Engineering Guide · Vendor-Neutral

Wave Bioreactor vs Orbital Shaker: Which Surface-Aerated System Should You Use?

Published October 6, 2026 · 13 min read
Wave bioreactor rocking a pillow bag versus an orbital shaker spinning a cylindrical bag — both surface-aerated Wave / Rocking O₂ ±5-12° @ 6-40 rpm Pillow bag · 2D rocking Surface-renewed by travelling wave VS Orbital Shaken O₂ 25-50 mm @ 50-200 rpm Cylindrical bag · orbital Surface-renewed by rotating ring Both are surface-aerated · no sparger · kLa comes from surface renewal rate × surface-to-volume ratio
Figure 1: Rocking (left) and orbital shaking (right) share the same aeration physics — oxygen enters only through the gas-liquid interface — but renew that surface by different mechanisms.
Quick Verdict

Both formats aerate through the headspace rather than a sparger, so kLa is set by surface renewal and surface-to-volume ratio, not gas flow. Rocking (Cytiva Xuri W25, Sartorius Biostat RM) owns the regulated-CGT installed base and scales to ~300 L in a pillow bag; orbital shaking (Kuhner SB-X) often reaches higher kLa at bench scale and is the natural step up from an Erlenmeyer flask for microbial and HEK293 transient transfection work.

Key differences at a glance

Side-by-side comparison

Factor Wave / Rocking Orbital Shaken
Mixing principle 2D rocking of a pillow bag; a travelling wave renews the surface Orbital motion of a cylindrical bag or vessel; a rotating liquid ring renews the surface
Typical agitation 6-40 rpm · ±5-12° angle 50-200 rpm · 25-50 mm shaking diameter
Typical kLa (10-50 L) 5-30 h⁻¹ 10-80 h⁻¹
Power input per volume ~60-200 W/m³ ~100-680 W/m³
Shear rate (max reported) 10-100 s⁻¹ 20-200 s⁻¹
Single-use ceiling Up to 300 L working (Cytiva WAVE 500) Up to 200 L working (Kuhner SB200-X)
Scale-up path from flask Flask → rocker is a format change (no Büchs correlation) Erlenmeyer → Kuhner SB-X preserves orbital physics
Regulated CGT installed base Approved CAR-T (Kymriah, Yescarta, Breyanzi) Published for CHO/HEK293 but limited GMP footprint
Capital (50 L working-volume platform) ~$70-120k ~$80-150k

Values reflect typical published specifications. Your vendor's current datasheet takes precedence; kLa figures in particular depend strongly on fill volume, geometry, and medium.

Wave / rocking bioreactor in detail

The wave bioreactor was commercialised in 1999 by Vijay Singh at Wave Biotech, acquired by GE Healthcare (now Cytiva) in 2007 and sold today as the Xuri Cell Expansion System and Xuri W5/W25/W500 bioreactors. Sartorius sells a competing 1D-rocking platform as the Biostat RM, Thermo Fisher sells the HyPerforma Rocker, and the CELL-tainer family goes a step further with 2D (rock + sway) motion that lifts kLa well above the 1D ceiling. The vessel is a single-use pillow-shaped bag that sits on a motorised platform. The platform tilts along one axis in classic Wave-style motion — at a rock angle of ±5-12° and a rate of 6-40 rpm. The angular motion forms a travelling wave on the gas-liquid interface, which continuously renews the surface film and transfers oxygen from the headspace into the bulk.

How it works

Because there is no impeller or sparger, mass transfer is set by three variables: rock angle, rock rate, and fill volume. All three affect how much of the liquid surface is renewed per unit time. The relationship is empirical rather than correlated to Van't Riet, and typical published kLa values sit between 5 and 30 h⁻¹ for working volumes of 10-50 L. Oxygen is supplied by sweeping gas through the headspace via a vent filter at 0.1-0.5 vvm. Carbon dioxide stripping is proportional to the oxygen transfer, which can be a problem in dense cultures where pCO₂ climbs above 100 mmHg and inhibits CHO growth; sparge-port modifications are available on some RM and Xuri models to add a direct gas path for CO₂ removal at higher densities.

When wave wins

Rocking is the format of choice when the shear-and-regulatory profile matters more than peak kLa. The maximum shear rate in a well-tuned rocking bag sits around 10-100 s⁻¹, two to three orders of magnitude below the impeller tip of a stirred tank, which preserves VSV-G lentivirus envelopes and T-cell viability. Just as importantly, the installed base for approved autologous CAR-T products (Kymriah, Yescarta, Breyanzi) runs on the Xuri platform, which means switching formats in a running clinical programme introduces a regulatory risk that few sponsors take. Rocking also scales further on a single-use bag — the Cytiva WAVE 500 handles 300 L working volume in a 500 L pillow — so a mid-size CGT programme can keep the whole upstream chain on one format.

Orbital shaken bioreactor in detail

Orbital shaken bioreactors are the industrial scale-up of the Erlenmeyer flask. The dominant vendor is Adolf Kühner AG with the SB-X series (SB10-X, SB50-X, SB200-X). The vessel is a cylindrical single-use bag or reusable vessel sitting on a platform that orbits — but does not tilt — at a shaking diameter of 25-50 mm and a rate of 50-200 rpm. The orbital motion forms a rotating liquid ring inside the vessel that renews the surface from the sides rather than the top. The Infors HT Multitron and the Enzyscreen GrowthProfiler cover the small-scale screening end of the same physics.

How it works

Shaking diameter, shaking rate, and fill volume together determine the Froude number and the surface renewal mode. At in-phase conditions, the liquid behaves as a well-mixed rotating ring and kLa scales favourably with increasing rpm; at out-of-phase conditions, the liquid sloshes incoherently and kLa collapses. Published studies on round EVA bags at bench scale report kLa values well above those of a Wave bag at matched working volume, in some cases roughly double; other studies find rocking competitive or ahead. The honest reading of the literature is that neither geometry has an inherent advantage — fill fraction, bag material, and the presence of surfactants in the medium each shift the ranking by more than the vendor-reported means. For microbial Erlenmeyer flasks, Büchs and colleagues developed the RAMOS respiration activity monitoring system, which measures OTR directly and remains the standard way to characterise orbital-shaken OUR without a dissolved oxygen probe.

When orbital shaking wins

Orbital shaking is the natural step up from a shake flask for microbial and HEK293 transient transfection work. The physics are preserved (Erlenmeyer → benchtop Kuhner SB10 → Kuhner SB50 → SB200), which means Büchs-correlation scale-up predictions from the screening phase carry across. For E. coli campaigns running at OD600 10-30, Pichia pre-induction, and HEK293 transient AAV packaging under 50 L, orbital shaking typically delivers higher OTR per invested dollar than a rocking bag of equivalent working volume. The reusable vessel option (vs single-use bag only on rocking) also matters for academic and PD groups running many media or clone sweeps where single-use consumables dominate the cost per experiment.

Pros and cons

Wave / rocking bioreactor

Advantages

  • Lowest shear of any single-use format — preserves VSV-G lentivirus and T-cell viability.
  • Regulated-CGT installed base: Kymriah, Yescarta, and Breyanzi all use the Xuri platform for T-cell expansion.
  • Single-use ceiling of 300 L working volume in a 500 L pillow (Cytiva WAVE 500) covers most clinical and CGT programmes.
  • Mature online DO/pH options via optical patches, Xuri Fluorescence Reader, and single-use probes.

Disadvantages

  • Lower peak kLa than a sparged stirred tank or a high-rpm orbital shaken bag — caps microbial OD600 around 10-30.
  • pCO₂ accumulation at high cell density unless a sparge port is retrofitted.
  • No direct Büchs-correlation scale-up path from Erlenmeyer flask screening data.
  • Single-use bag only — no reusable alternative, so consumable cost per experiment rises for PD campaigns running many conditions.

Orbital shaken bioreactor

Advantages

  • Higher kLa headroom per unit volume than rocking at bench scale (10-80 h⁻¹ vs 5-30 h⁻¹).
  • Scale-up from Erlenmeyer flask preserves the orbital-shaking physics — Büchs correlations carry across.
  • Reusable-vessel option on some Kuhner SB-X and Infors Multitron Pro configurations, cutting consumable cost in PD.
  • Strong fit for microbial seed culture (E. coli, Pichia pre-induction) and HEK293 transient transfection.

Disadvantages

  • Thin regulated-CGT installed base — few approved GMP processes run on an orbital shaken single-use bioreactor.
  • Out-of-phase operation at the wrong rpm/fill combination collapses kLa unpredictably.
  • Shear rates at the top of the rpm range (180-220 rpm) approach the lower end of stirred-tank shear.
  • Commercial single-use ceiling of 200 L working volume (Kuhner SB200-X) — above that, the surface-aeration regime runs out.

Which should you choose?

Pick based on the dominant constraint in your process. Most of the choice collapses to installed base, scale, and the format your screening data came from.

Autologous CAR-T, 1-25 L

Clinical T-cell expansion for individual patient batches. Shear-sensitive, low OUR, inherits the regulatory precedent of approved products.

Choose Wave / Rocking

HEK293 transient AAV, 10-50 L

Preclinical and early-clinical AAV packaging where OTR is the bottleneck and the screening deck ran in Erlenmeyer flasks.

Choose Orbital Shaken

E. coli seed, OD600 ≤ 30

Microbial seed train or low-density preclinical expression. kLa headroom and clean Büchs scale-up matter more than regulatory legacy.

Choose Orbital Shaken

Lentivirus packaging, 20-200 L

VSV-G pseudotyped lentivirus envelope is shear-sensitive; stirred-tank tip speeds destroy infectivity. Scale ceiling under 300 L fits rocking.

Choose Wave / Rocking

Real-world use cases

Typical setups where bioprocess teams have converged on one choice or the other.

CAR-T · T-cell expansion
Autologous clinical batch

Cytiva Xuri W25 at 5-20 L working volume, 6-15 rpm, ±6° angle. Perfusion-ready via Xuri Cell Expansion Chamber. The format used in Novartis Kymriah and Kite Yescarta manufacturing.

AAV · HEK293 transient
50 L preclinical packaging

Kuhner SB50-X at 20-30 L working volume, 90-120 rpm, 50 mm diameter. OTR high enough for triple-transfection at 1-2 × 10⁶ cells/mL; scale-up from 500 mL Erlenmeyer preserves orbital physics.

E. coli · seed train
N-2 seed into 500 L STR

Infors HT Multitron Pro with 2 L Erlenmeyers, 180 rpm, 25 mm diameter, feeding a Kuhner SB50-X N-1 seed and then transferring to a stirred-tank production vessel. RAMOS validates OTR at each step.

Insect cell · baculovirus
Sf9 vector production, 10-50 L

Sartorius Biostat RM 20 or Kuhner SB50-X. Sf9 cells tolerate higher rpm than CHO and both formats work; the deciding factor is usually which vendor the facility has qualified.

Not sure how your rocking or orbital bag scales to a stirred tank?

Our scale-up calculator takes your bench kLa, OUR, and target working volume and returns rpm, tip-speed and gas-flow setpoints for the next vessel in the train. Works for Wave/RM → STR and Kuhner SB-X → STR transfers.

Open the Scale-Up Calculator

Cost and lifecycle considerations

Three cost components, two very different consumable models

Capital on the platform (rocker or shaker + controller) is broadly similar at matched working volume. The split shows up in consumables: rocking is single-use bag only, while some orbital shaken vessels can be reused for PD. Over a 3-year horizon running one bag per week, that difference is $50-150k.

At a 50 L working volume, a Cytiva Xuri W25 system with a controller lands around $70-120k, with each 50 L pillow bag at $350-700 depending on sensor options and perfusion fittings. A Kuhner SB50-X with controller is $80-150k, with a 50 L single-use bag at $400-900, or a reusable vessel at $8-15k amortised over hundreds of runs.

Operating cost is dominated by bags on both platforms if you run GMP (every batch gets a new bag by policy), but PD groups can cut per-experiment cost on the orbital side by moving to a reusable vessel. Over 3 years at 1 run/week, the TCO difference is $50-150k in the orbital's favour for PD, and roughly a wash under GMP.

Cost component Wave / Rocking (50 L) Orbital Shaken (50 L)
Platform + controller$70-120k$80-150k
Consumables per batch$350-700 (bag only)$400-900 (bag) or $0 (reusable)
Calibration / maintenance (yr)$4-8k$4-10k
3-year TCO · 50 batches/yr · GMP$220-320k$240-360k
3-year TCO · 50 batches/yr · PD reusable$220-320k$150-230k

Vendor landscape

Major vendors in each camp, with one-line positioning notes.

Wave / rocking vendors

Orbital shaken vendors

Frequently asked questions

What is the difference between a wave bioreactor and an orbital shaker?
Both are surface-aerated single-use systems with no sparger, so oxygen enters only through the headspace. A wave bioreactor sits a pillow-shaped bag on a platform that rocks it ±5-12° at 6-40 rpm, forming a travelling wave that renews the surface. An orbital shaken bioreactor sits a cylindrical bag or vessel on a platform that orbits at 50-200 rpm with a 25-50 mm shaking diameter, forming a continuously-rotating liquid ring that renews the surface from the sides. Rocking has larger single-use ceilings (300-1,000 L working volume); orbital shaking tops out around 200-500 L but reaches higher kLa per unit volume at bench scale.
Which has higher kLa — rocking or orbital shaking?
The published literature disagrees, which is itself the main finding. For a 20-50 L working-volume bag, 2D rocking (Wave-style) can reach 20-30 h⁻¹ at 25 rpm × 10°; a round EVA bag on an orbital shaker at 100 rpm × 50 mm diameter has been reported above 60 h⁻¹ in some studies and below 15 h⁻¹ in others — the answer is sensitive to fill volume (typically 30-50% of total bag volume), vessel geometry, aspect ratio, and surfactants in the medium. Neither format has an inherent kLa advantage; both are set by surface renewal rate and surface-to-volume ratio, not by gas flow. The practical rule is to kLa-match empirically for your specific bag + medium combination rather than quote a vendor-reported value.
What is the maximum working volume for each format?
Rocking: Cytiva Xuri W25 goes to 25 L working volume, W5 to 100 L, W500 to 300 L working in a 500 L bag. Sartorius Biostat RM runs 1-100 L working. Thermo HyPerforma Rocker runs 5-25 L. Orbital shaken bioreactors: Kuhner SB10-X (10 L), SB50-X (50 L), SB200-X (200 L working volume) — the current commercial ceiling. Beyond ~200-500 L, surface-to-volume ratio falls below the point where surface aeration alone can keep up with cellular OUR, and vendors recommend transferring to a stirred-tank single-use bioreactor.
Which is lower shear — wave or orbital?
Both are low shear because neither has an impeller. Reported maximum shear rates in rocking bags are 10-100 s⁻¹ and in orbital shaken bags 20-200 s⁻¹, compared to 500-5,000 s⁻¹ at the impeller tip of a stirred tank. For VSV-G pseudotyped lentivirus, suspension CHO, and T-cell expansion, both platforms preserve functional titer and viability where a stirred tank would not. Orbital shaking at the top of its rpm range (180-220 rpm) in a baffled flask or non-baffled bag can approach the lower end of stirred-tank shear, which is why protein-expression E. coli campaigns in Erlenmeyer flasks are routinely run at those speeds.
Why do most T-cell therapies use wave bioreactors rather than orbital shakers?
Clinical T-cell expansion for Kymriah, Yescarta, and Breyanzi uses the Cytiva Xuri Cell Expansion System W25, a wave-mixed platform originally commercialised as the WAVE Bioreactor by Vijay Singh at Wave Biotech in 1999. The expansion is low-OUR and shear-sensitive, both of which rocking handles well, and the regulatory and user-training history for autologous CGT programmes is on the Xuri platform. Orbital shaken bioreactors are a credible alternative on OUR grounds and have been published for CHO suspension and HEK293 transient transfection, but they lack the installed base in approved cell therapy processes, so programmes rarely take the regulatory risk of switching platforms mid-development.
What is the power input per volume for rocking and orbital shaken bioreactors?
Published P/V values for surface-aerated single-use bioreactors sit in the range ~50-700 W/m³, overlapping with stirred-tank P/V for suspension CHO (typical 10-200 W/m³) and lower than microbial P/V (1,000-5,000 W/m³). Rocking at 25 rpm × 10° in a 10-50 L bag delivers roughly 60-200 W/m³. Orbital shaking at 90-150 rpm × 50 mm diameter in a 10-200 L bag delivers roughly 100-680 W/m³ depending on fill volume. Both scales are below the mechanical damage threshold for most mammalian cells, which is one of the reasons surface-aerated single-use is dominant for CGT seed trains.
Can you run E. coli or microbial fermentation in a wave bioreactor?
Yes, but with lower optical densities than a stirred tank. The ceiling is the surface-aeration kLa (5-30 h⁻¹ for rocking, up to 50-80 h⁻¹ for orbital in favourable geometries), which caps OUR around 50-150 mmol O₂/L/h. That is enough for low-cell-density E. coli expression (OD600 10-30) and most Pichia pre-induction phases, but not for high-cell-density fed-batch fermentation (OD600 100+). Orbital shaking has a modest edge for E. coli because kLa typically scales better in a cylindrical geometry, which is why the installed base of microbial Erlenmeyer and Kuhner SB-X runs is larger than for rocking bags in microbial campaigns.
Which is better for scale-up to a stirred-tank bioreactor?
Both platforms require re-tuning when transferring to a stirred-tank — neither has an impeller, so P/V and tip-speed criteria don't translate, and the scale-up path is to kLa-match and re-establish DO, pH, and feed control in the new vessel. Shake flasks have the richest published scale-up correlations (Büchs correlation, RAMOS-derived OTR targets), so for microbial processes the orbital route provides a smoother Erlenmeyer → Kuhner SB-X → stirred-tank path. For mammalian seed trains, Cytiva and Sartorius both publish scale-up guidance from Xuri/Biostat RM into their respective single-use stirred tanks (Xcellerex XDR, Biostat STR); the hybrid pattern (wave seed → stirred production) is the industry standard for commercial AAV and mAb.

Resources and references