steam sensor/Using it/09. When the reading is wrong
Using it · 09 of 9

When the reading is wrong

Five things go wrong, and the circuit says where to look for each: 0 even when wet is VCC; a wandering reading is the SIGNAL wire or GND; a reading that will not come down is something left on the loops; full scale is too much water or the wrong VCC; and steam that reads 0 has not condensed on the board.

Start with a wet finger

The quickest test needs no steam. Lay a damp fingertip across the loops: the reading should go most of the way up at once, and fall to 0 within a second or so of wiping the board dry. If it does, the wiring and the sketch are right, and anything else is about the water.

What it reads, and where to look
What you see
Water can only lift SIGNAL towards VCC, so if VCC never arrives, a soaked board still reads 0. Count from the square pad, GND, VCC, NC, SIGNAL, and check VCC goes to 3V3 or 5V. If VCC comes from a pin, check the sketch drives it HIGH before reading. Still 0: try tap water on a fingertip; very pure water reads very little.

Pick what you see. The circuit marks the part that causes it, and the verdict says where to look.

0: the supply

Everything on this board is a divider hanging from VCC. With no VCC there is nothing to divide, and the most sodden board reads 0. Count from the square pad and check VCC. If it comes from a pin, the sketch has to drive it HIGH and wait 500 ms first.

Drifting: the SIGNAL wire and GND

A dry board holds SIGNAL at 0 V firmly, through the 1 MΩ to GND. A reading that wanders with nothing on the loops means the pin is not connected to SIGNAL, or GND is missing. Check STEAM_PIN, and that the wire is not on NC.

Stuck up: the surface

A reading that stays up for minutes after the steam has gone is water, or something that holds water, still on the loops. The circuit forgets in half a second. Clean the loops with isopropyl alcohol, dry them, and store the board dry.

0 in a steamy room: the physics

If the room is full of steam and the reading is 0, the board is working. It reads water on its surface, and steam only lands there when it condenses on a surface cooler than itself. Put the loops in the path of the steam, and see water, not air.

When it does not work

It reads 0 with the loops soaking wet.

Water can only raise SIGNAL towards VCC, so no VCC means 0. Count from the square pad, GND, VCC, NC, SIGNAL. If VCC comes from a pin, check the sketch drives it HIGH and waits 500 ms before reading.

The reading drifts with nothing on the board.

An analog pin connected to nothing drifts. SIGNAL is not on the pin STEAM_PIN names, or it is on NC, or GND is missing, so the 1 MΩ has nothing to hold SIGNAL down to.

It will not come back down after the board dries.

Something still bridges a gap: a drop trapped between two loops, or a film of salt, flux or grime. Rinse the loops with isopropyl alcohol and let them dry. The circuit itself empties in about half a second.

It is pinned at full scale on my ESP32.

Either the loops are soaked with conductive water, or VCC is on 5V, which can put up to 5 V on a 3.3 V pin. Move VCC to 3V3 first, then dry the board.

Where this goes next

Temperature and relative humidity, measured in the air rather than on a wet surface.

For the air itself: the TK39 DHT22 →

Edit this page — content/books/steam-sensor/when-the-reading-is-wrong.mdx

Community

Questions about this product

See what other owners have asked, and read their solutions.

Ask a question ↗

Steam Sensor

Loading discussions…

Discuss this article

Ask about this page. The answer stays here, on the page it belongs to, for whoever hits the same wall next.

Browse Modules and blocks on the forum →