Work you can check.
Studies we ran ourselves: transonic pressure against wind-tunnel data, hull sinkage against towing-tank measurements, and a data-center layout taken from geometry to a defensible recommendation. Every page shows the result and the boundary of what it claims.
How much cooling capacity does your layout cost you?
An independent study we ran on our own bench: four ways of arranging 42 dry coolers around a 44 MW-class data hall. The dense baseline quietly lost about 12% of its cooling capacity to its own exhaust; a better layout cut that penalty to about 5%.
- +2.46 KBEST INTAKE RISE
- 4LAYOUTS COMPARED
- 5 DAYSILLUSTRATIVE SCHEDULE
The λ-shock, where the wind tunnel puts it.
The ONERA M6 wing at Mach 0.8395 against AGARD AR-138. 270 of 271 pressure taps, and the double-shock structure captured and merging where the experiment brackets it. Lift sits 1.1% from the midpoint of a NASA multi-code band — a code-to-code check, not validation; there is no measured lift or drag for the M6.
- 0.071POOLED RMS ΔCp · 270 TAPS
- −1.14%CL VS BAND MIDPOINT · CODE-TO-CODE
- 12.96 MCELLS
Sinkage within 1.03 mm, across five speeds.
The 3 m Wigley hull from Fr 0.177 to 0.408, against the 1983 towing-tank measurements. Every sinkage result landed within 1.03 mm — but only after a constant +6.6 mm offset gave away a datum bug that no amount of physics tuning would have fixed. Total resistance is verified to ±2.10% on a converged grid family.
- ≤1.03 MMSINKAGE ERROR · FIVE SPEEDS
- ±2.10%CT NUMERICAL UNCERTAINTY
- 0.99–1.07SKIN FRICTION CV/CF
You don't need a finished brief.
Send the geometry, the operating point, and the decision you're trying to make, and we'll tell you what can be answered and the fastest defensible way to get there.