-
|
I’m excited about this project! I’ll begin by outlining my plans and providing some context for the project. Project Details:
Questions:
|
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
|
Hey @Nej4218 That looks like an interesting project. It's not trivial at this scale. If you allow for some unsolicited advice: I would definitely start way smaller, specially since you mentioned not having a lot of experience. I would also simplify things a bit. For example, having one sensor per row or groups of similar plants. If it's a research project, make sure to read What to expect from the sensor -- it's going to be challenging to precisely compare values of two different boards anyway unless you at least calibrate each of them individually. Battery management alone at that scale is tricky, and so is handling with inevitable sensor failures and software updates. Tiny battery + wireless is great for spaced out sensors around the house. For 100+ sensors close together in a custom built setup, I would start to consider going wired + shared power.
I have never used anywhere near that many sensors in such close proximity, so I can't say how reliable they would be. I would guess not very, but some of those transmit infrequently enough that there may be very few active at any given time. There are also many parameters to tweak in each of those scenarios, such as transmission interval, power
The fabrication files are available in the
See #131 and #79 (comment).
Yes, the hardware is released under https://creativecommons.org/licenses/by-sa/4.0/. |
Beta Was this translation helpful? Give feedback.
-
|
I appreciate the swift reply!
I was planning on acquiring a few premade b-parasites from someone making a bulk order to keep the prices down, asking if they are willing to also preflash while I'm learning (which seems like there is) first to test how it works and play around with integration into HA. If everything works well I'll try to get 22 custom b-parasites for the zones of 12 plants each to actually do it myself
I am totally fine with the precision of the sensors, it's simply a project I'm doing because I love plants, data and I want to get into programming/home assistant. The sensors purpose will mostly be to adjust approximate watering schedule that fits the majority of the plants as the roots develop, as well as keeping an eye on my slightly more expensive aroids 😅
I will want try to calibrate all of them, but we'll see haha
My main issue is basically space, i use a plastic planter with drip system into local water reservoirs with wicks, the plants cover an entire wall from side to side with water reservoir just bellow. So batteries would be ideal (imo) I'd most likely try to adjust it to send data every 30 min or 1 hour to prolong the battery life as much as possible - I'm aiming for them to last around a year if possible, If that works out I dont mind changing the batteries yearly, i can get them in bulk / 200 batteries for ~ 57usd. In terms of sensor failure I plan on having a few extra on hand for sure, and software updates are probably unavoidable for any HA integration device i think?*, if i were to go with the alternative Xiaomi flora sensors (which imo suck ∆$$ compared to b-parasite) i would have to deal with similar issues and probably more expensive/less fun
Great idea about intervals! I'll have to look into that, and I read a bit further and I just have to have adequate amount of Zigbee router/extenders (if i decide to go that route) In the end the sensors don't play the most crucial part of the plant wall, but a "nice to have" rather than a "need to have". The water reservoir on the other hand is where I need pretty spot on readings, but for that I'm going with atlas scientific sensors/probes. I'll Def keep y'all updated with pics on the community gallery |
Beta Was this translation helpful? Give feedback.
Hey @Nej4218
That looks like an interesting project. It's not trivial at this scale.
If you allow for some unsolicited advice: I would definitely start way smaller, specially since you mentioned not having a lot of experience. I would also simplify things a bit. For example, having one sensor per row or groups of similar plants. If it's a research project, make sure to read What to expect from the sensor -- it's going to be challenging to precisely compare values of two different boards anyway unless you at least calibrate each of them individually. Battery management alone at that scale is tricky, and so is handling with inevitable sensor failures and software updates. Tiny battery + wir…