BCSS - Forum

ralphrmartin wrote: ↑Thu Feb 23, 2023 7:39 pm I use a home made wifi thermometer, based on an ESP32 board plus a BME280 sensor, programmed to send readings at regular intervals to a server, where they are stored. That way I can see the latest update, or draw a graph over time. You need a WiFi connection in the greenhouse to do that...

I used an arduino to make a temperature logger for the conservatory which recorded the temperature every ten minutes and logged it to an SD card, from which I could then take the data and graph it in a spreadsheet to see the highs and lows. I was already considering the Mark II which would send the data over WiFi to the cloud thus removing the need for the SD card step...and then in the Mark III I could have something which automatically produced a plot to the temperatures from the cloud stored data which I could view in a browser..

Then I realised that frankly I was just recreating and overcomplicating the functionality we already got from a £20 max/min thermometer....

Yes Andy,
But it was FUN!! and you enjoyed doing the job.... Perhaps I get around to doing something like that, if even I get the time.

I agree with Mel. Fun ... challenge .....
And I agree .... it's something I'd love to do too.
But did you fully exploit and explore the data you gathered?

Absolutely: when I get a power cut, I can figure out how long it has been going on - no readings!

I took a hybrid approach and used a couple of arduinos to modernise the system a bit. I already had an analogue set up hardwired back to the house that has a display unit, power fail alarm and heater activity indicator. I used one arduino to produce a signal conditioner and to convert the signals to engineering units and to out put a data stream, with CRC checksum over a RS422 connection. Also threw in a bluetooth module to allow calibration setup and configuration from a computer. A separate data logger then takes in the RS422 data stream and writes this to an SD card. File management and data upload is done over a bluetooth connection and finds its way into a spreadsheet. The end goal was to place the signal conditioner in the glasshouse and just have the RS422 cable running back to the house. however , temporary lash-ups have a habit of becoming permanent and that's how it has remained for the last 4 years. With what's available at the current time it should be easy to combine the two units using an up to date arduino that has more memory free digital lines to be able to run a string of Dallas one wire temp sensors.

By studying the data I can understand how the heater is performing. Humidity is a difficult parameter to have much control over and it's always a compromise between opening up some ventilation or keeping it closed to preserve temperatures, the latter seems to win at the moment due to energy costs.

I can also process the data from the heater duty cycles to work out how much it's costing me (too scared to look at any more this year). Light levels remain low for this time of year, we seem to have more overcast days (first half of January) where the GH is not capturing much additional heat.

With a bit of number crunching using spreadsheet macros I can plot out the daylight integral values. Again a compromise between shading and chilles and tomatoes having to sufficient light and able to coexist with Agaves over summer and not having the place turn into a furnace.

In the summer I can see when the fan has been running and how effective it has been in holding back the temperature, though when its pulling in radiated heat from the paving slabs outside, the best its going to do is keep the inside temperature not much worse that the outside.

Nice!

Hi,
Well some of those make my set-up seem a bit simple. I do like the idea of doing it myself with an Arduino or two, but it's having the time!!

Can you work this one out! or am I just missing something?
This unit has two sensors, put them side by side and there's about 0.2-3 degs out, and when their both in the Sunlight it's fine, but has the Sun moves around there is a time one gets more Sun than the other that's OK as expected.

But later when it's dark the bottom sensor (about 30cm from the ground) is reading higher then the sensor on the higher shelf (140cm from the ground), does not warmer air rise? or is the top sensor too near the cold roof, no insulation at the moment! Please see my other post on that subject...

Mel.

Smel,

If you are getting direct sun then it will not only elevate the reading but may be different between the two devices based on the reflectivity of the sensor and the surroundings. both my inside and outside sensors have a solar shield. Nothing fancy, just a 1 litre bottle sliced in half lengthways and covered with aluminium foil.

To check if the two units are reading the same they need to be in thermal contact, either cable tie the two sensors together in the shade or place them in a container of water and stir it for about a minute before taking a reading(external sensor only - assuming they are waterproof). A difference of 0.2-0.3 seems reasonable for an 'off the shelf consumer device' ie each sensor one could have a ±0.15 error.

The other thing to check is the temperature of the display unit, it may have a temperature coefficient if its own if one catches the sun more than the other. So similar for the display unit if they have an internal sensor, in a sealed box in the dark and in a stable temperature environment.

To give a comparison, the sensors that I'm using at work cost about £1200 each + ~£600 for the calibration and then there's ~£2000 for the meter itself. As soon as you start pushing accuracy the cost goes up exponentially. Another example , a IEC751 class B PRT temperature probe which is an industry standard wouldn't do much better than what you are seeing now.

Comparing the readings between two similar devices on a regular basis is good practice anyway. If the reading suddenly differ than what is normal then that's a good indicator that one units has calibration drift. You then need a 3rd device as a referee.

All the best
Dave

Rather than start a new thread I thought I'd update this one!

In a search for a simple thermometer/ hydrometer for my greenhouse and conservatory that won't need WiFi running down the garden; I have found this Switchbot set up (https://www.amazon.co.uk/gp/aw/d/B0BVZD ... asin_title)

It has a hub that connects to three waterproof sensors (many similar products had sensors that aren't waterproof and when I pushed one of the company's help desks said they were recommended to use in greater than 80% humidity!) and that connects to the WiFi so I can measure and log the temperature and humidity in greenhouse, conservatory and outside over time, have notifications if I wish (set at different levels for each sensor) and see the data via an app on my phone wherever I am.

Now I just have see how reliable they are and how long the batteries last!

All good so far...

Where does the hub live?