Onflux Labs starts, funds, and keeps companies built on results nobody was looking for.
Four times at Bell Labs, the noise was the finding.
Each began as a measurement that disagreed with what everyone expected. Each needed a second reader to say what it meant.
1927
Clinton Davisson and Lester Germer
- Looking for
- How electrons bounce off nickel, to build better vacuum tubes.
- Got
- A lab accident recrystallized the nickel. At 54 volts the electrons came back in a sharp spur at 50°.
- Read by
- At a 1926 Oxford meeting, Davisson heard Max Born cite his own earlier curves as evidence for de Broglie’s waves.
Electrons are waves. Nobel Prize, 1937.
Fig. 3, redrawn. Davisson and Germer, Bell System Technical Journal, 1928. 1932
Karl Jansky
- Looking for
- The sources of static on transatlantic radiotelephone calls.
- Got
- Under the thunderstorms, a faint steady hiss. It peaked each time his rotating antenna faced the same patch of sky, about four minutes earlier every day.
- Read by
- A colleague who studied astronomy, A. M. Skellett, saw that four minutes a day is the gap between solar time and star time.
Radio waves from the center of the Milky Way. Radio astronomy begins.
His record for September 16, 1932, 6 to 9 PM, redrawn. Proc. IRE, 1933. 1954
Daryl Chapin, Calvin Fuller and Gerald Pearson
- Looking for
- Power for telephone equipment in remote, humid places. The best photocells turned under 1% of sunlight into electricity.
- Got
- Silicon that Fuller had doped for transistor work put out a strong current when Pearson put it under a lamp. Its response (A) sat across the brightest part of sunlight (B).
- Read by
- Pearson told his friend Chapin to stop working on selenium.
The silicon solar cell, at 6%.
Fig. 1, redrawn. Chapin, Fuller and Pearson, Journal of Applied Physics, 1954. 1964
Arno Penzias and Robert Wilson
- Looking for
- A quiet radio antenna for astronomy, with every source of noise accounted for.
- Got
- After every known source was subtracted, 3.5 degrees of noise was left, from every direction, in every season. They evicted the pigeons nesting in the horn. The noise stayed.
- Read by
- Bernard Burke told them a Princeton group was predicting heat left over from the Big Bang.
The cosmic microwave background. Nobel Prize, 1978.
Antenna temperature budget at 4080 Mc/s, degrees Kelvin Noise measured, pointing straight up 6.7 ± 0.3 minus the atmosphere 2.3 ± 0.3 minus losses in the antenna .8 ± 0.4 minus pickup from the ground < .1 Left over, from everywhere 3.5 ± 1.0 Their noise budget at 4080 Mc/s, in °K. Astrophysical Journal, 1965.
We build and hold companies that start from a second reading.
We start
Every company we start begins with a result that did not fit the curve, and someone who knew what it meant.
We hold
We own for the long run. No fund cycle and no exit date on anything we start.
We keep the record
Negative results, raw data, and lab notes stay readable, so the next anomaly is not filed away as noise.