It seems that it uses bluetooth, as it can be paired directly with the smartphone (no gateway required).
The problem is that I am not sure it can be integrated in HA. I found this integration, but it seems that it works through a Gardena gateway (Cloud):
Yes and no, as the product info says it can be paired directly with a smartphone App (no gateway required)
Home Assistant has a Gardena bluetooth integration, but the sensor does not seem listed as supported (yet?)
Those Gardenas look to be almost $90 per sensor.
I moved to using Ecowitt WH51’s
https://www.amazon.com.au/dp/B07JM621R3?psc=1&ref=ppx_yo2ov_dt_b_product_details
They’re pretty cheap, and I bought a RTL-SDR supported dongle from AliExpress to get the readings into HA
So far they seem pretty reliable and battery life is no problem.
That’s pretty slick! I actually have a spare RTL-SDR dongle from another project. What kind of range can you get from the receiver? How do you turn the radio signals into HA data? How do you handle multiple sensors reporting on the same frequency?
I use the RTL-SDR addon for HA combined with the MQTT Autodiscovery script (just turn it off when your devices have been imported. GitHub - pbkhrv/rtl_433-hass-addons: Collection of Home Assistant add-ons that use rtl_433
It handles the multiple devices reporting.
So far I’ve not had any problem with range, the dongle is virtually in the middle of the house, and it picks up the rain sensor about 20m away (through a number of insulated internal walls and brick). There are three moisture sensors dotted around the gardens and I have no problems with those either.
Occasionally it even picks up some other 433Mhz weather stations from somewhere nearby (not sure where), but only every now and then. I suspect an upgraded antenna would mean I could just sniff that data, but not fussed just yet as I’d prefer my own station.
This is the MQTT explorer showing the 3x WH51 devices I use
FYI, a new project by ffenix113 aims to provide similar functionality to ESPHome, but for Zigbee devices, (initially based on nRF52840):
Other than following the ESPHome concept it also seems to take inspiration from PVTO (Zigbee Switch Configurable Firmware) project:
https://ptvo.info/zigbee-switch-configurable-firmware-router-199/
PS: ffenix113 has also upstreamed patches to the Zigbee2MQTT project so it can support devices that can change features dynamically:
PPS: ZHA integration might need something similar if want to use custom clusters and attributes, see:
After about 5 or 6 months of use, I can confirm that all my sensors are still working and have the same behaviour. One is placed in the ground outdoor with temps currently around 0 degree Celsius, it just received a new set of batteries while it was still at 20% battery-level. Two more sensors are on their second battery as well, while the others are still on the first batteries.
So, battery-wise the (probably 2nd gen) Zigbee soil sensors are working well. Concerning the values, they are good for smaller pots and for the ground outdoor. Being placed in bigger pots, I discovered that they are sometimes measuring a too high moisture level for a certain period of time. I think that this on one side may be related to uneven moisture levels in the bigger pots. And on the other side while the sensors are sitting flat on the ground surface, their own plastic body may prevent drying out the moisture of the ground right below them. I would wish they had longer probes to reach closer to the plants roots, and in a bigger distance to their body. To bypass this issue, I just placed 2 sensors in bigger pots and concentrated on the sensor with the lower reported moisture value.
which one are you using ? Link ?
Can you link the exact products you’ve purchased please @vobele ?
I saw your earlier post where you’d made purchases from 3 different places: Zigbee flower and plant sensors for soil moisture + light + temperature + humidity? - #239 by vobele
Maybe still a little too early in the development to even mention it as it currently is a one-man hobby project so progress is slow going, but FYI, the developer of the “Zigbee Home” project (open-source DIY firmware project aims to provide similar functionality to ESPHome) mentioned that he does have future plans to add a customizable soil moisture sensor as a “Zigbee Home” firmware example sample, so it is at least on his roadmap, and if and when available it should be possible to just use that CLI application to generate custom firmware for your own DIY Arduino-style hardware, like the “b-parasite” soil moisture sensor open-source hardware project.
If you can program code in Go programming language (or in C) then suggest check if can help out:
FYI, ffenix113 now added ADC (Analog To Digital Sensor) + support for required soil moisture sensors:
https://github.com/ffenix113/zigbee_home/discussions/5
That should as an example allow one to make a DIY Zigbee flower and plant soil moisture sensors with configurations that in theory can be more easily customized and tweaked via the YAML configuration:
Hi,
thanks for the great collection of sensors and solutions!
I had the same need a few years ago and built a solution for my balcony garden. I’m currently investigating whether there is a broader market demand for such a sensor. The features of the sensor I’m going to sell are described on this page. Any feedback is greatly appreciated! 
Making it out of real wood sounds like a bad idea, will the wood not rot and/or take on mold/fungi if kept burried in the soil in a pot with a plant that needs to be watered often? Using wood should logically also be limiting for outdoor use?
What is the RRP list price point you are aiming for and where will you be shipping it from?
Suggest that you consider also adding a temperature sensor and a sunlight sensor if that is not way to costly.
wow, that was a fast reply!
The idea is to use a special kind of wood that’s also used for outdoor children’s toys on playgrounds and can withstand many years with direct soil contact. Thanks for your input – I’ll try to make this point clearer.
I can’t tell a price yet, the target is a high quality sensor with a reasonable price. We’ll be shipping from europe.
Good point about the additional sensors, at the moment the sensor features are reduced to the bare minimum. But I’d also like to have those two types of measurements and I’ll try my best to get it in.
I am sceptical as know that even pressure treated impregnated wood and root risistant wood do sooner or later go bad if make constant contact with wet soil or sand. Done some carpentry myself and built wood structure outdoors (including my kids Jungle Gym) and I do not believe that wood at any modern playgrounds make direct contact with soil or sand all the time. As far as I am aware they design them thse dahs so that there always only inorganic materials like steal or plastic this making contact with sand, soil or even concrete, acting as a barrier layer of protection from moisture/wetness. Yeah that did not build like that in the past but building science and experince has gotten better since then.
Even if the wood does not rot, water will transmit through. All wood is permeable to some degree, so it will allow moisture into the electronics. Perhaps there is an additional waterproof layer or inner container to protect the PCB?
Thanks again for your input!
We will thoroughly examine the robustness and suitability of the wood. After all, happy customers are our top priority. Yes, the PCB will be protected against water.
Btw.: Sorry for hijacking your thread a bit, as it is not a final product yet. Right now you still have the chance to shape the final product to your liking with your contributions until it becomes available. If you prefer, we also have a contact form on our site. We love to hear your thoughts.
Recommend that you also start a new separate dedicated thread under the hardware segment for community forum discussion. There you can ask for feedback and suggestions.
PS: Another suggestion could be to start a crowdfounding campaign on a site like Crowd Supply or similar.
