Cancer Informatics Collaboratory
Home / Datasets / Medical device CFD
Dataset · Medical device CFD · US FDA

FDA Blood-Flow Benchmark Datasets for CFD Validation

Two standard test devices, a nozzle and a centrifugal blood pump, with measured flow, pressure and blood-damage data. Use them to check a computational fluid dynamics (CFD) model of a blood-contacting device before you trust it.

The research behind this page

  • Prasanna Hariharan, PhD: Mechanical engineer at the US FDA Center for Devices and Radiological Health since 2008; PhD, University of Cincinnati. ✓ FDA speaker profile

These researchers created the original work. They did not write or review this page. ✓ marks a credential or role confirmed on an official profile; "as listed" roles come from the original page. All researchers featured on this site.

Based on: 8 peer-reviewed papers, 2 government sources, 2 official project or publisher sites, 3 archived original pages. See all sources

Get the data

Publisher
US Food and Drug Administration (FDA)
Benchmarks
2 released: nozzle and centrifugal blood pump
What's measured
Velocity fields (PIV), pressure, hemolysis, pump head vs. flow (H–Q)
Files
CAD geometry, flow conditions, experimental and simulation data
Use and license
Work of US government staff; the original project page stated it is in the public domain in the US
Where it's hosted
FDA Regulatory Science Tools catalog and GitHub
Links checked 21 Nov 2026.

What you can use it for

  • Validate a blood-flow CFD model. Run the benchmark geometry under the published conditions and compare your velocities and pressures with the measured ones.
  • Test blood-damage models. The hemolysis measurements let you check a damage model against real blood-damage data.
  • Build credibility evidence. FDA describes the dataset as a way to do early-stage validation before device-specific testing.

Limit to keep in mind: FDA notes the data covers only the benchmark shapes. A real device still needs its own verification and validation.[1]

What's in each benchmark

Benchmark 1 · round robin 2008–2009

Nozzle

A straight throat that narrows through a cone at one end and widens with a sudden step at the other. Flow can run either way, so one part tests both a gradual and an abrupt change in diameter.

  • Velocity measured by particle image velocimetry (PIV) in three laboratories
  • Blood-analog fluid with its refractive index matched to the nozzle
  • Velocity and pressure along the model at five flow rates (table below)
Benchmark 2 · second round robin

Centrifugal blood pump

A simplified pump with a rotor in a chamber, run at set flow rates and rotor speeds.

  • Flow mapped with PIV in acrylic copies of the pump
  • Hemolysis (blood damage) measured experimentally
  • Pump performance as an H–Q curve: pressure head against flow rate
  • Geometry, CAD files and flow conditions supplied as an instructions package

Nozzle flow conditions

Reynolds number in the throatFlow regime
500Laminar
2,000Transitional
3,500Turbulent
5,000Turbulent
6,500Turbulent

[2][3]

What the comparisons found

FDA compared 24 participants' simulations of the benchmark pump with the measurements. Results varied widely between groups, especially under transient conditions. For modellers, that means checking a model against the data at every operating condition that matters, not just one.[4]

Key publications

  1. 2011Hariharan P, et al. Multilaboratory particle image velocimetry analysis of the FDA benchmark nozzle model to support validation of computational fluid dynamics simulations. J Biomech Eng. 2011.
    DOI: 10.1115/1.4003440 · PubMed 21428676
  2. 2012Stewart SFC, et al. Assessment of CFD performance in simulations of an idealized medical device: results of FDA's first computational interlaboratory study. Cardiovasc Eng Technol. 2012.
    DOI: 10.1007/s13239-012-0087-5
  3. 2013Stewart SFC, et al. Results of FDA's first interlaboratory computational study of a nozzle with a sudden contraction and conical diffuser. Cardiovasc Eng Technol. 2013.
    DOI: 10.1007/s13239-013-0166-2
  4. 2015Herbertson LH, et al. Multilaboratory study of flow-induced hemolysis using the FDA benchmark nozzle model. Artif Organs. 2015.
    DOI: 10.1111/aor.12368 · PubMed 25180887 · Free full text (PMC)
  5. 2016Raben JS, et al. Time-resolved particle image velocimetry measurements with wall shear stress and uncertainty quantification for the FDA nozzle model. Cardiovasc Eng Technol. 2016.
    DOI: 10.1007/s13239-015-0251-9 · PubMed 26628081
  6. 2016Taylor JO, et al. Analysis of transitional and turbulent flow through the FDA benchmark nozzle model using laser Doppler velocimetry. Cardiovasc Eng Technol. 2016.
    DOI: 10.1007/s13239-016-0270-1 · PubMed 27350137
  7. 2017Malinauskas RA, et al. FDA benchmark medical device flow models for CFD validation. ASAIO J. 2017.
    DOI: 10.1097/MAT.0000000000000499 · PubMed 28114192

Identifiers checked against PubMed and the publishers' DOI records, 22 Nov 2026.

Who uses these benchmarks

The working group tracked 35 outside studies built on the nozzle and pump data, including the ASME V&V 40 standard on the credibility of computational models for medical devices, and validation reports from COMSOL, Ansys LS-DYNA and Pointwise. Work that links to this page:

Sources

  1. FDA Regulatory Science Tools catalog · www.fda.gov · Government source
  2. Original nozzle page (archived copy) · web.archive.org · Archived original
  3. Original pump page (archived copy) · web.archive.org · Archived original
  4. FDA, interlaboratory comparison of CFD simulations of the benchmark blood pump (PDF) · www.fda.gov · Government source
  5. The working group's citation list (archived copy) · web.archive.org · Archived original

About this page

This page started as the home of FDA's Critical Path CFD/Blood Damage project wiki, hosted on NCIP Hub from 2016 until NCI retired the hub in 2024. The original wiki page was started by Prasanna Hariharan (created Sep 2016, last edited Oct 2018). We rewrote it to focus on the data and where to get it now.