Opensource DUCO Humidity and temperature sensor. (known as DUCO Vochtsensor 0000-4218)

Duco Vochtsensor

Two days ago I installed a new ventilation system in my home. My product of choice: DucoBox Silent 400 m3/h.
Why:
1. Made (or Designed) in the Netherlands.
2. On opening up the device, I saw no costs spared on Electrical Safety or components. (Images will follow)
3.Fitted nice in my technical room.

For the bathroom I went with the Duco Vochtsensor (0000-4218 or 0000-4545). First thing I noticed:

shocked homer simpson GIF
That’s expensive……………

95 euros for a sensor…. holy cow… (prices range from 90 euro – 170 euros….)

Having quite some experience with the SHT-21 Sensors from Senserion: https://www.sensirion.com/en/environmental-sensors/humidity-sensors/humidity-temperature-sensor-sht2x-digital-i2c-accurate/
I was hoping to find the same sensor (Or in the same family) within this ” Duco Vochtsensor” aaaand yes! it is.

After a couple of days of testing I noticed that it took quite a while for the Duco ventilator to turn on when I was taking my shower.
So I decided to do some measurements and improve the design a bit:

In the Duco design:

Duco airflow simulated in paint.

Somehow there are no holes on the bottom side of the sensor. Only on the two sides, where one side is heavily restricted by a cable. (so there is no direct airflow over the sensor, I think) Strange choice I cant get my head around this design.. (anyone from Duco wants to comment?)
maybe it’s a price issue so they can avoid the Senserion Dust cover? I truly have no clue.

In my design:

New 3D model (3D printable) of the Duco Sensor Humidity Sensor

Air flow: ( Professionally simulated in Paint)

Airflow simulation with paint

How did I measure the differences between the two designs:

On the left, my sensor, on the right, the duco vochtsensor.

First I wrote some software for an STM32L412KB Nucleo board I had laying around. I used two I2C channels to measure both sensors at approximately the same time.

Then calculate Humidity and Temperature.

After that I used Live Expressions and STM32CubeMonitor to Log the data and place them in a graph. (tutorial will follow)

Complete setup (and yes I closed the lid when I was testing)

Results:

When I turned on the shower (nice , hot and steamy) I saw the following graph:

Reaction graph

HUMIDITY: Duco Vochtsensor
HUMIDITY_2: My Sensor
TEMPERATURE_C: Duco Temperature
TEMPERATURE_C_2: My Sensor

At 32 Seconds I started the shower. As you can see my sensor is picking up higher moisture levels in the bathroom way quicker.

When both sensors reach approximately the same humidity I turn off the shower and open a window:

Turn off moment.

The dip you see is when I open a window.

These sensors have a 3% margin of error. So all results are with the margin of error.
The temperature Sensors have an accuracy of 0.3 percent. So to prove that my sensor is reacting quicker let’s have a look at those results:

Duco ventilation box vochtsensor results.

As you can read in the graph above there is an Delta T of almost 2 degrees C between the sensors. (maximum error is at that temperature is approximately: 0.067 Degrees Celcius)

After a while both sensors are display roughly the same temperatures again (within margin of error)

Same temperatures (within margin of error)

As you can see in the above graph, My sensor design reacts faster to temperature and humidity changes.

And Now:
I will open-source my design once my V2.0 PCBA’s are in. The following will be opensourced:

1: All hardware
2: 3D prints

So anyone can make their own improved Duco vochtsensor.

Leave a comment

Your email address will not be published. Required fields are marked *