Invitation/project suggestion: An automated potted plant

Hi Folks,

Wanted to give Flaura a shot, but didn’t want to buy the parts and invest in the time if it was in it’s current state unusable. Briefly looking through the thread here I see a new repo (not a fork) with some home-assistant yml code. What’s the current status on this? Are we trying to use the original Flaura code with home-assistant in lieu of Blynk?

Hi Tadcrazio,

Welcome on board :slight_smile:
No, we are rebuilding the code from scratch using ESPHome. If you don’t know it, see the website: https://esphome.io/.

Also, the physical parts are undergoing massive improvements. The original design was very hard to assemble but now all the printed parts can be made in one print/part with much less need of post-print assembly.

We are currently testing/improving the physical parts and making the code in parallel. There is some discussion around the PCB as well, but I think the consensus is that it’ll do for now (please correct me if I am wrong, folks).

If you’d like to chip in, you can print the pot from the git repo and see if it seems to work for you. Note that you need some silicone hose of about 3 mm inner diameter to do so as well as some screws. Otherwise the water will just run out from the hose connectors.

I’ll see if I can get my Nikolaj to upload his code tomorrow.

But the code hasn’t been updated in 3 months.

No. My friend Nikolaj is working on it locally. I’ve asked him to share. I’m sure he’ll do that soonish.

Hard to contribute on that basis.

1 Like

Yeah, I know. I’m sure I can get him to upload within a day or two.

So there are only 2 STL’s? The pot and the base? Looks like you combined the parts from the original Flaura and largely merged them into one?

There is just one file that needs printing - the pot. It prints in one part now. The base isn’t needed as it is part of the original, I’ve deleted it and committed again. I Printed with support touching base only, 20% support infill and high main infill.

Okay, I can try to print it. I was thinking if parts were separate it would be easier to deal with epoxy or reprint/replace certain areas like the water pot portion, or what would hold the board in place. That way upgrading a new board or other components would be easier to print just the needed area.

Hi there! I am new in that forum and stumbled over your work for the Flaura plant pot. I have built the original version a few months ago and implemented my own Arduino Cloud IoT remote control, but it was just an intermediate solution and I am now switching to writing a Flutter app (for iOS, Android and Web) plus Firebase and a ESP firmware to create a solution for everyone (not just Home Assistant Users) and possibly in the future a Home Assistant integration.

I hope I will have a usable first App / Arduino Code in around 2-4 weeks.
When you upload your code, I hope you don’t mind if I use it as a template. I will put my Arduino Code and the Flutter app on GitHub as soon as a first rudimentary version is available.
(I hope you don’t mind that we work on the same project but different approaches…)

Very happy to see the hardware changes btw!

2 Likes

An upgraded print would just need to have the same distance between the mounting screws, or it could be glued in place if need be.

I’d rather not have to use epoxy. Not healthy. At all.

The collective experience is, that the old version, which was printed in several parts was really hard to seal well enough to be safe to place in a windowsill for weeks or months.

Hi

Welcome on board, and thanks for sharing.

I think it’s ok you use whatever you can from this project :slight_smile:

I’ve asked my friend who said he would work on the code to share what he has, and also told him not doing so was blocking others from contributing. It has been a week and he hasn’t shared anything. It’s annoying, but let’s just assume he won’t share or doesn’t have anything after all and get on with our lives.

1 Like

Can we then specify a design brief.

As I recall the firmware needs to:

  1. sense the moisture levels in the pot;

  2. If too dry, turn the pump on, either for a defined period or until the moisture level comes up to a defined level;

  3. Define the water moisture level to use (as for example a cactus requires less water than some basil);

  4. sense the water level in the reservoir and alert user if it needs filling, also turn off the ability to pump water when the reservoir is dry;

  5. have a manual button to apply water when we think it is needed;

  6. Have an on/off switch (possibly combined with the manual button

is that it?

Would be nice, if we choose to have it cotrollable with your phone, to be able to manually water the plant using your phone. Would also be nice to be able to see water- and moisturelevel on your phone. Maybe also a timer showing when the latest water was done. Maybe this is super difficult and/or require more power from the pot, idk. I’m just brainstorming ideas to make it more “user friendly”.
Is this possible?

1 Like

Battery level and Deep sleep inbetween?

1 Like

I just received the water pump and it does not have 2 rubber diaphragm inside.
The original author mentioned it was important to work correctly with water: see here

Any of you got the right version, if so, can you provide a link where you bought it?

I got mine here: https://www.aliexpress.com/item/4000974989453.html?spm=a2g0o.order_list.0.0.7b971802lbIKC4

Yes, I think you are quite close. Just a few modifications:

2: The moisture will take time to disperse from the point of entry into the soil to the moisture sensor. So I think watering for a fixed time is the only option.
3: This could be a helper value.
5: It’s in the PCb design, connected to a wakeup pin. It just needs to be defined as such.
6: It’s in the PCb design. Toggles the whole thing.

I’ve made a flow-chart of the logic as I see it:

It can be edited here (just post a new link when you are done)

2 Likes

Thank you, helpful (as was the other response).