Bioreactor & Fermentor Fleet

Bioreactor & Fermentor Fleet (to 50,000 L) — Mixing, mass transfer, and control you can audit

Scope: Single-use mammalian stirred-tanks (development → 50 L & 250 L GMP), perfusion with cell-retention (ATF/TFDF), fixed-bed options for vector programs; stainless microbial/fungal line from benchtop and pilot through 10,000 L and 50,000 L production trains (CIP/SIP, pressure-rated gas, O₂ enrichment, methanol-safe). All reactors run inside a unified digital QMS (ALCOA+) with eBMR/eBR, LIMS/ELN, and cross-site CPV dashboards. Same recipe logic in San Diego and Montréal—kept dialed-in and audit-ready.

Why teams use our fleet

  • Operator-holdable ranges. We convert Stage-1 designs (kLa, P/V, shear envelopes, gas logic) into recipes with interlocks and alarms. PPQ lots feel routine, not heroic.
  • Predictable scale jumps. Geometric similarity, power per volume, and oxygen transfer modeling are verified up front; scale-down models predict—not explain after the fact.
  • Coverage without rework. Mammalian SU to perfusion to fixed-bed, plus stainless microbial up through 50,000 L—one control strategy, two hubs, no drama.
  • Evidence over optimism. We publish edge-of-failure, pressure–throughput curves, oxygen-enrichment limits, and membrane/TFF guardbands before validation.

Backgrounder: physics that actually govern success

Bioreactors don’t fail because “the clone changed.” They fail when kLa can’t keep up, P/V is mis-scaled, gas shear trashes viability, foam poisons filtration, or methanol/solvent envelopes aren’t engineered. We treat those as design variables—then hard-wire them into the batch logic and QMS so the floor runs them the same way every time (even on a Friday night when everyone should be at the beach).

The fleet—what’s on tap

Mammalian (single-use, perfusion-ready)

  • Development: 2, 5, 10, 20 L SU stirred-tanks with pitched-blade/elephant-ear impellers; capacitive/optical probes; gas blend (air/O₂/CO₂/N₂) via MFCs; PAT ports (Raman, dielectric).
  • GMP suites: 50 L and 250 L SU stirred-tanks (perfused or fed-batch), ATF/TFDF retention, sterile harvest manifolds, in-line pH/DO redundancy, recipe-driven antifoam.
  • Perfusion: residence-time control, bleed logic, cell-retention integrity tests, hold-up mapping; closed-loop transfers to downstream.

Vector production (adherent, suspension, fixed-bed)

  • Adherent: microcarriers where justified; harvest skids sized to prevent shear spikes.
  • Suspension: same SU family as above with gentle gas/sparge regimes.
  • Fixed-bed: production-relevant footprints with mass-transfer characterization and cleaning maps; dosing/harvest logic embedded in recipes.

Microbial & fungal (stainless, CIP/SIP)

  • Development/pilot: 5, 10, 30, 50, 100, 300, 1,000 L jacketed STBRs; Rushton/segment-blade trees; baffles & high-kLa spargers; back-pressure up to validated limits.
  • Production trains: 10,000 L and 50,000 L with O₂ enrichment, pressure-rated gas trains, antifoam controls, foam-knife options; methanol handling envelopes for Pichia (ATEX-aware); automated CIP/SIP with validated recipes.
  • Filamentous rheology: high-torque drives; staged sparging; viscosity-aware pressure limits; de-plugging-friendly filtration trains downstream.

(Exact counts, models, and utility loads are provided under NDA in the Equipment & Specifications dossier.)

Scale design—how we make the math real

  • Similarity rules: constant P/V windows for mammalian; constant Re/Fr checks where helpful; microbial targets by kLa and OUR balance with power and back-pressure ceilings.
  • Gas logic: cascaded air→O₂ blend→pressure; enrichment caps documented; CO₂ stripping windows for mammalian quality.
  • Mixing & shear: impeller/tip-speed limits enforced; sparger type and superficial gas velocity tied to viability/foam trends.
  • Foam discipline: antifoam feed under hard interlocks; nutrient/antifoam interactions tested to avoid downstream pain.
  • Instrument confidence: dual pH/DO or drift-guard procedures; PAT calibrated against reference assays before it’s allowed to steer anything.

PAT & automation (so each day is decisional)

  • PAT: Raman (glucose/lactate where proven), dielectric (biomass), off-gas analysis (O₂/CO₂), soft-sensors for OUR/OTR and growth rate; validated against reference analytics.
  • Control: PLC/SCADA or DCS with 21 CFR 11/Annex 11 compliance; recipe steps, interlocks, and alarms codified; exception capture in eBMR/eBR.
  • Historian & CPV: time-synced data historian; CPV dashboards charting titer, viability, metabolites, kLa proxies, pressure/flux, and residence-time.

Safety & EHS envelopes (engineered, not wished)

  • Methanol/solvent: ATEX-aware zoning; gas detection; inerting and ventilation; documented MOC; operator training; explosion relief where applicable.
  • Pressure: validated back-pressure limits and rupture-disc logic; alarm testing part of PQ.
  • Cleaning: CIP/SIP recipes validated; residues and bioburden acceptance; MACO/PDE cleaning limits; periodic re-qualification.

Integration to DSP & DP (start with the bottleneck)

  • Harvest rates & line sizing: pump NPSH, line IDs, and filter pre-stages sized to real broth; pressure–throughput curves in the file; surge/hold logic for downstream lag.
  • Closed transfers: sterile connectors/manifolds; steam-in-place stubs on stainless; bioburden/endotoxin trending before and after harvest.
  • DP cadence: for vector/LNP programs, buffer make-up, TFF, and filtration windows coordinated so no step waits “on hope.”

Validation & qualification—what we prove before PPQ

  • URS → DQ → IQ/OQ/PQ for every vessel and skid (impellers, spargers, MFCs, drives, probes).
  • kLa & mixing studies: scale-down/scale-up correlation reports; oxygen-enrichment limits; shear tests.
  • Boundary exercises: low/high gas, max antifoam, high cell-density (mammalian), high OUR (microbial), highest qualified flux to filtration; results and guardbands frozen into recipes.
  • Hold-time studies: seeds, intermediates, harvests; time/temperature limits; corrective-action trees.
  • CSV/CSA: audit trails, e-signatures, time sync (NTP), backup/restore; periodic review.

Species & modality notes (quick hits)

  • CHO/HEK: CO₂ management and osmolality shaping for charge/glycan windows; perfusion residence time validated for quality.
  • Adherent on microcarriers: shear-aware impeller profiles; verified harvest logic.
  • E. coli high-density: acetate avoidance via carbon-uptake caps; cooling capacity checks for induction shifts.
  • Pichia: glycerol→induction transitions with O₂/methanol envelopes; solvent sensors and EH&S sign-off baked into protocols.
  • Aspergillus: torque/viscosity mapping; staged aeration; filter trains that actually de-plug.

Materials & single-use—E/L and compatibility

  • Films & wetted parts: chemical compatibility files; extractables/leachables risk assessed; worst-case simulants used where warranted.
  • Connections: sterile welding, pre-sterilized manifolds, alternates qualified; integrity tests documented.
  • Inventory: mirrored kits at both hubs; barcode chain-of-custody in LIMS/ERP.

How scheduling & campaigns work (no surprises)

  • Changeovers: validated CIP/SIP cycle times; SU kit change windows; EM snap-back rules after maintenance.
  • Parallelization: dual-hub runs share standards and analytics; one QA disposition trail.
  • Utilities windows: gas/HPW/steam plans sized for peak; alarms escalate to Facilities + QA; RTO/RPO for data systems documented.

Program Onboarding (your first 30 days)

  1. Scale path & control strategy: target titers, CQAs, and candidate modes (fed-batch, perfusion, microbial batch/fed-batch/continuous) → proposed reactors with kLa/P/V envelopes and shear limits.
  2. Edge & boundary plan: low/high gas, enrichment caps, foaming windows; fixed-bed or perfusion residence-time tests; filtration pressure-time pilot.
  3. Harvest & downstream handshake: flow limits, transfer lines, filters, surge/hold logic; UF/DF/MF alignment.
  4. Safety envelope: methanol or solvent plan (if Pichia), pressure limits, alarm tests; training.
  5. Validation calendar: URS/DQ/IQ/OQ/PQ matrix, CSV/CSA items, and CPV chart set.
  6. Dual-site parity: equipment class mapping, recipe/alarm parity, method transfer metrics; comparability outline if both hubs will run.

You send: molecule class, expected titers/OUR, phase and timelines, and any historical reactor data. We return: a signed scale plan with ranges you can hold—and a calendar QA will approve.

Indicative timelines

  • Weeks 0–2: scale path, control-strategy draft, safety envelope; boundary/edge test protocols written.
  • Weeks 3–6: mixing/kLa studies, boundary exercises, CSV/CSA closure; first engineering lot.
  • Weeks 6–10: additional engineering lots; downstream/harvest coupling finalized; guardbands frozen into recipes.
  • Weeks 10–12+: PPQ readiness review; validation annexes (hold, filtration, cleaning) finalized; CPV dashboards live.

Deliverables (what you can hold)

  • Scale design report (similarity rules, kLa/P/V targets, boundaries, guardbands).
  • Reactor qualification pack (IQ/OQ/PQ, alarm/interlock tests, probe qualifications).
  • Safety dossier (methanol/solvent, pressure, ventilation, MOC).
  • Harvest/transfer spec (lines, pumps, filters, surge/hold).
  • Recipes in eBMR/eBR with setpoints, NOR/PAR, interlocks, and alarms.
  • CPV dashboard seed set and SOPs for periodic review.
  • Regulatory text for Module 3 process description and validation summaries.

FAQs (straight answers)

How do you size oxygen at 50,000 L? By OUR targets, power limits, enrichment caps, and back-pressure—tested at pilot with verified kLa correlations and carried in the recipe.
Perfusion or fed-batch? We choose what meets CQAs with the least complexity; if quality or cadence demands perfusion, we build residence-time and retention controls with guardbands.
Can you keep mammalian shear low while hitting titer? Yes—tip-speed caps, sparger selection, and gas velocity limits; antifoam logic proven not to punish DSP.
What about Pichia methanol safety? ATEX-aware design, detection, ventilation, and training—validated before induction; enrichment and pressure rules enforced.
How do you prove scale-down models? By back-predicting pilot performance with the SDM (kLa/P/V/RTD), then freezing correlation limits into the file.
Cross-site sameness? Same equipment classes or mapped deltas, recipe/alarm parity, shared standards, and CPV overlays; bridging lots if regulators expect them.

Conclusion

A great bioreactor fleet is not the sticker on the vessel; it’s the ranges, guardbands, and evidence that let operators run the process the same way every time—and let QA show exactly how we know. Our San Diego and Montréal hubs execute the same control strategy, with the same batch logic and analytics, so scale feels clean and inspections stay chill. If you want a scale plan that holds at 2 L and 50,000 L, we’ll map the physics, codify the recipe, and prove it—Design → Data → Decision, without detours.

MycoVista | San Diego, CA & Montréal, Canada
Start Program Onboarding → Share target titers, mode, and timelines. We’ll return a scale path, boundary tests, and a validation calendar you can run.