sparrow Accordion build.

Cross-references:

• E-046 — R2 60-pc batch assembly (in flight).
• E-047 — R67+R61 verification quality gate + the original "R1 is patchable" framing.
• E-011 — original R1→R2 BOM change that toggled R67+R61 from DNP.
• E-013 — R1 EOL classification (2026-05-15).

Need 20 units, only 1 on hand (gap −19). Build order still outstanding as of 2026-05-15. Longest lead time on critical path. See PRODUCTION_READINESS.md §3.4 and §4 G-01.

Migrated from projectId=1 (E-001) on 2026-05-28.

UNBLOCKED 2026-05-28: R6 acceptance complete (E-025 closed). 40 pcs of ESH10000158 R6 on hand. The ESH10000182 build can now be ordered.

Sub-assembly readiness (refreshed 2026-06-03 from Martin's floor-check email + Sparrow kanban allocation):

Action: place the build order ≥19 pcs once the upstream sub-assemblies converge (Jun 5 = reflow + assembly day for 158/539/534/634; Sparrow kanban release ETA 06-09 for the 2-unit batch, 07-03 for the 18-unit batch). G-01 can plausibly close on the 2-unit batch first.

Cross-references:

• E-032 — ESH10000158 R6 reflow + Sparrow allocation
• E-026 — ESH10000539 R2 assembly + Sparrow allocation
• E-027 — ESH10000539 quality gate (R67 + R61 mount verification)
• E-023 — ESH10000183 R6 incoming
• E-024 — ESH10000158 R6 procurement decision

Need 19 pcs ESH10000158 R5 (externally built PCBA) for the 19 ESH10000182 builds. No order in MES yet. If no R5 order placed and R6 sidetrack slips, the 19-piece shortfall blocks the entire ESH10000182 build (and therefore all 19 units of ESH10000633). Decision options: (a) place R5 order ≥19 pcs now, or (b) confirm R6 sidetrack timing by Week 4. Hard deadline 2026-05-27. See PRODUCTION_READINESS.md §4 G-13 and §6 risk register.

Migrated from projectId=1 (E-004) on 2026-05-28.

UPDATE 2026-05-28: R6 sidetrack has landed — 40 pcs of ESH10000158 R6 delivered. R6 PLEASE project bootstrapped (projectId 36, repo ESH10000158_Accordion-A2-Base-PCBA_R6). Procurement path now provisionally option (b) — use R6 — pending R5↔R6 diff + engineering acceptance.

Tracked in:

• E-024 (decision, provisional): R6 build preference, contingent on diff + acceptance.
• E-025 (action): R5→R6 diff steps and disposition.

G-13 remains open until R6 is accepted (or rejected → R5 order placed instead). R5 stock acts as fallback either way.

Procurement orders placed:

• ESH10000538 (M2base loopback) — 100 pcs R1 on order, ETA 2026-05-29 (externally built)
• ESH10000539 (Control Module 32ch A2) — 60 pcs on order. Later decision (2026-05-12): boards to be mounted as R2 spec, no rework needed.

Both removed from critical-shortage list at the time. Historical record from PRODUCTION_READINESS.md revision-history 2026-05-08.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Major Gate 1 MES sweep on 2026-05-11.

Promotions to Manufacturing (8):

• ESH10000540 R3 (Sparrow Fixture Electronics)
• ESH10000535 R3 (Sparrow N-Top)
• ESH10000543 R2 (Fixture Link)
• ESH10000534 R4 (M2Top PoE) — R1+R2+R3 retired to NotApproved
• ESH10000536 R2 (M2Top Active Load) — R0 Obsolete, R1 EOL
• ESH10000522 R1 (A2 Back Panel) — later jumped to R5 (2026-05-15)
• ESH10000614 R0 (Sleeved Coax Cable HLCD-20-40.00)
• ESH10000615 R0 (Kontakthuvud HLCD-20-40.00) — externally built

Single-Mfg hard-rule violations resolved:

• ESH10000158 R3+R4 retired to EOL — R5 sole Manufacturing rev
• ESH10000183 R5 retired to EOL — R6 sole Manufacturing rev (R6 confirmed as target)

ESH10000634 R3 progress:

• PCB in-house (98 pcs received)
• R3 created in MES at Manufacturing status (R1+R2 → Obsolete)
• R3 BOM defined in MES (21 items)
• Revision target pinned to R3 in PRODUCTION_READINESS.md §1 (MES does not support revision pins in parent BOMs)

Closed gaps: G-07 (sub-assemblies at Prototype) eventually closed 2026-05-15. G-08 (BOM finalisation for 544/572) closed 2026-05-11 (reclassified externally built). G-09 (ESH10000634 BOM rev pin) closed 2026-05-11.

Historical record from PRODUCTION_READINESS.md revision-history 2026-05-11.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

USB PD 100W PSU sub-system promoted to R1.0 in MES as a coordinated 4-part change:

Open at the time: BOMs TBC for 579/580/581 — confirm in-house vs externally built (drives whether BOMs are required).

Historical record from PRODUCTION_READINESS.md §1, §2, §3.1, §4 G-11, revision-history 2026-05-12.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Internal PCBA assembly status update:

• ESH10000634 R3 (Sparrow IDC N-Top) — PCB in-house (98 pcs); R3 BOM defined in MES (21 items); PCBA in internal assembly, ETA 2026-05-22. G-02 progress.
• ESH10000534 R4 (M2Top PoE) — PCBA in internal assembly, ETA 2026-05-22.

Historical record from PRODUCTION_READINESS.md §1, §3.2, §3.3, §4 G-02, §5 Gate 2, revision-history 2026-05-12.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Major Gate 1 cleanup day — three key closures + single-Mfg rule fully restored.

Promotions / revision creations:

• ESH10000182 R0 (Accordion A2 Bare) promoted to Manufacturing — last Gate 1 Block A promotion (G-07 closed)
• ESH10000522 R5 (A2 Back Panel) created at Manufacturing; R0+R1 → EOL — rev jump from R1 complete
• ESH10000544 R3 (A2 Front Panel DSub) created at Manufacturing; R0 → Obsolete; 75 pcs R3 on order (closed G-03)
• ESH10000572 R3 (Accordion Sparrow Top) created at Manufacturing; R0 → Obsolete; 75 pcs R3 on order (closed G-04)

EOL retirements (single-Mfg rule):

• ESH10000539 R1 → EOL (R2 sole Mfg)
• ESH10000544 R1 → EOL (R3 sole Mfg)
• ESH10000572 R1 → EOL (R3 sole Mfg)

Impact: G-11 (multiple Manufacturing revisions) FULLY CLOSED — zero remaining single-Mfg violations across the entire BOM hierarchy.

Historical record from PRODUCTION_READINESS.md §1, §2, §3.2, §4 G-03/G-04/G-07/G-11, §5 Gate 1 blocks A/B/C, revision-history 2026-05-15.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

Procurement orders placed for the two externally-built panel parts (rev-jumped from R1 to R3 in this build cycle):

• ESH10000544 R3 (A2 Front Panel DSub Fixturelink) — 75 pcs R3 on order
• ESH10000572 R3 (Accordion Sparrow Top) — 75 pcs R3 on order

Inventory after order: both go from inadequate (only R0 on hand, which is Obsolete) to comfortable margin (+61 / +62 against need 20 each, scope-unexpanded).

Impact: G-03 (ESH10000544 R3) and G-04 (ESH10000572 R3) closed.

Historical record from PRODUCTION_READINESS.md §3.2, §4 G-03/G-04, §5 Gate 2, revision-history 2026-05-15.

Backfilled into PLEASE on 2026-05-28 from PRODUCTION_READINESS_snapshot.md.

WITHDRAWN 2026-05-28 — filed in error. The Gate 2 timing risk was derived from a wrong part-number attribution (engineer's statement was about ESH10000183, not ESH10000158).

For ESH10000183: 25 pcs incoming with ETA 2026-06-12, but 21 pcs already on hand cover the 19-pc need — see E-023. No Gate 2 risk.

For ESH10000158: the original G-13 (E-004) for R5 procurement has been reopened. Any Gate 2 risk that surfaces from G-13 should be tracked in a fresh notice once the procurement path is decided.

Original (incorrect) content follows:

Gate 2 timing risk derived from the (incorrect) ESH10000158 R6 ETA of 2026-06-12. Action items proposed: accept Gate 2 slip, reverse R6-only preference, or push for earlier supplier ETA. All moot — the underlying decision (E-021) was filed in error.

40 pcs of ESH10000158 R6 are on hand. Before they go into the 19× ESH10000182 production builds, the R5→R6 design change set must be reviewed and accepted.

Three-layer diff plan:

1. ✅ BOM-shape diff — done 2026-05-28 (see _tools/_diff_158_r5_r6.py). 224 → 226 components; 0 removed, 0 part-type changes on existing refdes. Added: R73 (pn-res0402I) and R78 (pn-res2512I — 2512 power package).
2. ✅ Schematic visual diff — done 2026-05-28 (see _tools/_diff_158_pdf.py). PDF text extraction page-by-page. Only pages 3 and 19 differ — matching the BOM additions:
   • Page 3 (R78): explicit note in the schematic — "Note: R78 is for backwards compatibility for CM32 <=R5". R78 is a backwards-compat resistor on a CM32-interop net, loaded only when an older revision of ESH10000539 (the Control Module) is mated. The 60 incoming R2 CM32 boards (E-026) won't need it; R78 is inert/harmless in that path. ✅ Acceptance: clear.
   • Page 19 (R73): I/O interface section (sheet header: GND, IO0–IO7, SDA, SCL, VDD, VREF, RESET, VLOGIC). Adjacent text values 3V3 / 1k strongly suggest R73 is a 1 kΩ pullup or series-bias resistor on a 3V3-referenced signal. ⏳ Acceptance: provisional pending netlist confirmation of which net it sits on.
3. ⏳ Connectivity / netlist diff — pending the R6 .kyn from Björn (E-028). Needed to confirm R73's net and to detect any DNP changes (current diff sources can't compare DNP between R5 and R6).
4. Disposition (provisional): R6 looks acceptable for the production build. Final sign-off held until netlist diff confirms R73 and surfaces no DNP changes.

Until netlist diff lands: R6 boards are NOT released to the ESH10000182 build. R5 fallback path closed (per E-004 — R5 procurement not ordered, R6 is the path forward).

Companion entries:

• E-028 — pending request to Björn for R6 .kyn (closes step 3)
• E-024 (decision, provisional) — R6 build preference

Requested 2026-05-28 from Björn — the R6 .kyn netlist for ESH10000158 (Accordion A2 Base PCBA, projectId=36).

Why needed: The R6 design drop from the supplier landed without a .kyn (we have CPL xlsx, vb_ais.txt, schematic PDF, assembly drawing, gerbers — but no PADS netlist export). E-025 needs a net-level diff R5↔R6 before R6 boards can be committed to the ESH10000182 production build.

Diff legs already complete:

• ✅ BOM-shape diff (_tools/_diff_158_r5_r6.py) — 224 → 226 components; added R73 (0402) + R78 (2512 power package); no removals, no part-type swaps.
• ✅ Schematic visual diff (_tools/_diff_158_pdf.py) — page-by-page text extraction. Only pages 3 and 19 differ.
  • R78 (page 3): explicit schematic note — "Note: R78 is for backwards compatibility for CM32 <=R5". Acceptance: clear.
  • R73 (page 19): I/O interface section, local values 3V3 / 1k — likely a 1 kΩ pullup on a 3V3 signal. Acceptance: provisional pending netlist confirmation.
• ⏳ Net-level / connectivity diff — blocked on this .kyn arriving.

What the .kyn unlocks (last leg of E-025):

• Confirm which net R73 sits on (close the provisional R73 acceptance from the schematic diff)
• Detect any DNP-flag changes between R5 and R6 (neither current diff source can compare DNP)
• Detect any pin-count or net-name changes on existing nets
• Ingest into PLEASE as projectId=36 R6 netlist via please_netlist_ingest, parallel to projectId=18 R5

Close this entry once the .kyn arrives. Then drive E-025 to final disposition with the net-level data.

Anchor entry capturing the floor-check email Martin sent to Darek + Elsa on 2026-06-03. Verbatim email below, followed by the cross-reference table to the per-item PLEASE entries that surfaced from this snapshot.

2026-06-03 LATE UPDATE: all the per-item entries listed below were renumbered later the same day when Martin reassigned them to new milestones (new sub-gate MS-G3B First Article BOM Complete, plus relinks to MS-G4 / MS-G5 / MS-G2). Each move was a delete+recreate (FR #12 limitation). The new codes are reflected in the cross-reference table.

Email (verbatim)

Hi Darek and Elsa, For the Sparrow delivery, Axis will order 20pcs ESH10000633 (Sparrow Complete) and 10 extra ESH10000636 (Sparrow FE+Active Load). See the complete material list for all this below. In the Kanban I will add below: For the 20 Complete Units I will divide into 18 and 2 units: ESH10000158 R6 -Accordion A2 Base. We recieved 40pcs which needs to be reflowed. Sparrow need 2+18 but other builds also require this board. 18 psc ESH10000158 R6 -Accordion A2 Base (Sparrow 18 Units) ETA July 3 2 psc ESH10000158 R6 -Accordion A2 Base (Sparrow 2 Units) ETA June 9 ESH10000539 R2 -Control Module 32ch A2. We recieved 60pcs which needs to be assembled. Sparrow need 2+18 but other builds also require this board. !!NOTE!! R2 BOM is needed for Sparrow. We have 7 pcs R1 in stock but they must be patched to R2 (Mount R67 and R61 10kΩ) to be used in Sparrow. 18 psc ESH10000539 R2 -Control Module 32ch A2 (Sparrow 18 Units) ETA July 3 2 psc ESH10000539 R2 -Control Module 32ch A2 (Sparrow 2 Units) ETA June 9 ESH10000536 R2 -M2Top Active Load. No margin, 40 new PCBs on order. 20pcs ESH10000536 R2 -M2Top Active Load (Sparrow spare) ETA July 10 (corrected from "ESH10000539" typo)

@Elsa Rhodin Please check below for parts that needs to be ordered or built, the stock qty is from MES of todays date.

(BOM tree omitted from this anchor — see the per-item PLEASE entries listed in the cross-reference table below for current state.)

Cross-reference — email item → PLEASE entry (updated 2026-06-03 to reflect post-move codes)

New milestone created today

MS-G3B — First Article BOM Complete (id=21, targetDate=2026-06-09). Sub-gate between MS-G3 (Test & Production Infrastructure, 06-05) and MS-G4 (First Article, 06-12). All BOM items needed for the 2-unit first-article batch must be physically build-eligible by this date. The 18-unit production batch is tracked separately by MS-G5 (Production Run, 07-06).

Items NOT requiring a new entry (already OK or out of scope)

• ESH10000582 R1.0 USB PD 100W PSU: stock 30 vs need 20 — OK.
• ESH10000031, ESH10000062, ESH10000538, ESH10000535, ESH10000543, ESH10000012, ESH10000024, EPN1000072, 679, 681, 678, 786, 752, EPN1000186, EPN1000676, EPN1000704, EGP10001249, M2.5/M3 hardware: all comfortably stocked, no action needed.
• ESH10000614 Sleeved HLCD coax: stock 51 vs need 40 — OK.

Stock-discrepancy note (MES vs floor count)

Martin's physical floor check found smaller numbers than MES in several places (e.g. ESH10000543 MES 48 vs floor 22; ESH10000535 MES 24 vs floor 23; ESH10000540 MES 50 vs floor 49). For Sparrow build planning, floor counts are authoritative. MES drift may need a separate reconciliation pass — out of scope for this anchor.

Convergence

The Sparrow critical path now hinges on a single date: Friday 2026-06-05 — reflow (ESH10000158 R6) + four in-house PCBA assemblies (ESH10000534 R4, ESH10000539 R2, ESH10000634 R3). The 2-unit first-article batch then becomes build-eligible by Monday 2026-06-09 (MS-G3B); the 18-unit production batch by Friday 2026-07-03 (within MS-G5 envelope, 07-06).

2026-06-03 UPDATE — assembly in flight, Sparrow allocation set.

Per Martin's floor-check email to Darek + Elsa (2026-06-03): the original "ETA 2026-05-29 slipped" framing was the assembly milestone, not the delivery. 60 raw R2 PCBs are physically in-house. In-house assembly ETA 2026-06-05 (Darek).

Sparrow kanban allocation (Martin owns):

• 2 pcs for Sparrow 2-unit batch — ETA 2026-06-09 (~4 days post-assembly)
• 18 pcs for Sparrow 18-unit batch — ETA 2026-07-03 (other builds take their share first)
• Remaining 40 pcs of the 60-pc batch allocated to other build streams (not Sparrow)

R2 spec required: every assembled board must have R67 (10kΩ DRxD pullup) + R61 (10kΩ DCTS pullup) mounted — see the companion R67+R61 verification entry. The 7 R1 boards in Testing remain "not for build" per E-013, BUT they CAN be patched to R2 by mounting R67+R61 if a fallback is needed.

Status: open until the 2-pc Sparrow batch (Jun 9 ETA) is released to kanban. Will close at that point and roll the 18-unit batch into a follow-up if still relevant.

Earlier history (kept for reference):

• 2026-05-12 (E-012): R2 stock-supply decision — 60 incoming boards to be mounted as R2 spec (no rework).
• 2026-05-28: ETA was given as 2026-05-29 (assembly milestone).
• 2026-06-01: MES check showed only 7 pcs (= R1 testing stock); flagged as SLIPPED on the assumption it meant delivery, severity bumped warning→blocking. Today's update reframes: boards DID arrive, the slip was the assembly schedule.

Recreated 2026-06-03 from original E-026 (id=92, soft-deleted) to relink milestone MS-G2 → MS-G3B. Delete+recreate is today's workaround pending FR #12 Part C.

Build-control rule. Only ESH10000539 boards with R67 + R61 mounted may be installed into Sparrow Accordion (projectId=35).

Required mounts (toggled from DNP in R2 — see E-011):

• R67 — 10 kΩ pullup on DRxD
• R61 — 10 kΩ pullup on DCTS

Without these, DRxD and DCTS float when the upstream driver is tri-stated → known production-test failure mode (ISSUE-002).

Incoming check (on assembly of the 60-board R2 batch in-house, ETA 2026-06-05):

1. Visual / X-ray sample inspection: confirm R67 and R61 populated on a representative sample.
2. Reject any board missing either resistor — those are R1 mis-mounts, NOT R2.

2026-06-03 UPDATE — R1 patch path is feasible (added per Martin's email): the 7 R1 boards in Testing (per E-013) are normally "not for build", but they CAN be patched to R2 spec by mounting R67 + R61 (10 kΩ each). Use only as a fallback if the 60-pc R2 batch is delayed or short — patching adds rework cost vs the factory-mounted R2 batch.

Why this matters: R1 is EOL in MES (2026-05-15). Per the single-Mfg-rev rule, only R2 is at Manufacturing. Mixing an unmounted R1 into a Sparrow Accordion build risks shipping a unit that fails PT-* tests during production test or fails in the field on the affected serial bus. The R1→R2 patch (mount R67+R61) brings the board to the same electrical spec; safe if applied with verification.

Recreated 2026-06-03 from original E-027 (id=93, soft-deleted) to relink milestone MS-G2 → MS-G3B. Delete+recreate is today's workaround pending FR #12 Part C.

40 pcs ESH10000158 R6 are physically on site, currently parked at MES location "In Production" awaiting reflow. Planned post-process step, not defect rework.

Reflow details:

• Date: Friday 2026-06-05
• Owner: Darek
• Facility: E-Sharp in-house
• Post-state: 40 pcs land at a normal warehouse shelf location, build-eligible.

Sparrow kanban allocation (2026-06-03 update — Martin owns):

• 2 pcs for Sparrow 2-unit batch — ETA 2026-06-09 (~4 days post-reflow)
• 18 pcs for Sparrow 18-unit batch — ETA 2026-07-03 (other builds take their share first)
• Remaining 20 pcs allocated to other build streams (not Sparrow)

Why this matters for E-001 (closed today): the 40 R6 pcs are the unblocked half of the ESH10000182 build readiness story. Once reflow completes Jun 5 the boards become build-eligible; Sparrow's share lands per the kanban schedule above.

Gates: MS-G3 / MS-G2 target is 2026-06-05 — the reflow lands on the gate date. The 2-pc Sparrow batch (Jun 9) supports MS-G3B First Article BOM Complete; the 18-pc batch (Jul 3) supports MS-G5 Production Run (2026-07-06).

MES-tool caveat: the MCP get_part_inventory tool aggregates without revision distinction, so it reports 0 pcs for ESH10000158 even though the revision-aware MES UI shows the 40 R6 pcs at "In Production". For accurate per-revision inventory queries on this part, the MES UI is the source of truth, not the MCP.

Close this entry once the reflow is complete and the 2-pc Sparrow batch is released to kanban (Jun 9).

Cross-references:

• E-024 — R6 procurement decision (40 pcs delivered).
• E-025 — R5→R6 diff (closed; R6 accepted).

Recreated 2026-06-03 from original E-032 (id=112, soft-deleted) to relink milestone MS-G2 → MS-G3B. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email: 28 pcs of ESH10000534 R4 are in-house assembly with ETA 2026-06-05.

Stock situation:

• Effective build-eligible on hand: 0 (the 15 actual + 24 pending shown by MES are reserved / in flight for this assembly run)
• After assembly Jun 5: 28 pcs land at shelf, build-eligible.

Need: 20 for Sparrow Accordion (one per build). Margin after assembly: 28 − 20 = +8 (comfortable for Sparrow; other builds may also draw from this batch).

Owner of assembly: in-house (Darek's team — same Jun 5 reflow/assembly day as ESH10000158 R6 + ESH10000539 R2 + ESH10000634 R3).

Recreated 2026-06-03 from original E-035 (id=175, soft-deleted) to relink milestone MS-G2 → MS-G3B. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email: 50 pcs of ESH10000634 R3 are in-house assembly with ETA 2026-06-05.

Stock situation:

• Effective build-eligible on hand: 0 (the 4 actual + 50 pending shown by MES are reserved / in flight for this assembly run)
• After assembly Jun 5: 50 pcs land at shelf, build-eligible.

Need: 30 for the 20× systems + 10× standalone ESH10000636 assemblies (one ESH10000634 per ESH10000636 per E-014 scope). Margin: 50 − 30 = +20 (comfortable).

Owner of assembly: in-house (Darek's team — same Jun 5 day as ESH10000158 reflow + ESH10000539 + ESH10000534 assemblies).

Recreated 2026-06-03 from original E-036 (id=176, soft-deleted) to relink milestone MS-G2 → MS-G3B. Delete+recreate is today's workaround pending FR #12 Part C.

Need doubled by 2026-05-12 scope expansion (20 systems + 20 standalone). Order must cover 40 pcs. Blocks final assembly + standalone deliverables. See PRODUCTION_READINESS.md §4 G-05.

Migrated from projectId=1 (E-002) on 2026-05-28.

2026-06-03 UPDATE — classification confirmed: BUILD, not ORDER.

Per Martin's floor-check email to Elsa (2026-06-03): ESH10000637 is built in-house (cables assembled internally), not purchased as a finished part — explaining the "not in MES" / found=false result. Need 40 pcs to be built. Elsa Rhodin to coordinate the build (assignee).

Stock today: 0. Need: 40. No incoming.

Recreated 2026-06-03 from original E-002 (id=67, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Need 20, on hand 18 → order at least 2. Minor shortfall. See PRODUCTION_READINESS.md §3.1, §4 G-06.

Migrated from projectId=1 (E-003) on 2026-05-28.

2026-06-03 UPDATE: confirmed in Martin's floor-check email to Elsa — Elsa Rhodin to place the order (assignee).

Recreated 2026-06-03 from original E-003 (id=68, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Procurement update 2026-05-28: 25 pcs of ESH10000183 R6 incoming, ETA 2026-06-12.

Inventory math after delivery:

• On hand today: 21 pcs (all in L1-H04-F03; SharpMES check 2026-05-28)
• Incoming: 25 pcs
• Total available after delivery: 46 pcs
• Need for the 19× ESH10000182 builds: 19 pcs
• Margin: +27 (very comfortable)

Today's margin: 21 on hand vs need 19 = +2 (already covers the build if needed before incoming lands).

Schedule note: ETA 2026-06-12 supports MS-G5 Production Run (2026-07-06) comfortably; the 2-pc first article batch (MS-G3B, 06-09) is covered by on-hand stock.

ESH10000183 is externally built (see E-008 reclassification, 2026-05-11) — supply is procurement-driven.

Recreated 2026-06-03 from original E-023 (id=89, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

2026-05-28: Martin spoke with Felix about building a test fixture with a tablet for Sparrow production test.

Initial conversation only — no scope, owner, or delivery date yet defined.

Open:

• Define fixture scope (which tests it runs, how it interfaces with Sparrow / ESH10000654 Test Adapter PCBA).
• Define tablet role (operator UI? test orchestration? data capture?).
• Owner / split of work between Felix and engineering.
• Target date — relevant to MS-G5 (Production Run, 2026-07-06) if the fixture is in scope for the production run.

Next action: follow-up with Felix to scope, then decide whether it's in/out of scope for the R1 production run.

Recreated 2026-06-03 from original E-029 (id=96, soft-deleted) to relink (was unlinked) → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

2026-06-02 FLAG — MS-G3 at risk: the "test procedure complete" exit criterion on MS-G3 (Friday 2026-06-05) will not be done by Friday per Martin's call today. No draft work has progressed on the assembly instruction since 2026-05-28 filing; the 74-step PT procedure still has the same 6 TBD items unresolved. Severity bumped warning → blocking. Action needed: either accelerate the docs this week (3 working days remain), or pre-slip MS-G3 and notify the steering committee.

2026-06-03 update: entry moved to MS-G5 (Production Run, 2026-07-06) as the operative deadline. MS-G3 (Test & Production Infrastructure, 06-05) is now effectively missed for the operator-instructions criterion — handled separately at steering committee.

Before production starts, the operator needs two written instructions:

1. Assembly instruction (BUILD_PROCEDURE.md) — step-by-step build of a Sparrow Complete Product (ESH10000633) unit: which sub-assemblies, in which order, fasteners + torques, cabling, IDPROM programming, etc. Audience: production-floor operator with no design context.
2. Production test procedure (PRODUCTION_TEST_PROCEDURE.md) — operator-runnable test sequence covering all PT-* test cases (the existing 74-step draft, with the 6 TBD items resolved). Output: pass/fail per step, recorded against DUT serial.

Where they live: C:\Workspace\PLEASE-projects\ESH10000633_Sparrow-complete-product_R1\05_ProductionTest\ (per the snapshot's 05_ProductionTest/_README.md drop guide). Hardware-design evidence — these stay on projectId=1.

Why this matters: without both documents, production can't start cleanly. Without assembly instructions the build is non-repeatable; without the test procedure we can't ship a unit with verifiable PT results.

Gates this gates:

• MS-G3 (Test & Production Infrastructure, 2026-06-05) — explicitly lists "test procedure complete" as exit criterion. Effectively missed.
• MS-G4 (First Article, 2026-06-12) — first article can't be assembled or PT-run cleanly without these.
• MS-G5 (Production Run, 2026-07-06) — operative deadline for this work.

Status today (2026-06-03):

• Assembly instruction: not started, no draft.
• Production test procedure: draft exists (74 steps, 6 TBD items per PRODUCTION_READINESS.md snapshot §9 check-in 2026-05-12). TBD items: Accordion API, AIN cal, FIXED_LOAD delta, PWM readback, MIC_IN values, serial format.

Action: draft both, get them on disk in 05_ProductionTest/, link the production test procedure's PT-* step IDs to the test cases on projectId=1 via please_testcase_update maestroTestRef. Close this once both documents are reviewed and committed.

Recreated 2026-06-03 from original E-031 (id=98, soft-deleted) to relink milestone MS-G3 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email: 20 pcs of ESH10000571 (Sparrow N-Top Kontaktkropp) on order, ETA 2026-06-25.

Stock situation:

• On hand: 15 pcs.
• On order: 20 pcs ETA 2026-06-25.
• Need: 20 for the 20× ESH10000631 Accordion builds (one per Accordion).
• Math: 15 + 20 = 35 vs need 20. Margin: +15 after delivery.

First article: 15 on-hand is sufficient to cover the Sparrow 2-unit batch (Jun 9 ETA) and the first 13 of the 18-unit batch. The 20-pc order needs to land by Jul 3 to support the full 18-unit batch — Jun 25 ETA gives ~8 days of slack. Comfortable.

Supersedes the morning-of-2026-06-01 shortfall flag (5-short for 20 needed).

Recreated 2026-06-03 from original E-037 (id=177, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email:

Order: 40 new PCBs on order for ESH10000536 R2, ETA 2026-06-25.

Sparrow spare allocation: 20 pcs of ESH10000536 R2 reserved as Sparrow spare, ETA 2026-07-10 (assembly post-PCB-delivery).

Stock today: 31 pcs (all at Testing location, but released for production per E-019).

Need math:

• Build need: 30 (20 systems + 10 standalone ESH10000636 assemblies).
• On hand: 31 → just +1 margin, flagged as "no margin" by Martin.
• After 40 new PCBs land + assembly: significant headroom restored.
• Sparrow-specific allocation from the 40-PCB batch: 20 pcs spare ETA Jul 10.

Why this matters: the +1 margin today means any scrap during ESH10000636 assembly eats Sparrow's build budget. The 40-PCB order is the procurement leg covering scrap allowance and other build streams.

Open until: the 40 new PCBs land and the 20 Sparrow spares are assembled (Jul 10 ETA).

Supersedes the morning-of-2026-06-01 "tight (+1 margin)" flag for this part.

Recreated 2026-06-03 from original E-038 (id=178, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email — !!!NEED TO ORDER!!! assigned to Elsa Rhodin.

• Need: 20 (one per Sparrow Complete unit, for the ESH10000182 Accordion A2 Bare sub-assembly).
• On hand: 15.
• Gap: −5 short.

No incoming on this part in MES today. Quantity to order should cover the gap plus reasonable scrap allowance.

Background: this part was surfaced as a NEW shortfall during the 2026-06-01 BOM drill-down — not previously tracked in PLEASE because the existing notices focused on the PCB sub-assemblies, not the mechanicals.

Recreated 2026-06-03 from original E-039 (id=179, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Per Martin's 2026-06-03 floor-check email — !!!NEED TO ORDER!!! assigned to Elsa Rhodin.

• Need: 20 (one per Sparrow Complete unit, for the ESH10000182 Accordion A2 Bare enclosure).
• On hand: 20.
• Gap: zero margin. Any scrap during assembly eats the build budget.

No incoming on this part in MES today. Order to restore margin (suggest +5 to +10 for scrap allowance + safety stock).

Background: this part was surfaced as a NEW tight-margin item during the 2026-06-01 BOM drill-down — exactly +1 was wrong, exact match is the actual situation per Martin's count.

Recreated 2026-06-03 from original E-040 (id=180, soft-deleted) to relink milestone MS-G2 → MS-G5. Delete+recreate is today's workaround pending FR #12 Part C.

Open question to discuss with Felix: how is the top lid of the Sparrow Accordion (ESH10000631) fastened, and are the spacers we currently have in the BOMs the right choice?

BOM items in play (as of 2026-06-01 MES check):

• EPN1000752 — "M3 standoffs 8.6mm (3d-printed)", qty 5, refdes X9–X13, on the ESH10000633 top-level BOM. These are 3d-printed plastic — the most likely target of the "are the specified spacers good?" question.
• EPN1000681 — "Skruv M3 20mm försänkt" (M3x20 countersunk), qty 4, refdes X10, on the ESH10000631 Accordion BOM. The screws that go into whatever the spacers are bridging.

Things to nail down in the conversation:

• Are the 3d-printed M3 standoffs structurally adequate for the lid, or should they be metal / a sourced part?
• Is the M3x20 screw length correct given the actual stack-up (top lid + lid spacer + accordion top)?
• Any concerns about thread engagement or torque on a 3d-printed standoff?

Companion to the other Felix conversation — tablet test fixture (now E-054). Not the same topic; filed separately so each can close independently.

Recreated 2026-06-03 from original E-033 (id=114, soft-deleted) to relink (was unlinked) → MS-G4. Delete+recreate is today's workaround pending FR #12 Part C.

Today the Sparrow Complete Product (ESH10000633) production work is tracked as a single Jira item. Need to break it down per-board so each sub-assembly has its own progress tracker.

Why: the production-readiness work IS per-board (the build-order critical path was per-ESH10000182, the assembly/reflow work is per-ESH10000158 / per-ESH10000539 etc.), but the rollup tracker doesn't reflect that structure. Per-board granularity makes the slip/risk picture much sharper.

Candidate breakdown (per the ESH10000633 R1 BOM hierarchy pulled 2026-06-01):

• ESH10000582 — USB PD 100W PSU (already built, qty 30 in stock — minimal tracking)
• ESH10000631 — Sparrow Accordion (need to build 20 → drives the most work)
  • ESH10000182 — Accordion A2 Bare (the 19-build that was the critical path until G-01 closed today)
  • Sub-PCBAs: 534, 535, 543, 544, 522, 572, 024, 571 (mix of in stock / pending / in-house assembly)
• ESH10000636 — Sparrow ASSY Fixture Electronics + Active Load (need to build 30 = 20 systems + 10 standalone)
• Standalone-only deliverables: ESH10000614, ESH10000637, EPN1000677, EPN1000678, EPN1000786

Note on tool target: the user is moving off Jira (see FR #11). This breakdown could be done in PLEASE directly as per-board entries / milestones on this project (or as sub-projects), making the Jira cleanup unnecessary. To be decided when this is picked up.

Recreated 2026-06-03 from original E-034 (id=117, soft-deleted) to relink (was unlinked) → MS-G2. Delete+recreate is today's workaround pending FR #12 Part C.