@cjborchert Thank you for your quick reply. I’lll be reading the threads as I have an acute problem.
The sensors I have now are on solar panels and it looks like the dally light in the winter. So they are dead now. Secondly netrohome doesn’t have an integration to HA . And the API doesn’t give all the info needed.
Hence the need for new sensors… ideally zigbee based. Hence my interest in this channel.
Mine is exhibiting the same issue with rapid fluctuations in the moisture reporting. It only seems to happen during the day though, so maybe the heat is affecting it (not sure where you are, but I’m in Australia).
I suspect the rapid reporting is also what is affecting the battery life. I believe these would only report back once a change exceeding a certain threshold is met (and probably ever X minutes/hours otherwise), so if there are lots of rapid fluctuations that need to be reported it will impact on battery life
Mine is currently connected via ZHA and the SkyConnect usb dongle but I have ordered a Tuya gateway, so hoping might be a firmware update that might address it.
I wonder how accurate the humidity would be if you just buried one of those Aqara temp/humidity sensors in the plants dirt? Would be interesting to see what it reports.
Prvably it will not last long as moisture go inside through the opening damaging it soon.
I’m in the UK and pretty cold at the moment. From what I can tell the humidity is greatly affected by the room temperature. I’ve been messing with my underfloor heating for the past few days to keep it at a constant temperature for most of the day, and since I’ve done this, the humidity has settled on 70% after lots of fluctuation. Seems unlikely to be a coincidence in my opinion. Maybe this is to be expected?
Battery also seems to be depleting a bit slower but still down to 50%. Can’t see me getting the advertised 1 year.
Capacitive probes measure the dielectric of the soil+water. Dielectric constant of the materials change with temperature. I cannot tell you how much in numbers without calculating it.
Low costs sensors work at low frequencies where salt composition play an important role in the dielectric constant so fluctuations can be bigger.
In my opinion it is expected.
Many of these have thermal probes. Wouldn’t a simple model that accounts for the dielectric changes keep it pretty stable?
It is not that easy to create models for these kind of probes. As these sensors work below 50MHz the dielectric constant of the water also depends on salinity, so you need to know your salinity every moment to try to create a formula just for the water. Then, each soil in each house is going to have a different dielectric constant with temperature relation as you have different soil compositions.
That is why I said that most of these products are toys or just fancy gadgets if you want accuracy. They can work for rough estimations.
Profesional probes normally use time domain reflectometry, not low single frequency capacitive measurement.
If you want to read more you can take a look https://www.ictinternational.com/casestudies/which-soil-moisture-sensor-should-i-choose-2/
Hi Stan,
I am looking to set up a plant monitor for the European consumer market. Your project looks promising and I would be very keen to share thoughts. I just registered on your website, you should find my contact details there.
Would you please share your myhomeiot modifications? I concur with the idea that there’s some problem with the disconnect, as it is supposed to be making the connections to read the values serially and only use one connection, but adding just one more miFlora (for a total of six) to a working esphome configuration tipped the whole thing over.
@fluxsmith please start a seperate new thread or find a other thread to discuss MiFlora Bluetooth configuration/optimization with ESPHome Bluetooth proxy as that is off-topic in this specific thread.
Thanks for this document!
Is the most common (non-professional) issue not over / under watering rather than optimization of crop production?
As a consumer / home user (it is home assistant we are talking about) i am ok with “toy” as none of my plants are for production, they just should not dry out or rot.
Over-watering / too much fertilizer is my biggest challenge inside my (very dry house) and very wet outdoors.
Quality for me is completely different from the quality required by my neighbors and their greenhouse production operations.
Sure, i was just saying it because many people in this post are expecting accuracy and they do not accept 20-30% moisture error in the measurement that most of these capacitive sensors show. Everything will depend on your needs and expectations
I’m also testing the new Tuya-Zigbee-Soil-Sensors in my home/garden the last two weeks and compared the measurment with 7 MiFlora-Sensors directly next to each other (which I’m using for one year pretty successful already).
I’m pretty dissatisfied with the Tuya-Sensors and would definitely return them, if I wouldn’t have ordered them from AliExpress. Two problems with them:
First: sensors change constantly (see picture)
Above that: their measurement is very different from each other even though thay are directly next to each other and also very different to what the seven MiFlora-Sensors are measuring. And the MiFloras are pretty much all in the same range (± 5%).
Picture below:
Mine is also having this major quality issues.
After reading through the thread and getting an idea of where things are at, I was going to pick up a few MiFlora-type sensors for now, with the idea of upgrading to one of the open source zigbee designs in 6-12 months once they’re a bit more mature & easily available.
Top review on the AliExpress listing for the “Timethinker HHCC Flora” I was looking at says: it is working with zigbee2mqtt and Home Assistant without problems. I’m used to mistakes in the description on that site, but it seems oddly specific to be in a review. Is it possible that the latest revision has a chip & firmware with zigbee support, or is it more likely to just be a mistake?
I was considering buying a few even expecting BLE, so I’m also happy to report back on what turns up if there’s interest.
I came to the same conclusion, the B-Parasite open source option looks great but they need a PCB Way project or similar to allow easy ordering of the pre-built parts. At the moment it seems that you have to submit the Gerber files, make the changes listed on the issue thread and hope for the best.
I don’t see any MiFlora with Zigbee support, that would be great though.
I’m going to order a BLE model and test it with an esp32 as the receiver.
It looks like yours will arrive first so I’d love to hear your feedback.
I am begging you! There is simply no off the shelf zigbee soil sensor available right now. Tried the tuya one but it is complete garbage.
Cool, well the ETA from China is showing 30 days right now, so we’ll see what turns up! After some additional digging for information I’m 95% expecting BLE only, so anything more would be a pleasant surprise.
My guess is either it’s a fake review to optimise the SEO for people like us, or a real review from someone using a combined Zigbee/BLE bridge and not realising which radio these specific devices are connected to. Price was good either way though, so I’m not worried if that does turn out to be the case.
Hi Raul, the product is sold on Aliexpress by different sellers. I bought it from yieryi Offical Store.
The ideea is not how often values are reported but how accurate are them. In my case moisture changes from 90 to 76/77 and than back to 90 in seconds. Many times per minute. This leads to a huge battery drain - 30% in just 3 days. Tested outside in the garden and inside in a pot with flower. Also in my case the reported moisture value is double than of the HHCC Mi Flora.
So if values is constant there is no need for reporting too often. You could report for example hourly if values are unchanged, but if for example there is a decrease/increase of 1%, 2% or 5% it should be reported immediately.
Hope it helps. If you need some more testing details please let me know as I am interested in having a reliable ZigBee product for indicating moisture on my lawn, my garden and for the flowers in the pots.
Best regards,
Silviu
