Tuya SGS01 Plant sensor

[Grillbert]

... it´s not an air humidity sensor, but intended to measure humidity in soil - but could be pretty similar to the devices targeted by this project:

Also it´s a standard tuya dev-board included in the device.

Does anybody experiment with this device already? Would be great to be able to use it in HA without tuya apps/cloud/login ...

[pvvx]

I have such a sensor and it works with my firmware.
PLANT-MON - Tuya BLE "Plant monitor"
But I haven't finished embedding such sensors into the general firmware source code yet...

It has a Tuya MCU installed. This is an extra chip that eats up the battery and limits the measurement interval, and so on. Tuya MCU controls the BT3L module and without getting rid of this chip, nothing good can be done. This is the main problem with these sensors.
Not everyone can remove this chip and make a couple of jumpers...

As a result, embedding into the general pile is abandoned. And I will not create a special repository and support for one sensor.
When I have the desire and time, then perhaps I will finish integrating this sensor into the overall project. But only with modification - soldering jumpers...

[Grillbert]

Great!
If your photo sais it all i think i should be able to do that easily.
Let me quickly recap:

Removal of the 5 SMD components,
3 Bridges (white Wires)
3 Bridges (MCU)

Connect schematic
| 1 2 2  |  |  |  |
| | 1 |  |  3  |  3

1 Bridge U4

-> is that complete or did i miss something?

I already flashed some of the Mija Bluetooth Thermometers (these are doing a really good job).
I would expect to flash the new firmware first using the known tools and do the HW change afterwards?!?

Thanks!!

Gilbert

PS: I wonder why Tuya goes with this complicated setup... just to make hacking harder and force customers to use their tools?
And that at the cost of: more hardware cost / higher battery drain / larger housing / and finally: fewer customers - because being
OOTB open (e.g. for HA integration) would surely attract some people to buy this thing...

[pvvx]

Measurement scheme:

The input is fed with a PWM signal from TLSR8258.
After a fixed time in ms, measurements are taken on the ADC.

As a result, all other components are not needed.

[pvvx]

Tuya simply copied the 555 timer circuit from Arduino.
https://lastminuteengineers.com/capacitive-soil-moisture-sensor-arduino/
And Tuya MCU is a standard system with a program library from Tuya.
Tuya device manufacturers don't care how many times you change the battery.

[Grillbert]

I´ve done some power measurements too (using a NRF-PPK2)...
Battery would last for ~400 days using Amazon Basic AAA without any measurement
(see attached) ... if i don´t break it i´ll add a comparison later :)

[pvvx]

Your graph shows a measurement being made using a 555 timer. It is a 4 second interval and 10 mA.
The tested measurement process on the modified circuit is no more than 7..25 ms at 3..4 mA.

[pvvx]

I was mistaken. You have not signed a measurement

The work graphs of the original device are given in Plant-Mon

[Grillbert]

Hmm ... I figured out that the standard flasher i used before does not work here, but most likely the Telink OTA flasher.
This one requires a "Service" and "Characteristic" (i am no bluetooth expert) which i got using a "BLE Scanner" from my mobile.
So i was able to connect.

I only tried the OTA Firmware Version so far which stops with "Update error: some error while sending char data" .

(and i guess the OTA Firmware is the one to use here, correct?? )

[pvvx]

I have already described that this sensor has not yet been integrated into the general project.

When I have the desire and time, then perhaps I will finish integrating this sensor into the overall project.

Today I have no free several days to work with this.

[Grillbert]

OK ... I was not aware what "not integrated" means - i thought it´s at least ready to flash something and get values without the tuya stuff.
I stay patient - thanks for your work so far - if i can support with anything let me know.