Mysteries behavior of Sonoff S20 - how can that be?

Moderators: grovkillen, Stuntteam, TD-er

Post Reply
Message
Author
User avatar
EDsteve
Normal user
Posts: 61
Joined: 11 May 2018, 21:13

Mysteries behavior of Sonoff S20 - how can that be?

#1 Post by EDsteve » 27 Nov 2022, 11:07

Hi,

since years i am using a few Sonoff S20 flashed with ESPeasy without issues. Two days ago one device wasn't accessible any more so I looked into it and found out that in only works when connected with a 3,3V power source on the pin headers. But when i plug it into the 220V power plug. It is not working any more. Even though i can measure 3,3v on the pin header while plugged into 220v.
Reading the serial while plugged into 220v shows this symbol in an endless loop: ⸮⸮⸮⸮⸮⸮⸮⸮⸮⸮⸮⸮⸮

After flashing a blank bin file i flashed the new version of ESPeasy. But still same behavior.

I don't think it is an ESPeasy issue. More like a hardware problem. But it just doesn't make sense to me. Anybody an idea?

Cheeers
ED

User avatar
Ath
Normal user
Posts: 3416
Joined: 10 Jun 2018, 12:06
Location: NL

Re: Mysteries behavior of Sonoff S20 - how can that be?

#2 Post by Ath » 27 Nov 2022, 13:08

Most likely the power conversion from mains to 3.3V is defective, resulting in an unstable voltage. Though the average is still 3.3V when measured, it isn't stable enough for the ESP to properly start. That's a common issue with Sonoff devices, many have reported about the poor power-supply behavior of these and other Sonoff units.
/Ton (PayPal.me)

TD-er
Core team member
Posts: 8643
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: Mysteries behavior of Sonoff S20 - how can that be?

#3 Post by TD-er » 27 Nov 2022, 17:59

Capacitors do age. And when their capacitance gets below some threshold, the supplied voltage is no longer stable.

Typically a capacitor is rated for a life span of 1000 hours at some temperature.
For every 10 degrees lower (average operating temperature) the life span doubles.

Good quality capacitors have 1000 hours at 105C and really good ones 2000 hours at 105C.
But the low cost ones will probably be rated at 1000 hours @85C.
This means that at an average temperature of 35C (which is not uncommon in a Sonoff device) the life span is 2^5 * 1000 hours = 32'000 hours. That's 3.6 years.

So my guess is that replacing the caps will restore the Sonoff to be working again.

User avatar
EDsteve
Normal user
Posts: 61
Joined: 11 May 2018, 21:13

Re: Mysteries behavior of Sonoff S20 - how can that be?

#4 Post by EDsteve » 30 Nov 2022, 12:37

Sorry for the late reply a virus tied me to bed for a few days.

Both your answers pointing in the direction of the power conversion from 220V which of course makes sense.
I wasn't surprised that capacitors don't last forever. But it's a bit shocking to hear that they might create problems already after almost 4 years. Nobody want's to hear that...!! We should remove your post @ TD-er :lol:

For sure i want to try the capacitor solution before i through that thing away and feel bad about it.
After a quick read it seems that ceramic capacitors have the lowest life expectancy. So i guess it's either one of the blue ones on the frontside or some unidentified SMD capacitors on the back side.
Which one would you suspect to be troublemaker? A wild guess?
Attachments
backside.jpg
backside.jpg (195.4 KiB) Viewed 1582 times
frontside.jpg
frontside.jpg (127.3 KiB) Viewed 1582 times

TD-er
Core team member
Posts: 8643
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: Mysteries behavior of Sonoff S20 - how can that be?

#5 Post by TD-er » 30 Nov 2022, 12:57

The cylindrical black ones surrounding the transformer thingy.

Please take note of the polarity. When used with reverse polarity their life expectancy is even worse and rather explosive too.
Can you see the mentioned temperature on the side?

The capacitance and the voltage are important when looking for replacement.
The capacitance should be at least what's written on the side and the voltage also at least what was used here.
Problem when you pick one with either higher capacitance or higher voltage is that those will be larger and probably won't fit in that tight spot.

Also with the caps so close to the transformer, I think my estimated average working temperature may be a bit low.

The blue droplet shaped one is probably a varistor. (voltage dependent resistor)
It is often used to protect against voltage spikes.
Even if that one is broken/blown, it won't affect the performance of the unit. Unless it is causing a short, but then you would probably see other indicators like magical smoke, blown fuses, etc.

User avatar
EDsteve
Normal user
Posts: 61
Joined: 11 May 2018, 21:13

Re: Mysteries behavior of Sonoff S20 - how can that be?

#6 Post by EDsteve » 30 Nov 2022, 14:48

One of these capacitors i had lying around (the single one on the right which is 330uF and 16V).
I changed it... no smoke... or fire... and it works again like nothing ever happened.

I love this forum specially Mr. TD-er :)
Thanks for taking your time i hope this thread will save many SonOFFs from the bin

EDIT: Maybe a better title would be: "Sonoff doesn't boot / start any more" so people can find it more easy?

TD-er
Core team member
Posts: 8643
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: Mysteries behavior of Sonoff S20 - how can that be?

#7 Post by TD-er » 30 Nov 2022, 14:54

That's the one on the output of the power supply.
The one that's probably the most likely one to make the voltage unstable.
However the other ones will eventually cause issues too. Perhaps they have a higher tolerance.

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 27 guests