History: how light learned to count people

Why count. A shop needs two numbers. "Total in" — to see how many visitors became buyers: two hundred came in, fifty bought something — so something is wrong with the window or the prices. "Inside" — for safety: a hall, a museum, a club has a limit on how many people may be in at once. In 2020 such counters were put on shop doors everywhere. Our kit shows both numbers — and works the same way as the devices over real doors.

Becquerel1839light givesa currentselenium1873conductsin lightEinstein1905explainedthe effect"electric eye"1930sbeam + relaycountsphototrans.1950an eye withno valveIR LED1962Texas Instr.invisible beamCMOS 40001968counterand displayfrom "light gives a current" to the E and R pair over the door — 130 years
How the photosensor appeared and started counting people: the discovery, the explanation, the "electric eye" with a relay, the phototransistor, the invisible infrared beam and chips that count by themselves.

1839 — light gives a current. Edmond Becquerel noticed that lighting an electrode in a solution makes a current. 1873 — Willoughby Smith found that selenium conducts better in light than in the dark. The first photocells were made of selenium.

1905 — the explanation. Heinrich Hertz saw in 1887 that light knocks charges out of metal, but nobody knew why. Albert Einstein explained it: light comes in portions, and each portion can knock out a charge. He got the Nobel Prize for this — not for relativity.

Segment LCDAlbert Einstein1921, Vienna — at a lecture1905: light comes in portions — quanta.A quantum hits the metal and knocks out a charge:the brighter the light, the more charges,and so the bigger the current.The Nobel Prize in Physics for 1921 —"for the discovery of the law of thephotoelectric effect".Every photocell works by this law —and so does receiver R of our IR pair.
Albert Einstein, 1921 — on an LCD of 96 × 120 dots. After a photo by Ferdinand Schmutzer (public domain).

1930s — the "electric eye". A photocell, a valve amplifier and a lamp opposite. Cross the beam — a relay clicks: a door opens or a mechanical counter with digit wheels ticks. That is how parts on a conveyor were counted, and then people at the entrance. The weak point — one beam does not know which way a person goes. So there were separate doors for in and out, or two beams one after another — exactly as in our lesson 5.

doorway (top view)streetshop floorE1R1beam AE2R2beam B
Two beams across the doorway, top view. Emitter E shines across into receiver R. Beam A is nearer the street, beam B nearer the shop floor. Someone coming in blocks A first, then both, then only B; someone leaving — the other way round.

1950 — the phototransistor. John Shive at Bell Labs made an "eye" the size of a pea — no valve, no high voltage. 1962 — Texas Instruments engineers Bob Biard and Gary Pittman got the infrared LED: light the eye cannot see but a phototransistor can. The beam over the door became invisible — this is the E and R pair in the simulator.

1968 — CMOS 4000. RCA released chips on pairs of P and N transistors — exactly like our gates. They use almost no current while nothing switches: a device over a door can run for years from a weak supply. The series had the CD4029 counter and the CD4511 display decoder — the whole visitor counter now fitted on a small board.

The datasheet you build from. The CD4029B datasheet header says "Data sheet acquired from Harris Semiconductor", and the figure numbers — 92CS-…, 92CL-… — are RCA drawing numbers. The series went from RCA to Harris, from Harris to Texas Instruments, and the Fig. 9 circuit you carry into lesson 3 was drawn more than half a century ago.

Today. Cameras and thermal sensors hang over doors, with a microcontroller inside. But the idea is the same: two sensors, direction, counting. We build it on chips of the 1970s, where every step can be seen.