Buoyancy manufacturers
Our initial focus: one formulation or supplier-change decision, using the manufacturer's ingredients, process range and historical test evidence.
Subsea buoyancy · material development
Choose what
to make.
Plan what
to test.
Explore formulation tradeoffs, compare material changes and use reviewed test results to choose the next experiment.
Inside the materialHollow glass microspheres
MICROSTRUCTURE STUDY / 001Illustration · not simulation data
RESEARCH INTO PRACTICE
Research guidance from Prof. Nikhil Gupta · Composite Materials and Mechanics Laboratory · NYU Tandon
01 / The platform
A material change.
A clearer next step.
Explore exact filler loadings against your density and stiffness requirements. Compare candidate tradeoffs, then bring the recipes and source evidence behind a material change into a saved engineering case.
Plan a first compression-test campaign without local test data, clearly labelled as literature screening. As reviewed batch results arrive, build a local model and plan the next round with that evidence.
01.1 / Formulation engine
Explore the
tradeoffs.
Evaluate exact loadings with the fitted model. Compare density and stiffness, inspect uncertainty and export your candidate.
Open the formulation engine01.2 / Material change review
Bring the evidence
to your team.
Compare recipes and source documents, then turn a saved case into an evidence brief with gaps and proposed tests.
Review a material changePlan a first campaign or the next tests
The engine currently models one listed filler grade at a time. Local updates learn compression modulus within a declared process window; pressure performance still requires direct tests.
02 / Starting beneath the surface
Built around
the next test.
Subsea materials face real pressure. Connect your service requirement with candidate estimates and a laboratory request. Direct pressure tests establish hydrostatic performance.
Plan a pressure testDSV ALVIN · NOAA PHOTO LIBRARY
03 / The workflow
From formulation targets
to the next experiment.
01
Compare your inputs
Bring your recipes and supplier documents. Review exact ingredients, process changes and compatible source evidence, keeping missing information visible.
02
Evaluate exact loadings
Set density and stiffness limits. Evaluate the fitted model between catalogue points, compare tradeoffs and export the exact candidate with its model version.
03
Plan the next tests
Start a first campaign from literature predictions, or continue with a model fitted to reviewed batches. Set targets and export a worksheet that identifies its evidence and exact loadings.
04
Share the decision brief
Save a case revision and export its evidence, calculations, gaps and next-test plan as a PDF. Reopen the case as the work develops; an edited case needs a new brief.
1,781
stored literature records from 130 papers; mixed provenance, not all independent measurements
5–60%
particle volume fraction: direct model predictions at arbitrary loadings within this range
16.4 %
median stiffness error on 15 held-out literature campaigns; 58% within ±20%
PDF
a shareable evidence brief with source excerpts, calculations, gaps and next tests
Evidence you can inspect
Model benchmarks come from literature cross-validation, with unresolved source independence. Nominal 90% intervals cover 72.9% of modulus and 74.9% of uniaxial-strength observations; they are not calibrated guarantees. Catalogue candidates have no factory or hydrostatic validation. Full benchmark, scope and limitations.
05 / One development workflow
Built around
a real material decision.
Vehicle and system teams
Define service requirements and acceptance criteria with the material manufacturer. Published vendor grades provide context, not validation of our candidates.
Microsphere and resin suppliers
Contribute traceable ingredient specifications and application data to a manufacturer's evaluation. Supplier comparisons require comparable test methods.
The formulation engine supports modulus estimates for four polymer matrices with ten legacy glass-microsphere grades at 5–60 vol%; strength requirements are supported only for epoxy. Material Change can record other ingredients and compare compatible source evidence without claiming model predictions for them. Inspect model coverage.
Build the evidence together
Bring your material.
And your next question.
Start with one formulation or supplier change and the records you are permitted to use. We first assess data suitability, then agree an evaluation against your current method. Accuracy and savings must be measured.
Explore the pilot