Something that would be good to see is a system to measure temperature and moisture level.
Dai Jenkins suggested something like this, which is very cool:
Would it be possible to make them in-house and have it so they keep records centrally?
One for @niccokunzmann
They can be made. The sensor costs not much, below a pound. We might have one actually, so we can build a prototype. We can use the open sense map to publish the info.
I guess, LoRa or WIFI would be it. I can also install free as in freedom WIFI (not the free beer but give me your soul) in other places.
For WIFI, I would use WeMos - little WIFI Arduinos - or the Raspberry Pi Pico W that we use in other projects on Sundays.
I can imagine a purpose to reduce heating costs or alarm for poly tunnel frost. I like to work with a purpose.
What is the context?
We can also offer a workshop where people build their own and take them home.
It can be integrated with other software like apps and diagrams and the like.
The context is to track temperature vs moisture in the workshop to get best use from the boiler. Keeping fuel use as low as possible this year will be important, but we also need to make sure the machines are happy.
This would be a big help in getting the balance optimal. I am sure other similar venues will want the same sort of thing.
I will try to find out more what you mean.
Tracking does not influence the boiler. Our system probably has knobs to push to change it.
Who are possible actors in the system?
When can we see that best use is achieved?
Machines and house. What are the temperature requirements to keep the machines happy?
I am thinking that a purpose could be over the period where nobody is there to see if it is too cold or if the heater was left on accidentially. Does that match?
While this system can’t directly control the boiler, it will allow me to play with the settings to ensure best usage.
Best use looks like no damp in the rooms and all machines working reliably/not forming rust. We won’t be able to tell this directly from the readings themselves but we will be able to observe them and associate the relevant readings with them to establish a baseline below which temp/humidity should not drop/exceed.
We will have to experiment to see exactly what is optimal for our situation.
Thanks! I wondered where the S.M.A.R.T.ness came in and I am glad it is you!
Does rust have the necessary precondition of condensation? Condensation might be computable.
I am writing down some design ideas:
This should be a little box, 9V powered. The sensor is exposed but will not get dusty and is protected from breaking off. There should probably be a manual of where to hang it up. I would like to get graphs on the floor, on the ceiling, below and above a radiator and close to a machine just to see how much the data varies and if we can use it to make sense of it and where it is most expressive. It is probably worth getting one from the picture to check a reference value.
After checking the condensation wikipedia article, it might be useful to add something to measure the machine temperature or that of metal.
For microcontrollers: Raspberry Pi Pico W allows easy updating by drag&drop of a file. Another one is the WMOS. You could update it over USB and WIFI.
In any case the controller must support deep sleep or the battery empties too fast. They all do usually. WIFI costs a lot of electricity. Data upload should be done every day or when you click a button.
The protocol could be MQTT or HTTP. It is probably worth using MQTT as it is specifically designed for that IOT purpose and we should be able to plug e.g. a graph website into the system.
Rust can form with any humidity, but we rarely get the combination of humidity and heat in this country to cause rust. Condensation however provides a perfect storm of moisture concentration and air exposure.
I was thinking we could use a plug in power supply, readings and uploading doesn’t need to be constant, twice an hour and however many times a day necessary to empty full memory would be fine.