Megaplan

Powered CEB Plant
+ All-Earth Dome

Salida, CO & Mars College, Salton Sea — ~$700 machine, ~100% on-site materials, zero added water

Buy: splitter→ Bolt on mold · weld box→ Press ~2,100 bricks→ Dig pit (spoil = bricks)→ Stack dome ~3 days→ Berm it — done

Phase 0 — The Buy Run (all links verified live 11/15 — click any row)

ItemWhere$
Log splitter = THE PRESSFt Garland350
Gauge tee + minor adapters (stock plumbing already closed; $40 Florence bag = optional)local10–20
Mold plate 3/8–1/2″ → Pueblo steel supplier cut drops (buy new) — ribs not needed at 25TPueblo60–110
In-line gauge 0–5,000 psibuy-new20
Gr-8 eject pin + boltsbuy-new15–25
Hydraulic + engine oil + fuelbuy-new50–80
Total~460–635
  • Route: Salida → Florence → (CO Springs steel) → home; Ft Garland splitter separate
  • Force: stock 25T cyl — products sized to it: 8×10 dome = 625 psi ✓ · 6×12 wall = 694 ✓ · 10×10 paver = 500 ✓ — no cylinder swap, no beam mod
  • Message first: does splitter start? cyl bore/rod/closed-length
  • Eyes-on checks: beam straight (= the press frame), rod unscored, no pump whine
  • Fallbacks: free splitter Littleton · $160 CO Springs · $750 Mesa (all live)
  • Rule: scrounge everything EXCEPT seals, pins, fluid, gauge — pressure-critical = new

The Machine — The Splitter IS the Press

Splitter as press
  • The insight: a log splitter is already a horizontal hydraulic press — I-beam + cylinder + engine-driven power pack in one unit
  • The build: mold bolts horizontally on the beam end; cylinder pushes the platen through the bore. No press frame purchased, none fabricated
  • Force: system ~3,000 psi × stock ~5″ bore = ~25T → 8×10=625 psi, 6×12=694, 10×10 paver=500 — all in band
  • Cost: splitter $350 + mold plate ~$110 + parts → ~$460–635 — one machine, one purchase
  • Bonus: fully reversible — splitter still splits logs
  • Tradeoff: brick sized to stock force (8×10 = 625 psi ✓) — no swap, no beam mod; ~18 lb units handle easy

■ $350 log splitter: engine, 2-stage pump, valve, tank, beam (the frame), cyl · ■ stock cyl: ~5″ bore = ~25T = the force · ■ Fabricated ~$150: mold box, platen, eject door, shoe, liners

The Fabricated Weldment — Exploded

Mold exploded

All fabricated (~$150 steel): platen (groove underneath) → mold box + stiffeners → brick on shoe → door dropped flat.
Stock 25T cyl = production force — products sized to it (8×10=625 · 6×12=694 · 10×10=500 psi).

Parts — Everything We Weld

Mold box · 14.1×12.1″ master bore · 3/8″ plate + angle
Platen · 1″ plate + stub + dead-stop collar (+ optional key insert)
The product · 8×10×3.5″ flat-face, ~18 lb — stacks like real tholos (keyed insert ALL courses)
Liner set → 6×12 (also 8×10 dome, 10×10 paver)
Eject door · bottom-hinged → drops = take-off shelf · ¾″ gr-8 pin
Bottom shoe · slides out for drop-eject (+ optional keyway bar)

Build — Week 2 Checklist

  • 1. Splitter service: fluids, tire, cold-start + cycle test — cylinder stays stock
  • 2. Cylinder stays stock 25T — verify bore/stroke + pin-mount condition; bench-test at relief 5 min, zero weep
  • 3. Mold bracket bolted to beam end — beam rated for exactly stock force
  • 4. Weld mold box + stiffeners; shoe + hinged fill gate
  • 5. Platen: plate, stub, collar; door: hinge barrels, pin ears — drops flat = shelf
  • 6. Plumb: splitter valve → gauge → stock cyl; relief ≤3,000 psi at deadhead
  • 7. Commission: dry cycles, wood-block test, first soil pucks, check beam for flex, re-torque
  • CAD: cad/params.scad — FIELD-measure beam before cutting

Hydraulics — One Line

Kohler engine~6–8 hp 2-stage pumpfast / strong splitter valverelief ≤3,000 psi gauge0–5,000 psi stock cyl~5″ = 25T tank / filter Force = system psi × bore area → stock ~5″ cyl = ~25T → products sized to it

Soil Gates — Test Before You Commit

  • 1. Test hole ~3–4 ft, watch 24–48h — seep = shallow sink or pad (SLV water table)
  • 2. Jar test — 15–30% clay target; SLV sandy → +5–10% native soil only — crush gate decides
  • 3. Soak/efflorescence test — salt-weak soil fails the 24-hr soak directly (Salton)
  • 4. Selenium — Salton ag-drain history → cheap lab screen
  • 5. Native moisture 4–6% — weigh→dry→reweigh; dig-fresh spoil carries it
  • 6. Bearing walk — tamp, walk it; pumping/prints = rethink siting
  • 7. Pressed validation — waist-high drop survives; 24h soak; SLV freeze-thaw cycle
  • Bone-dry <3% → mist liters/ton only (still ~zero-water)

Production Line — Press at the Pit Edge

Dig pit→Sieve→Tarp-mix (±native soil only — crush gate decides→ Dose scoop→Press + dwell→Door drops→ Slide to stack→Carry to wall
  • ~30–60 s/cycle → 1–2 bricks/min
  • ~2,100 units/dome ≈ 10–21 press-hours
  • Excavation = soil source — spoil never leaves the site
  • Fines → mix · coarse sand → drainage · rest → berm
  • ~18 lb units — one-person stacking, ground-row curing at wall height
  • QC loop: gauge every cycle · thickness/flatness every 20th · drop test every 50th · key-fit every batch
  • The press IS the crush rig — gauge at failure = MPa

The Brick Cycle — Soil In, Brick Out

measured dose → open bore
1 DOSE — soil into the bore
600 psi → dead-stop collar
2 PRESS — platen to dead-stop
brick formed in bore
3 RETRACT — platen up
shoe slides out → brick held by bore walls
4 SHOE OUT — bottom opens
brick through to bed → out over door
5 EJECT — platen pushes through
pull brick → stack
6 TAKE-OFF — shelf → stack

Cycle ~20–40 s/brick powered · bottom shoe IS the mold floor — slide it out, platen pushes the brick through to the bed; front door drops open = take-off shelf at bed level, pull brick over it → stack

Automation — $300 Spec, ~$60 Scrounged

ItemNewScrounged
Valve actuator (junkyard linear actuator on stock lever)90–1405–35
Dose-gate + shoe actuators700–30
Limit switches ×4 (microwave door switches)200–5
Pressure transducer 0–5000 psi4015–25
Arduino Nano + SD + RTC + relays558–15
E-stop + buttons + box + wire6015–25
Total~300–380~40–90
  • ONE donor does ~80%: actuators + relay board + pendant + wire → FREE power recliner (live) · $50 reclining bed backup · limit switches: FREE microwave (live)
  • Controller written: automation/ceb_controller/ceb_controller.ino — DOSE→PRESS→RETRACT→SHOE→EJECT→LOG, timeout-guarded, fail-safe center
  • Auto-QC: every brick logs peak psi + dwell to SD — the certification file writes itself (the thing commercial vendors can't give you)
  • Manual override: stock valve stays plumbed — unplug one connector → 100% manual again
  • Safety: E-stop hardwired (never software) · mechanical relief ≤3,000 psi · both-hands-clear start

The Dome — Exterior + Stomion Entry

Dome exterior — view-dormer window through the mound
Mound + view-dormer: framed 48×30″ reveal through ~14″ of flank soil — true outside view · oculus cap ring at apex
Stomion entry
Rimmed open run + sump + threshold sill → covered vault → doorway never sky-exposed

~22 TRUE-helix spiral courses of 8×10 flat-face · uniform brick-cap apex · mound burial · 12″ apex cap (pit-spoil self-funding) · covered tholos-stomion entry

The Dome — Section

Dome section

Pit floor 36″ below grade → springs from raised collar → 12″ wall + 24″ base band (below grade) → corbel to brick cap → oculus + two 48″ view-dormers (~2 ft shafts — deep reveals, true view) → tholos stomion: stairs under a corbelled vault

The Venue Module — Unburied, No Mound

Interactive — drag to orbit, wheel to zoom, buttons for Exterior/Section/Wireframe/Top. THE GATEHOUSE — earth-sheltered TWO-CELL vaulted gatehouse centered on the square — 14×8-ft hall + 14×5-ft livable back cell behind a shared party wall (door through it; rear vault leans FORWARD onto it — opposing thrust cancels, the multi-bay trick the Dunes basilica repeats; thin liner rear wall — the bank closes the arc): Nubian leaning vault (no formwork) under a soil cap + spring-height berms (27°, placed BEFORE the vault — they carry the thrust) + plinth keys + one continuous counterforts + haunch banks → ~5 g composite-kern bound (cad/venue_battle.py — computed 10–12 g, contingent on compacted lifts + keys) + fully open 14-ft arched mouth under a stepped arch-ring face — the facade's silhouette follows the vault's own arc (retains the cap, no square wall, no vertical soil face) — the echea bass vents moved INTO the rib walls (~56 Hz, woofer discs flanking the entrance = the walls ARE the PA), centered on the square — dancers on bare earth, no pad. ~3,050 units + ~55 yd3 scrape-earth ≈ ~9–11 d all-in, NO machine (counterforts replace the deep bank; jumbo bricks on the same press) w/ pre-pressed bricks + excavator. Dome deferred to Trip 2 (2.67 g dry / 2.18 g wet when built)

The Gatehouse — What's Inside

  • Nubian leaning vault 14×8 ft — vertical front arch + 11 courses at ~15°, each leaning on the last → full 14-ft span, zero formwork; ~200 sq ft covered floor, two cells (hall + livable back cell — the party wall's door links them)
  • Ribs + party wall + rear cell — deep-battered wythe+soil-core walls; the party wall's gable is the shared abutment BOTH vaults lean into (rear cell's courses lean FORWARD — opposing thrust cancels); rear gable wall + vent stack = a livable second room
  • Earth-shelter — load-bearing — berms (~27°) cresting over springing + ~20″ crown cap + mouth wings (the open face gets a kern) + plinth keys below grade (kills sliding → kern rocking); berms go in BEFORE the vault — they carry the thrust
  • Fully open mouth + arched front ring — the 14-ft arch free-spans rib-to-rib, nothing in the doorway; the spandrel steps along the vault's own arc (one arched face, no square wall) and still retains the cap
  • Echea bass vents ×2 — in the RIBS — ~44 L Helmholtz cavities embedded in each rib's soil-core zone, necks + dark woofer discs on the front faces ≈ 56 Hz → the walls ARE the subwoofers (passive ports)
  • Dust-storm package — mouth threshold (saltation lip) + masonry vent stack through the cap + paver armor on the crest + echea shutters: seal the hall, live in the back cell
  • No pad — gatehouse centered on the square; dancers on bare site earth; vent shutters = the acoustic "mute"
  • Honest rating: ~5 g composite-kern bound — battle-tested (venue_battle.py: computed 10–12 g, contingent on compacted lifts + plinth keys + wings; incoherent-soil floor ~1.6 g; Salton liquefaction flagged) / unburied ribs FAIL static thrust — shelter is structural; dome = Trip 2 at 2.67 g dry / 2.18 g wet; 32-ft dance dome = crew build

Build Sequence — ~5 Days On Site

  • Day 1: test hole ✓ → excavate ~36″ pit · swale + levee same day
  • Day 2–3: press ~2,100 keyed units at pit edge
  • Day 3–5: compass arm → spiral courses · clay-dust bedding · place-don't-drop · level each course
  • Day 5–6: brick-cap apex + view-dormer shafts · doorway returns · stair treads + stomion vault
  • Day 6: paver floor · apron skirt · mound berm — spoil self-funds it · drain daylighted
  • TRUE helix spiral (igloo): each brick ~H/n higher than the last — bears on course below + previous block, self-supporting at placement
  • Corbel 1–3″/course · hemispherical · 12″ wall + 24″ base band (courses 0–7, below grade) · +20 kN cap · batter
  • Flat beds + keyed insert ALL courses — single SKU, no course-tracking
  • Fallback: sand mound only if free corbelling proves too fussy

Water Defense — 7 Layers, All On-Site

  • 1 Siting — highest ground, never in a "dry" wash
  • 2 Diversion swale — contour ditch around both sides; spoil = levee; doubles as permaculture water harvest
  • 3 Rim collar 12–18″ — floor below grade, springline above — sheet flow can't climb in
  • 4 Flared apron — kills splash/ponding
  • 5 Air-gap lining — 1–2″ void behind wall (better than rubble, zero material) + drain channel to daylight
  • 6 Covered entry — stomion vault — corbelled vault roofs the inner stair run through the berm: doorway never sky-exposed; open outer run → drain sump; raised sill lip at door
  • 7 Infiltration floor — loose pavers + coarse sand
  • Residual: catastrophic flood → ~free soil + 3 days = rebuildable by design

Disaster Scorecard — Full Ancestral Toolkit

ThreatDefense
Seismic — VERIFIED PASS 5.0g12″ wall + 24″ base band · keys ALL courses · helix spiral · mound w/ 12″ cap · pit-buttress + soil-shear credits. ~1.7× the strongest recorded motion ever (~3g). Site ceiling = liquefaction susceptibility, not structure
FireBest-in-class — earth doesn't burn
Flood7-layer system; inundation remains fatal — siting IS the answer
Wind / heatOptimal shape · ground-coupled ~55–60°F
Freeze-thaw (SLV)Below-frost base · dense low-absorption brick · berm · durability gate
Snow / lightningSymmetric sheds · earth = ground

Igloo · clochán · trullo · tholos — every proven dry-stack technique, zero tensile elements, ~100% site soil, uniform units throughout

Cost · Schedule · Next Actions

Phase$
0 — soil gates (jars + soak tests)~20
1 — working plant~460–635
2 — first dome standing+68–108
3 — throughput parity pkg+450–900
4 — movers (per region, optional)2.5–15k
5 — automation (donor scrounge + Nano)+40–90
  • W1: messages + buy run + soil tests + test hole
  • W2: weld mold + mount on beam + plumb
  • W3–4: tune, durability gates, ~2,100 units
  • ~5–6 days: excavate → press → stack → mound → dome · strikes in ~4–6 h: remote-pole collapse into the pit (15-ft exclusion, nobody on the shell), mound backfills on top — returns to land clean · or ~1 day careful unstack to reuse bricks
  • Machine = splitter only: ~$650–750 core, reversible, still splits logs — slide 3
  • Next: send seller messages · measure door · dig test hole

Docs: ~/Documents/ceb-press/ · CAD: cad/ (OpenSCAD parametric + STL) · Calc: brick_calc.py

← → navigate · click to advance