search

basnijholt / miflora

☘️🌡🌼🥀🏡 Mi Flora Plant sensor Python package
MIT License
362 stars 99 forks source link

How does the misture sensor work? #119

Closed GAZ082 closed 5 years ago

GAZ082 commented 5 years ago

Hey there. I've been trying to get info about this regards in several forums/pages and could not find any. Perhaps you guys will be able. In theory the value is in % (as commented in the code) but i did a simple test of putting the sensor in a glass of water and then drying it out with some paper towel.

The value i got under the glass of water is 70 and after well dried, 20. How is this supposed to work then? I understand that the lower value may be influenced by ambient moisture or another things, but the top value? I saturated it with moist in theory... dont know!

Any hit would be appreciated!

ChristianKuehnel commented 5 years ago

The electrical conductivity of tap water is probably not so good. If you add a bit of salt (or fertilizer) you should probably get different values. And my sensor does reports zero moisture if you just put it on the table.

However, I don't really care about the specifics. What I do is for my plants: I defined a minimum moisture level per plant (between 15% an 30%, depending on the plant) and I'm adding water whenever the moisture level is below that. That works nicely. I never really bothered to check the precision of the sensors...

Am Mi., 9. Jan. 2019 um 20:22 Uhr schrieb Gabriel Alejandro Zorrilla < notifications@github.com>:

Hey there. I've been trying to get info about this regards in several forums/pages and could not find any. Perhaps you guys will be able. In theory the value is in % (as commented in the code) but i did a simple test of putting the sensor in a glass of water and then drying it out with some paper towel.

The value i got under the glass of water is 70 and after well dried, 20. How is this supposed to work then? I understand that the lower value may be influenced by ambient moisture or another things, but the top value? I saturated it with moist in theory... dont know!

Any hit would be appreciated!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/open-homeautomation/miflora/issues/119, or mute the thread https://github.com/notifications/unsubscribe-auth/ATCWCBAwyOz84OYbwYkSX0acxkw_Wte1ks5vBkFjgaJpZM4Z4EGz .

GAZ082 commented 5 years ago

Thanks Christian! I'll do more rigorous testing, with salty water and will get a container and put some something to dry out the air inside it. Perhaps my ambient moisture is too high, dont know.

ThomDietrich commented 5 years ago

If you want to benchmark the sensor readings, pure water (or with salt) won't tell you as much. The sensor was probably never developed/tested/calibrated for this, therefore the reading is not meaningful.

I second the opinion that relative changes should be more meaningful for you/the user.

If you want to get a better feeling for the value range, how about a measurement in dry (i.e. fresh) soil and in soaked soil. After this simple experiment you'll know your extremes in-between which you can scale your warning and alerting levels.

GAZ082 commented 5 years ago

Yes, that's actually how i'm working now, normalizing in the 20-70 scale I got with "synthetic" benchmarks. What i'll do is redo the scale with desert-dry and flood-moisted soil and see how it works.

Thanks a lot!

ThomDietrich commented 5 years ago

I'm not sure about the real value of those measurements but I'd sure be interested in your results!

ChristianKuehnel commented 5 years ago

closing old issues