Research & institutes

Every build is a data set. Treat it like one.

Euler turns the layer images your printer already saves into data you can plot, compare and export, layer by layer.

Ten watch-case frames in 3D, with the defects found glowing where they sit, and the build's per-layer data beside them
watch-case-frames-wf-12.csv
layerseverityspatterstreaking
615minor71
616minor91
617critical146
618critical138
619critical158

In the lab

Measure the build without building the instrument

Every layer of every build is analyzed from the images your printer already takes. No sensors to buy, no code to write and nothing to calibrate before the first experiment.

A metal 3D printer with its operator PC: Euler Connect on the PC sends each layer image to Euler in the browser, where it is marked up

On the machines you already have.

Connect a printer and layers arrive as they print, or upload a build you have already finished.

  • Example of obstruction on a layer image
    Obstruction
  • Example of warpage on a layer image
    Warpage
  • Example of pinning on a layer image
    Pinning
  • Example of spatter on a layer image
    Spatter
  • Example of smoke on a layer image
    Smoke
  • Example of streaking on a layer image
    Streaking
  • Example of hopping on a layer image
    Hopping
  • Example of hole on a layer image
    Hole

Defects where they happened.

Researchers step through a real build and see the defects on the image, not on a slide.

Where defects sit inside the part.

The build in 3D, reconstructed without CAD, so a cluster on one part or one side of the plate is plain to see.

Compare builds

Which parameter set stands out, and why

Compare a group of builds and Euler tells you which ones stand out, and why. Standing out is not always bad. A build can stand out because it has fewer defects.

TypicalSlightModerateStrongBuild 09 · Set E
Build 07 · Set C
Flared bushing · IN718
Strong
Obstruction
Warpage
Spatter
Streaking

Five or more builds, one map.

Pick the builds and see what makes one different from the rest.

Findings
2 of 10 builds stand out
1 strong, 1 moderate
Strong · 1
  • Build 07 · Set CSpatter
Moderate · 1
  • Build 09 · Set EWarpage
Deviation
  • Typical8
  • Slight0
  • Moderate1
  • Strong1
Print jobs · 10Deviation
  • Build 07 · Set CSpatter
  • Build 09 · Set EWarpage
  • Build 01 · Set A
  • Build 02 · Set A
  • Build 03 · Set B
  • Build 04 · Set B
  • Build 05 · Set D
  • Build 06 · Set D
  • Build 08 · Set C
  • Build 10 · Set E

Tagged the way you run the study.

Tag builds by parameter set, material or student project, and see how far each one deviates.

Exposure#184
Exposure image of layer 184 of Build 08 · Set C, with its detections
Build 08 · Set CTypical
Exposure#184
Exposure image of layer 184 of Build 07 · Set C, with its detections
Build 07 · Set CStrong

Then look at the layers.

Open two builds on the same layer, side by side, and see the difference for yourself.

Your data, your tools

Chart layer data next to your own logs

Everything Euler records about a build is yours to take out. Line it up with your own machine logs and plot the two together.

Using Euler with an AI assistant
  • layer-series.csvOne row per layer
  • detections-1-1000.csvMarked defects, layers 1–1,000
  • detections-1001-1107.csvMarked defects, layers 1,001–1,107
  • report.jsonThe build report

Export the data.

Per-layer results and every detection with its position on the plate, as CSV or through the API.

Spatter last week
Which builds last week had more spatter than usual?
Read from Euler

Three builds had more spatter per layer than the rest. The turbine bracket had about twice the average.

Spatter per layer, against the average
  • Turbine bracket TB-42.1×
  • Pump housing PH-61.4×
  • Heat exchanger HX-21.2×
Reply to Claude…
Euler

Or ask for the chart.

Ask Claude or ChatGPT, and it pulls the data and draws the chart for you.

Customers

What customers say, and what was measured

DTI logo
“Euler is close to perfect in spotting the issues. We have already adopted Euler on all our machines and the vision is to switch completely to automatic QA reporting from machine data.”
Portrait of Jeppe Byskov
Jeppe Byskov
Director of Additive Manufacturing, DTI
Danish Technological InstituteWhite paper27.8%of all print time in the study went to builds that failed.Many of those were experiments that push the process on purpose.Read the white paper
“There's a lot of things that can go wrong during the process especially for an RTO like DTI where we're really trying to push the boundaries of what's possible.”
Danish Technological Institute · Customer Spotlight

FAQ

Before you connect a printer

Start with the free plan

Connect a printer, upload a finished build, and have two of them analyzed on us.

No credit card.