I don’t want to offtopic in your post but off-topicing is one of the most fruitful activities to achive an objective. Brainstorming, giving different ideas and points of view avoid getting stuck in a dead hole.
Coming back to the sensors this post, in my opinion, I see some problems in most of them. They use FR4 PCBs without coating in the side. FR4 is non-isolant water absorvant material. This mean water will penetrate slowly thought the sides leading to permanent increasing offset and corroding the electrodes. How long does this take? Difficult to say but this is a problem that I have had at work.
Anothe possible problem in my opinion is related with electrochemistry. These sensors use AC currents for measuring. Currents below 20MHz polarize the salts in the water. These mean that working below that frequency you will measure humidity+salt content. I do not know the working frequency of aall sensors.
In any case if you want something with not many modifcations you can always buy b-parasite and reflash it modifying the firmware for zigbee that it is in github as its MCU supports zigbee.
I think if the goal is to really measure soil moisture (not even talking about high accuracy) it is required to do much more than just picking a 5$ chinese sensor and attach a zigbee transceiver.
Dear @Hedda it looks that you hijacked your own thread the most by this lecture. I will try to not share ideas with you, if you are niggly focused on rules. Please, accept my deep, wide and sincere apology.
Hello, today I found a sensor that claims to use Zigbee, but does not sport a model number or any details - nor does it exist on any of the Zigbee2mqtt herdsman converters after searching.
Sounds like it’s worth getting a couple to test as the probes are stainless and shouldn’t corrode. I like that it uses AA batteries and not coin batteries too.
This looks really cool. Expensive little bugger though. I’d love one half the size and half the cost, but can’t complain. This might be what we’re looking for
FYI, rbaron has now merged that Zigbee sample to b-parasite upstream repository and he hasd tested it with Home Assistant’s ZHA integration (which works without modifications) as well as with Zigbee2MQTT (which requires zigbee-herdsman converter b-parasite.js from his PR that needs to be added as per Z2M instructions to support new devices):
Are you looking in his ”hardware-v2” branch? Also maybe re-post that question in his repo’s discussion section since he probably do not read this forum thread here:
I’ve got it connected with Z2M. Just topped up the soil with some water and humidity is jumping up and down. I’ll report back later with some more readings.
Anything you want me to test let me know
They could have added a pump and reservoir to let you inject fertilizer.
On the bright side, it would be easier to find if you’re using it in your yard (I have a couple of those smaller FlowerCare ones in my yard and I sometimes forget where they are).
I’ll preface this by saying I’m only just starting out with these type of sensors and have limited knowledge on the subject. This is the first sensor I’ve owned.
My main concern is the humidity readings vary a lot from hour to hour. Considering I haven’t touched it for a week, in my opinion, it should be a fairly stable reading but seems to fluctuate between 40-70%. Have a look at the attached image. Another concern is the battery seems to be discharging quicker than advertised. I’m down to 70% after one week with fresh batteries.
I’ve got a Mi Flora on the way so will have a comparison soon.
Hard to stay about using it outside. It’s advertised as being IP67 so can withstand being in water for upto 30mins. So it will depend on how you intend to use it and if it will be in water for long periods of time.
Most of these products in this price range do not give reliable measurements and they are just toys. They use capacitive measurements in low frequencies that are affected by salinity. This means changes in the salinity give changes in the measurement for the same moisture.
I am interested in having a way to check if my netrohome actually has watered the plans or not based on the moisture levels. Based on what I’ve read here the mi flora seems to be the way forward ?
Hey @Jens_Wymeersch: There are a number of people in this thread specifically working on a much better solution than bluetooth for outdoor/indoor plant health monitoring (in order to integrate into an existing common mesh network with reasonable range and excellent usability). However, if you can’t wait for such a solution to mature a bit further, then MiFlora via ESP32 BLE hub has indeed filled the gap for some people. For questions on the subject feel free to check out one of the other ESPHome/MiFlora dedicated threads in the community. If you get stuck you can also send me a DM or tag me in another thread as I have done this myself for the interim.
