Overview · Clients · RUBISCO2

RUBISCO2 — Sand Separator

Fresh · 0d

Production design for ~100 hotels. Spark Automations contracts Pablo González for the design work. The June 9 demo is the production unveiling, not a prototype reveal — frame every recommendation as "manufacturable at scale by multiple Mexican fab shops" + "service supply chain at 100 sites."

Demo
2026-06-09
43 days out
Annex B
2026-05-31
34 days out
Sempiternus
2026-04-29
2 days out
CAD-hire trigger
2026-04-28
tomorrow

Esta semana auto · scripts/weekly_client_refresh.py

Bloque regenerado el 2026-06-29 08:00 PdC desde el event log de los últimos 7 días.

2026-06-29 · Sin actividad registrada en bitácora (22 al 28 de junio) No se detectaron eventos ni actualizaciones de ingeniería para el proyecto RUBISCO2 en el sistema durante los últimos 7 días. [Tier 3: Se asume continuidad operativa latente sin novedades reportables en el log principal].
sin cambios
Seguimiento: Cierre de Demo de Producción (Hito: 2026-06-09) Han transcurrido 20 días desde la presentación del diseño de producción para ~100 hoteles. No se han registrado minutas de retroalimentación ni actualizaciones del pipeline de fabricación en GDL. [Tier 3: Se requiere validación directa con Pablo González para formalizar el estatus de aceptación del diseño].
revisar

Iterate-render projects live · phone-first

Fullscreen 3D viewers with click-to-isolate sub-parts + plain-language iterate dock (GPT-5.5 + CadQuery rebuild ~30-60 s). Each project's details.json drives the per-sub-part info pill (material, mfg process, fasteners, tolerances, mass).

Full machine alignment →Current feeder-dev + drum-dev combined viewer. Uses the trompa-to-drum-axis centerline for top-view alignment and shows the trompa entering the drum cone.
V2 drum (full machine) →Bible v3.0 §3-§10 production design. Drum core + frame + drive + rollers + brushes + sand deflector + HMI + guards. Tap any sub-part for material/mfg/fasteners/tolerances.
Plan D feeder →Vibrating bin 1000 × 2500 with integrated long gentle funnel (taper 1000→700 over second half). Springs + cups + L-brackets + jack screws + X-bracing + sand deflector. Q-Drum-17 interface.
Test rig →Calibration apparatus for funnel-angle tuning before committing to the production design. 10 PRT-9xxxx parts: frame + walls + legs + test plate + motor mount + VFD box + sand tray + jack handle.
V1 drum (ref) →Historical V1 (700 OD × 2500). Reference for the iterate-loop demo. Not the production design.

Catalog of all PRT parts (cad-twin pipeline) at cad-viewer. State badges: green = native-verified parametric SW, yellow = CadQuery only, grey = concept mesh.

V2 Specs (from bible v3.0) production

Drum2.50 m × 700 mm nominal OD (pending field verify)
DriveV2 electric gearmotor + VFD + belt + drive roller + external raceway
RPM24 RPM hard limit
Capacity target3,000–6,000 kg/h
Screen panels316L modular, 4 mm round holes baseline (PU/rubber = V3, not V2)
Raceway band (PRT-80001)1045 + hardfacing, 120 mm wide, 720 OD / 700 ID. NOT A36 flat ring.
Lifter wearA36 body + Hardox 450 bolted wear bar on leading edge
Sand chemistryCaribbean = CaCO₃ Mohs 3-4, NOT SiO₂ Mohs 7. Corrosion is #1 threat, not abrasion.

V1 (gasoline + chain) is dead. Anything referencing those is stale.

PRT Numbering parts

70xxxFeeder family
80xxxDrum family
90xxxTest rig family

Notable parts

PRT-80001Raceway band — 1045 + hardfacing, 120 mm × 720 OD / 700 ID
PRT-90001Test plate small — gold-standard reference component (cad-twin gold)

Team humans on the ground

Marcoout of project — moved to the reactor (still inside RUBISCO2, just a different project)
PedroWeld / fab in GDL. Sempiternus rig.
Eduardo MonMotors. OLI quoted USD $4,870/motor.
AndyDesign
Pablo G.Design lead — Spark Automations contracts him for this work

Field Trips PdC

Playa del Carmen field test April 7 — sand chemistry confirmed CaCO₃, not SiO₂. Reframed material strategy from abrasion to corrosion.
done
GDL flight (Pablo G. → Sempiternus) Pablo flies to GDL to supervise feeder rig build at Pedro's shop.
scheduled

Sempiternus Meeting (Apr 26, 15:30) tactical, not strategic

Was tactical, not strategic. Decisions: full-size feeder rig · Pedro builds in GDL · Pablo G. flies to supervise · Wed Apr 29 checkpoint · Tue Apr 28 emergency-CAD-hire trigger.

NOT discussed: Iberostar, royalty fix, post-May-31, Anexo B, $200k pricing. V2 launch plan still has gaps.

Pedro accidentally re-anchored as the fab shop without feeder-vs-drum distinction. Marco functionally out (reactor priority).

Vendor Pulse n8n 3Giq79kdfUODIP5P

Last 7 days outbound Apr 26: 7 sends (Sempiternus cluster). Apr 24: 2. Apr 21: 5 sends + 2 errors. Apr 27: 0 so far.
flag Apr 21
Apr 21 send errors Execs 2618 + 2619 errored at 06:08. Never followed up. Worth a one-line check whether those messages ever landed.
unresolved
Standing rule Vendor / outreach status checks must include n8n executions. Memory + supplier_quotes README lag reality. Disclose: outbound-only, no inbound replies.

Q-Drum Design Space 5 themes · 21 questions

Group ABody
Group BInterior
Group CMotion
Group DEnvironment
Group EFleet-wrapper (was empty → Q17–Q21 promoted)

Map: RUBISCO2/Sand Separator/docs/v2/Q_DRUM_DESIGN_SPACE_MAP.md. Working drafts in docs/v2/qdrum_annex/. Q22–Q25 deferred post-demo.

Open RUBISCO2 Decisions 2

  1. Whether tomorrow's emergency-CAD-hire trigger fires. If feeder build is slipping by Tue, you commit to bringing in external CAD support. Decision posture: have Pedro's status + your GDL flight both confirmed before Tuesday morning. Spark recommendation: pre-empt the decision today by pinging Pedro for status. If green, no trigger. If red or no-response, escalate.
  2. SAT regime for SA ↔ Pablo G. revenue split. Spark researches LISR + RIF/RESICO + IMSS, builds comparison sheet for salary / per-deliverable / royalty shapes. You file via SAT online portal with e.firma. No human accountant. Highest-leverage open item — only blocker on routing future RUBISCO2 income through SA cleanly.