moisture sensor

Moderators: grovkillen, Stuntteam, TD-er

Message
Author
User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#51 Post by dynamicdave » 17 Nov 2018, 12:23

Hi,
Can you share the schematic/wiring diagram of how you have connected/wired your moisture sensors?

I've drawn a very quick diagram of what I think is on the YL-38 board.
I'm only using the analog part of the board - which I've coloured in 'red'.
This section is really just a potential-divider formed by the 10K resistor and the moisture sensor.
Although I could have made exactly the same circuit using a discrete resistor, the YL-38 board came with the sensor and is an easy way to connect the sensor to a breadboard.

The second sketch shows how I think the analog part operates with a single moisture sensor.

I think you are correct in saying "why use the transistor?".
At the time I made up this node, I was concerned in case a student accidentally shorted-out the the two input pins on the YL-38.
Now I've studied the cicruit diagram I can see there is a 10K resistor in the path, so my fears are reduced.
So I suppose you could "power" the YL-38 from an output on the WeMos D1 Mini.
Including the current drain for the two LEDs I suspect the total current consumption is about 8mA.

Note: I'm still ultra-cautious and for the sake of the cost of a cheap NPN transistor I would use that, rather that risk destroying a WeMos.

Note: I'm not using the digital part of the YL-38. This part contains an LM393 Op Amp that is used as a "comparator".
It compares the voltage input from the sensor against a threshold set by the 10K 'pot'.
The OpAmp outputs either a logic 1 or 0 (not much use in this situation to determine mositure content).

Please let me know what your circuit looks like.
Attachments
yl_38_analog.jpg
yl_38_analog.jpg (34.26 KiB) Viewed 24354 times
yl_38_complete.jpg
yl_38_complete.jpg (47.83 KiB) Viewed 24354 times

alabama
Normal user
Posts: 90
Joined: 24 Nov 2017, 10:04

Re: moisture sensor

#52 Post by alabama » 18 Nov 2018, 19:17

Here is a really very primitive sketch of my wiring, without the sensor-fingers and without the WeMos. But I think you can imagine that parts. Please excuse the poor quality of the drawing.

I think it's clear after have a look on your sketch of the internal analog part of the sensors. Even iv they are not activated the analog signal is devided from the active sensor to the other sensors and to teh A0-input of the WeMos. So only a very small part (someone other perhaps can calculate it - I'm not) comes really to A0.
I tried some diodes betwen the sensors and A0, but I fear this ones I have here (1N4001) are not the right ones.

Now I run the prototype with the ADS1115 and it runes fine. The analog lines from the four sensors come to the A0 ... A3 of the ADS1115 and this is connected by I2C with the WeMos. The calibration is a little different, because the WeMos-ADC uses 12 bit (?), the ADS1115 16 bit (?), but this is very clear written in the wiki of letscontrolit. I use the 2-point-calibration written there to set the output of the signals to values between 0 and 10.
https://www.letscontrolit.com/wiki/index.php/ADS1115
Attachments
IMG_1544.JPG
IMG_1544.JPG (1.62 MiB) Viewed 24324 times

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#53 Post by dynamicdave » 18 Nov 2018, 21:04

Hi,
Thanks for posting your original wiring diagram.

It is as I suspected - you have commoned all the outputs together and fed them to the A0 pin on the WeMos.

If you think about the circuit all four moisture sensors are in parallel (i.e. a parallel resistor circuit).

Although you only switch-on one moisture sensor at a time (so only one of the 10K resistors is joined to +3.3V) all the bottom parts of the circuit are paralleled together.
If you know Ohms Law you could work out the resultant resistance value - it's going to be low, which will lower the combined voltage.
I can send you a sketch if you need one to show the effect.

Anyway, I'm glad you have got the circuit working with the ADS1115.
The ADS1115 is the correct component to use as it's a four channel analog switch followed by an ADC (analog-to-digital converter).

Kind regards from David.

alabama
Normal user
Posts: 90
Joined: 24 Nov 2017, 10:04

Re: moisture sensor

#54 Post by alabama » 18 Nov 2018, 23:39

Thank you for the offer, that is very nice. But I think it's not necessary for you to work that out especially for me. I already understood the principle. I learned some electronics at school about 35 years ago, and I'm always happy when I remember something after such a long time. That's why this project is so much fun for me. Sometimes it's just that you want to try an idea, although the better / right solution is different.

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#55 Post by dynamicdave » 19 Nov 2018, 07:59

Hi,
Glad to have been some help to you.

Really pleased you have got the ADS1115 and moisture sensors working with the WeMos.

PS:
I've added the Node-RED node called "telegram" to my system so when a plant is getting dry it sends a message to my mobile-phone "I'm getting dry, please water me!!"

Kind regards from David.

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#56 Post by dynamicdave » 19 Nov 2018, 09:51

Hi,
I decided to sketch-out the circuit as I might explain it to my IoT students (sometime).
Basically the mositure sensor is acting as a voltage divider.
If you look at Fig-1 it shows a simple voltage divider.
The output voltage is proportional to the value of resistors R1 and R2.
If R1 and R2 have the same resistance value, then the output is 50% of the supply voltage (i.e. 2:1 division ratio).

Fig-2 is what I think your original circuit looked like.
Assuming that just one transistor is switched on, then one of the 10K resistors is connected to +3.3V.
To keep the maths simple/easy, I've assumed the resistance value of all four moisture sensors is 10K.
You can try changing these values later if you want to see the effect it has on the divison ratio.

Fig-3 shows the circuit in a clearer way (I've just tidied up the circuit).
As there are four (10K) resistors in parallel the resultant resistance is 2.5K ohms.
(E.g two 10K resistors in parallel gives 5K, three would give 3.3K and four gives 2.5K)

The final circuit on Fig-3 shows the equivalent circuit as a simple voltage divider.
So applying our original formula (R2/(R1 + R2) ) gives ( 2.5K/ (2.5K + 10K) ).
The final division ratio is therefore 0.2 or 20%.
This means the voltage that goes to A0 (on the WeMos) is much lower than you would expect and is not directly related to one sensor (it is a combination of all four sensors).

Hope this makes sense.

Cheers from David.
Attachments
Fig_3.jpg
Fig_3.jpg (20.24 KiB) Viewed 24305 times
Fig_2.jpg
Fig_2.jpg (24.29 KiB) Viewed 24305 times
Fig_1.jpg
Fig_1.jpg (30.53 KiB) Viewed 24305 times
Last edited by dynamicdave on 19 Nov 2018, 09:55, edited 2 times in total.

alabama
Normal user
Posts: 90
Joined: 24 Nov 2017, 10:04

Re: moisture sensor

#57 Post by alabama » 19 Nov 2018, 09:54

Do you know the FHEM? I use this since longer time. My QNAP has a IoT system based on Node-RED, I tried to use it, but until now it don't work like I want. Some years ago I tried to concentrate as much as possible to one central system. But now I changing my opinion to separate as much as possible to the end nodes. So my watering system of my garden runs per rules independent on a NodeMCU-V3 an is olny started from the central-FHEM. So I try to arrange that every node gives me a message by mail using the notification. Up to nw it works with my provider with the standard mail (no encripting). The watreing system gives me information when each line starts watering. Now the WeMos shall send me a mail when one plant needs water, this will be one of the next steps. But this is only the test run for the wintertime, next year the watering of the plants at the windos shall start autonom. This will be a bigger project ....

Best regards, many thanks for your time,
Harald


EDIT: You already answered in the meantime - I'll send this message anyway.
Your calculation makes absolutely sense, it shows very good what my measuring gives me. Of cause it is not exact because not all senors have the same resistance (different state of every plant), but in general it's very similar: About 20% of the real values. I think this can be a nice example for your students to show the "wrong" way.
Last edited by alabama on 19 Nov 2018, 10:04, edited 1 time in total.

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#58 Post by dynamicdave » 19 Nov 2018, 09:57

What does the abbreviation FHEM Mean ??

And snap... I have a QNAP NAS as well.
The only problem is my QNAP is a few years old and does not support Node-RED.
No problem as I run Node-RED on a couple of Raspberry Pi 3+ boards.

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#59 Post by dynamicdave » 19 Nov 2018, 10:03

If you need any ideas for additional projects (ha, ha, ha) then take a look at some I've published on Discourse - Node-RED.

https://discourse.nodered.org/c/share-your-projects

Cheers from David

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#60 Post by dynamicdave » 19 Nov 2018, 10:55

Don't worry answering this - I've found out about FHEM.

alabama
Normal user
Posts: 90
Joined: 24 Nov 2017, 10:04

Re: moisture sensor

#61 Post by alabama » 19 Nov 2018, 11:27

Okay, thanks. Meanwhile my collegue disturbes me longer time and my writing was removed before I can send. But my conclusion: The best system is that one you can serve. For me it is FHEM this time, because its the center of all my subsystems from TV and telephone to watering the garden.

Cheers,
Harald

alabama
Normal user
Posts: 90
Joined: 24 Nov 2017, 10:04

Re: moisture sensor

#62 Post by alabama » 21 Nov 2018, 11:52

Yesterday I removed my transistors and run the YL-38 direct with the GPIO-pins. It works fine so far.

I see one - not really important but for me interesting - fact: First I wonder why my analog input now shows 10 if there is no activity of the YL-38. With use of the transistor in the ground line it shows 0 in this stage. Of cause, with your simple sketch above of the main functionality of the analog part it is very clear: If I switch of on the ground line the A0 is connected to the 3.3 V through the 10R and shows 3.3 V -> 0 in the display. If I switch of the 3.3 V-line (like you do and I also do without transistor) the A0 ic connected through the sensor fingers with grnd -> 0 V -> 10 in the display.

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#63 Post by dynamicdave » 21 Nov 2018, 14:24

Hi,
If you have removed the transistor and now make use of a GPIO output pin, then you have changed the electrical conditions of the circuit.

When a transistor is ON, the collector to emitter resistance is very low (so +3.3v appears on the emitter).
When a transistor is OFF, the collector to emitter resistance is very, very high, so it appears as if the collector to emitter is open circuit i.e. high impedance.
This means the top log of the 10K resitor is effectively floating or open circuit.

Using a GPIO pin on a WeMos is a lot different.
Most GPIO output pins have a "totem pole" output (formed by two transistors in series: one is On and one if Off).
Google it for more information.
This gives a low impedeance path to +3.3V when the output is logic 1, and a low impedance path to 0V (GND) when the output is logic 0.
The bottom line is when the output is logic 0, then the top leg of the 10K resistor is effectively connected to 0V.
i.e. you have another 10K resistor in parallel with the moisture sensor.

Trust this makes sense.

Kind regards from David.

alabama
Normal user
Posts: 90
Joined: 24 Nov 2017, 10:04

Re: moisture sensor

#64 Post by alabama » 22 Nov 2018, 09:06

Thank you very much. I think I have understand the principle - and learn a new word :"topem pole output". The Internet shows me a lot of nice pictures about that with two, three or even more transistors and other parts.

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#65 Post by Lale » 15 May 2020, 20:55

dynamicdave wrote: 30 Aug 2018, 09:15 In case anyone is interested I've made use of ESP Easy rules, timers and a generic dummy device (i.e. variables) to control one of the inexpensive resistive moisture sensors.

I'm using a Wemos D1 Mini which is triggered by sending an event (via MQTT) called "activate" from my Raspberry Pi (running Node-RED).
This action turns ON a BC547 transistor (its base is connected to D6 via a 1K resistor) to supply 3V3 to the moisture sensor.
It also kicks-off a 6-second timer.
The analog input (from the moisture sensor) is set to sample at 5-second intervals.
The samples will only be "live" if the transistor is turned on, and will only be sent to my Raspberry Pi (running Node-RED) if the timer is active.
To ensure the system only sends a single sample I set a variable called "sent" once the first sample has been published.
This action effectively shuts the door after the first sample.

The Wemos is triggered once every 15-minutes for 6-seconds, so hopefully corrosion should not occur on the metal fingers of the moisture sensor.

Hope someone might find this useful.
Cheers from David.

on System#boot do
timerSet,1,0
TaskValueSet,12,1,0 //Set var#activation = 0
TaskValueSet,12,2,0 //Set var#sent = 0
endon

on activate do
gpio,12,1 //Set GPIO(D6)=1 (Turn on the transistor)
publish node26/status, "activated"
TaskValueSet,12,1,1 //Set var#activation = 1
TaskValueSet,12,2,0 //Set var#sent = 0
timerSet,1,6
endon

on paw#value do
if [var#activation]>0 and [var#sent]=0
publish node26/reading, {"paw": [paw#value]}
TaskValueSet,12,2,1 //Set var#sent = 1
endif
endon

on Rules#timer=1 do
gpio,12,0 //Set GPIO(D6)=0 (Turn off the transistor)
publish node26/status, "deactivated"
TaskValueSet,12,1,0 //Set var#activation = 0
endon
Hi all, first of all thanks to @Dynamicdave for this project. Trying to develop my never complete project i searched for an example to better understand Espeasy rules. This one seems a good systematic example, enough complex to get confidence with the language and enough simple to let me understand it.
I tried to write my version as i don't have a moisture sensor, but many tests has gone bad. I usually read everything again and again before giving up and ask for a help, actually after some debug i've found a lot of my mistakes.
Now i'm stuck at the condition

Code: Select all

if [var#activation]>0 and [var#sent]=0
it's like the first part is working fine: the rule notices activation=1 , the cycle starts and i get the reading, but the second part doesn't work because i get 6 readings until the counter expires, even if sent=1. To verify this value i've put a

Code: Select all

publish /met1/dht_1/reading {"sent":[var#sent]}
only for debug: except the first cycle, sent is = 1
Any idea?
Thanks

User avatar
grovkillen
Core team member
Posts: 3621
Joined: 19 Jan 2017, 12:56
Location: Hudiksvall, Sweden
Contact:

Re: moisture sensor

#66 Post by grovkillen » 15 May 2020, 21:43

Never use VAR as task name. Change that and it should work.
ESP Easy Flasher [flash tool and wifi setup at flash time]
ESP Easy Webdumper [easy screendumping of your units]
ESP Easy Netscan [find units]
Official shop: https://firstbyte.shop/
Sponsor ESP Easy, we need you :idea: :idea: :idea:

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#67 Post by Lale » 16 May 2020, 00:59

grovkillen wrote: 15 May 2020, 21:43 Never use VAR as task name. Change that and it should work.
Hi Grovkillen, i tried to replace the string "var" with "va2" but the result is the same. Thank you for your answer however.
I took a look at the wiki, the "IF / ELSE" section says it's not possible to use AND/OR for older versions. My version is 2_3_0 build 120, is it right?
I tried to avoid "AND" nesting the statements

Code: Select all

on dht_1#temperature do
 if [va2#activation]=1
  if [va2#sent]=0
  publish /met1/dht_1/reading {"Temperature":[dht_1#temperature]} // debug?
  publish /met1/dht_1/reading {"activ":[va2#activation]} // 4 debug
  publish /met1/dht_1/reading {"sent":[va2#sent]} // 4 debug
  TaskValueSet,2,2,1 //Set va2#sent = 1
  endif
 endif
endon
It worked fine once just modified, not after a reboot though: i get an unwanted read at system boot.

User avatar
grovkillen
Core team member
Posts: 3621
Joined: 19 Jan 2017, 12:56
Location: Hudiksvall, Sweden
Contact:

Re: moisture sensor

#68 Post by grovkillen » 16 May 2020, 05:46

Aha, well then please try the latest version found on GitHub.
ESP Easy Flasher [flash tool and wifi setup at flash time]
ESP Easy Webdumper [easy screendumping of your units]
ESP Easy Netscan [find units]
Official shop: https://firstbyte.shop/
Sponsor ESP Easy, we need you :idea: :idea: :idea:

User avatar
dynamicdave
Normal user
Posts: 257
Joined: 30 Jan 2017, 20:25
Location: Hampshire, UK

Re: moisture sensor

#69 Post by dynamicdave » 16 May 2020, 09:53

Hi,
I found an old post about the problem in using VAR as the name for a variable.
Although I've been using VARS since that time, I didn't have time to update the example/tutorial, sorry.

The other thing.. I think you should have single quotes around your json string in your publish command

Code: Select all

    publish /met1/dht_1/reading '{"sent":[var#sent]}'
I would actually write this as... (with commas rather than spaces)

Code: Select all

 publish,/met1/dht_1/reading,'{"sent":[var#sent]}'
The other thing to note is... a leading slash in the topic name is not recommended anymore.
If you do delete it in the command, remember to also adjust the setting in the MQTT controller tab.


=======
#6 Post by dynamicdave » 20 Nov 2018, 11:48
Hi,
Just noticed in one of your rules the named-variable 'var'.
You might run into trouble using this becuase as from November 2018 release of EASY Easy Mega, 'var' is a reserved command in rules.
You might want to change it to 'variable' or 'vars'.

Here's what I was told...
- - - -
The problem with your rule is that [Var#n] is a reserved command that retrieves the system variable %Vn%.

Wiki should be updated with this new commands and system variables.
=======

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#70 Post by Lale » 17 May 2020, 16:38

grovkillen wrote: 16 May 2020, 05:46 Aha, well then please try the latest version found on GitHub.
May you help me to understand versions?
The version installed on my board is the same stable ESPEasy_R120_4096.bin i can find in http://www.letscontrolit.com/downloads/ESPEasy_R120.zip
Should i go into (that i don't know very well) github -> branch -> v2.0? Or Mega? or is it the same?

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#71 Post by Lale » 17 May 2020, 16:57

dynamicdave wrote: 16 May 2020, 09:53 The other thing.. I think you should have single quotes around your json string in your publish command

Code: Select all

    publish /met1/dht_1/reading '{"sent":[var#sent]}'
I would actually write this as... (with commas rather than spaces)

Code: Select all

 publish,/met1/dht_1/reading,'{"sent":[var#sent]}'
Thank you Dave, unfortunately if i replace spaces with commas i get in nodered

Code: Select all

Unexpected token ' in JSON at position 0
each line in rules among

Code: Select all

 publish,/met1/dht_1/reading,'{"Temperature":[dht_1#temperature]}' // debug?
  publish,/met1/dht_1/reading,'{"activ":[va2#activation]}' // 4 debug
  publish,/met1/dht_1/reading,'{"sent":[va2#sent]}' // 4 debug
also adding single quotes. There should be something else wrong in my rules or in the Json node in nodered.
On the other hand if i remove single quotes, i get in nodered

Code: Select all

/met1/dht_1/reading : msg.payload : undefined
To take stock of the situation i have to fix the AND/OR issue probably downloading a newer version or finding a workaround with more varibles and
undestand where is the bug that doesn't let me use single quotes in publish lines.
I would remove slashes in mqtt rules for the last, coz i can find everything published in the mqtt log.

mrwee
Normal user
Posts: 225
Joined: 31 Aug 2016, 12:52

Re: moisture sensor

#72 Post by mrwee » 17 May 2020, 17:02

You need to:

1) go here: https://github.com/letscontrolit/ESPEasy/releases
2) Select "Assets" on the most recent version (Right now it's "Release mega-20200426")
3) Download the .zip file
4) Extract the .zip file
5) I'd first flash a blank .bin file, to make sure everything is clean (4MB or 16MB depending on how much flash your wemos has). I'm using the "FlashESP8266.exe".
6) Flash the "ESP_Easy_mega-20200426_normal_ESP8266_4M1M.bin (Again, it's the latest version today) to your Wemos
7) Reset Wemos
8) After boot & initialization (You can follow on serial port), reset again
9) Ready :)

This usual works for me :)

User avatar
grovkillen
Core team member
Posts: 3621
Joined: 19 Jan 2017, 12:56
Location: Hudiksvall, Sweden
Contact:

Re: moisture sensor

#73 Post by grovkillen » 17 May 2020, 17:27

Lale wrote: 17 May 2020, 16:38
grovkillen wrote: 16 May 2020, 05:46 Aha, well then please try the latest version found on GitHub.
May you help me to understand versions?
The version installed on my board is the same stable ESPEasy_R120_4096.bin i can find in http://www.letscontrolit.com/downloads/ESPEasy_R120.zip
Should i go into (that i don't know very well) github -> branch -> v2.0? Or Mega? or is it the same?
Here you find the releases: https://github.com/letscontrolit/ESPEasy/releases
ESP Easy Flasher [flash tool and wifi setup at flash time]
ESP Easy Webdumper [easy screendumping of your units]
ESP Easy Netscan [find units]
Official shop: https://firstbyte.shop/
Sponsor ESP Easy, we need you :idea: :idea: :idea:

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#74 Post by Lale » 18 May 2020, 00:13

mrwee wrote: 17 May 2020, 17:02 This usual works for me :)
Thanks both.
Now i'm fighting against the new flashed firmware...

TD-er
Core team member
Posts: 8642
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: moisture sensor

#75 Post by TD-er » 18 May 2020, 00:21

Lale wrote: 18 May 2020, 00:13
mrwee wrote: 17 May 2020, 17:02 This usual works for me :)
Thanks both.
Now i'm fighting against the new flashed firmware...
What's not working in the newer versions?

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#76 Post by Lale » 18 May 2020, 00:32

TD-er wrote: 18 May 2020, 00:21 What's not working in the newer versions?
I moved all the settings to the new settings page, i can't connect the access point, when i set the WpaKey (8 chars) in Wifi settings and submit, when i reload the page it shows five bullets in place of 8. I'm trying to understand if i misconfigured something.
I wonder if could be a test if i set those with my home router instead of the raspberry AP.

TD-er
Core team member
Posts: 8642
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: moisture sensor

#77 Post by TD-er » 18 May 2020, 00:35

In order to protect the used key, the shown password field always displays ***** if something is set.
Only if the field is empty, it will also be shown as empty.

If you have had some old firmware installed on the node, please wipe the complete flash first using one of the included blank flash files. (4M file for 4M flash etc.)
Then flash the desired firmware bin file again and please let me know if that helps.

User avatar
Lale
Normal user
Posts: 10
Joined: 07 May 2020, 20:45

Re: moisture sensor

#78 Post by Lale » 20 May 2020, 19:22

TD-er wrote: 18 May 2020, 00:35 Then flash the desired firmware bin file again and please let me know if that helps.
The flashing process it's ok now. To keep it short just to avoid off topic: the first time i flashed, i put the blank image, then the standard4M1M and skipped the wifi wizard because i was sure to have enough settings saved in my pc as screenshot, txt, etc. I don't know why after setting everything by hand i got "Last Disconnect Reason=(203) Assoc fail", anyway i kept a little log from the serial tty output (if needed for study i can provide it, maybe in another topic or in PM). I flashed blank4M again and standard4M1M, used wizard, everything ok.
So now i can get back to my in-topic rules, now with AND/OR conditions.
Thanks TD-er.

TD-er
Core team member
Posts: 8642
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: moisture sensor

#79 Post by TD-er » 22 May 2020, 12:44

Lale wrote: 20 May 2020, 19:22 [...] I don't know why after setting everything by hand i got "Last Disconnect Reason=(203) Assoc fail", [...]
It probably tried to connect to some AP, and failed as there were no credentials stored.
But I've also seen that the first disconnect reason sometimes is bogus.
Not sure where the bogus value may come from, but it seems like it may be not initialized well if a unit just reboots (not power cycle).
I would not pay much attention to it right now.

Post Reply

Who is online

Users browsing this forum: Ahrefs [Bot] and 14 guests