ESP32 SonOff Dual R3 Schematics.

Moderators: grovkillen, Stuntteam, TD-er

Post Reply
Message
Author
miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

ESP32 SonOff Dual R3 Schematics.

#1 Post by miazza » 10 Jan 2022, 17:00

Hello to everybody.
I connected the subject switch but I made a mistake so that S1 and S2 are no more working (I burnt some devices).
Output are working and I would like to fix the issue by changing the burnt components.

Is there any chance to have the original schematics ?

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

Re: ESP32 SonOff Dual R3 Schematics.

#2 Post by TD-er » 10 Jan 2022, 17:30

iTead does have a wiki with lots of schematics of their products.
Not sure what is the difference between the various revisions of the Sonoff Dual ones.
Found this one: https://wiki.iteadstudio.com/Sonoff_Dual
But I think that's for an ESP82xx version. Maybe you can find the R3 on their site, as probably overlooked it (if it is there).

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#3 Post by miazza » 11 Jan 2022, 08:34

Thankyou for the feedback.
I cannot find an ESP32 schematic for the Dual R3.

Just for complete the issue, I attach a pic of the C19 burnt.

I need to find the schematic in order to understand the function of C19 and its value and rated voltage.
May be I can write them ...
Attachments
DUAL R3 Burnt.jpg
DUAL R3 Burnt.jpg (3.6 MiB) Viewed 17652 times

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

Re: ESP32 SonOff Dual R3 Schematics.

#4 Post by Ath » 11 Jan 2022, 08:55

miazza wrote: 11 Jan 2022, 08:34 I need to find the schematic in order to understand the function of C19 and its value and rated voltage.
May be I can write them ...
You should probably go to their new site for support: https://sonoff.tech/support/

Maybe you already found this review: https://notenoughtech.com/home-automati ... f-dual-r3/ but still no schematic to be found
/Ton (PayPal.me)

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#5 Post by miazza » 11 Jan 2022, 09:04

Yes. I have opened a dedicated ticket few minutes ago.
I hope they are willing to share the schematic.
Thanks for the assistance.

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

Re: ESP32 SonOff Dual R3 Schematics.

#6 Post by TD-er » 11 Jan 2022, 09:12

At least here is a better view of the undamaged board: https://templates.blakadder.com/sonoff_DUALR3_v2.html
Just based on the distance of the traces, it should be low voltage, but based on how you need to wire it, it is quite easy to have mains voltage across it.
Image (found via image search, on this page: https://allegro.pl/oferta/sonoff-dual-r ... 0698140924 )

It is located next to a coil/inductor (L3), which either is a filter or part of a DC/DC converter.
Since it is close to the S-input screw terminals, I think it is part of a low-pass filter:
Image
Source

If you had wired the cable to the switch on neutral (N) instead of L, then it is clear why the capacitor was blown.
I would be surprised if only that cap was destroyed. See also: https://notenoughtech.com/home-automati ... f-dual-r3/
Since there is only 1 inductor was used while there are 2 inputs, I think it is a filter between live and GND of the board.
It looks like this is used to prevent false triggers as the wires to the actual switches may be quite long.

If it is not a LC filter, then it is part of a DC/DC converter to power the unit, but that's a really small one which will probably cause a lot of coil whine noises.
Can you see what kind of part the black cylinder is next to SW2?
That does look like an inductor too from the top and its size does make more sense to be part of the power supply of the unit.

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#7 Post by miazza » 11 Jan 2022, 09:44

WOW, you are providing a lot of very usefull information.

I try to explain the mistake I made (I did not mixed neutral with phase ... at least not directly).

This is what I did:
I wanted to use the DUAL R3 for my bedroom shutter that is right now commanded with a remote (RT3 - https://www.telecommandes-toutes-marque ... -MOOVO.pdf)

The idea is to:
- connect RT3 Remote Output PIN 1 and 3 into Dual R3 S1 and S2
- connect motor phases (UP and DOWN) to Dual R3 L Out 1 and L Out 2.

The problem is that I messed up the wires and I connected RT3 Remote Output PIN 1 and 3 to Dual R3 L Out 1 and L Out 2 and motor phases (UP and DOWN) to Dual R3 S1 and S2.

The mistake lasted about 10 seconds and during that time the motor was in some way powered and making a strange noise that suggested to disconnect immediately the power ....

After the mistake, I restored the right connection and the motor was well working when Sonoff Dual R3 is commanded via app but nothing happens when I send the command via RT3 remote device.
Then I felt a smell of burnt and I disconnected everything .... :(

Now, I do not know what indeed happened that burned C19 but I guess the motor winding has somehow shorted S1 and S2 (may be via the N).

Edit: the black box close to SW2 han nothing written on the body ... is seems a capacitor but the polarity is not clear... there is a pink mark on the top suggestig that the positive is close to SW2 but I'm not sure at all.

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

Re: ESP32 SonOff Dual R3 Schematics.

#8 Post by TD-er » 11 Jan 2022, 12:23

That remote output, is that intended to connect to the motor directly?
If so, then you may have put a voltage differential of 230V (AC) on the S inputs.
There is probably a short circuit or load detection in the remote, so it could be that the remote first tries to check if something is connected by checking with a lower voltage first, or only for a short period.


Not sure if there is any other over voltage protection on the board, but I think you may have damaged more than just this capacitor.

Is the motor even intended to run at mains voltage?
The Sonoff dual seems to connect the Live wire to the relais output pins.
So there may be quite a voltage difference between the COM line and the UP and DOWN lines, where it was never intended in the design of that controller.

First you have to check the voltages on the UP and DOWN line when the motor is connected to the remote.
You actually need to know all combinations, but first start with checking whether the voltage between UP and DOWN is either AC or DC and what order of magnitude.
I think you may be very lucky if the remote and the motor have survived this.

Does the remote have relais in it, or doesn't it "click" when operating the motor?

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

Re: ESP32 SonOff Dual R3 Schematics.

#9 Post by TD-er » 11 Jan 2022, 12:25

Hmmm, in my search for "mhouse moovo" I found this topic:
viewtopic.php?t=2386

You can perhaps also look for sending the RF signals.

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#10 Post by miazza » 11 Jan 2022, 13:05

TD-er wrote: 11 Jan 2022, 12:23 That remote output, is that intended to connect to the motor directly?
If so, then you may have put a voltage differential of 230V (AC) on the S inputs.
There is probably a short circuit or load detection in the remote, so it could be that the remote first tries to check if something is connected by checking with a lower voltage first, or only for a short period.
Not sure if there is any other over voltage protection on the board, but I think you may have damaged more than just this capacitor.
Yes, the remote output is intended to connect to motor directly.
As you can see in the link of my post above, it drives a 2230V (AC) on the UP and DOWN windings of the shutter motor.
This is done via a relay (I can hear the click) that is commanded by the remote.
Because the phase of the remote device is the same of the Sonoff (they are phisically connected at the same wire) it is not expected to have any voltage differencial on the S input.
May be I can check with a tester the 230V on the pin 1 and 3 of the Remote device.
TD-er wrote: 11 Jan 2022, 12:23 Is the motor even intended to run at mains voltage?
YES
TD-er wrote: 11 Jan 2022, 12:23 The Sonoff dual seems to connect the Live wire to the relais output pins.
So there may be quite a voltage difference between the COM line and the UP and DOWN lines, where it was never intended in the design of that controller.
I'm not sure to have well understood the point.
The COM is phisically connected to the Neutral (from motor manual: Terminal 7 (Neutral) is directly connected to Terminal 2 (Common) of the motor).
TD-er wrote: 11 Jan 2022, 12:23 First you have to check the voltages on the UP and DOWN line when the motor is connected to the remote.
You actually need to know all combinations, but first start with checking whether the voltage between UP and DOWN is either AC or DC and what order of magnitude.
I think you may be very lucky if the remote and the motor have survived this.
Voltages of UP and DOWN line when the motor is connected to the remote is 220V otherwise the motor would not work.
The Motor works at 230V AC.
Anyhow I will check.
For sure motor and remote device have survived because the shutter is now perfectly working with only the remote.
TD-er wrote: 11 Jan 2022, 12:23 Does the remote have relais in it, or doesn't it "click" when operating the motor?
YES. As stated above.

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#11 Post by miazza » 11 Jan 2022, 14:24

Just to avoid misunderstandings about what I wanted to do and the mnistake I have done I post a pic.
Dual R3 Set-up.jpg
Dual R3 Set-up.jpg (152.97 KiB) Viewed 17621 times

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

Re: ESP32 SonOff Dual R3 Schematics.

#12 Post by TD-er » 11 Jan 2022, 14:24

OK, a bit more info on my thoughts here..

Typically an AC motor has only 1 rotation direction, regardless of how you plug it in a wall socket.
So I assume your "COM" is connected to Neutral and you either apply Live to "UP" or "DOWN"
It then depends on what is put on the non active line.
It could be "floating" or connected to "COM".

If it is connected to "COM", then you have a voltage difference on S1 and S2 which is equal to mains voltage.
This will for sure cause some sparks, and maybe more.
It is effectively the same as wiring it the wrong way as we discussed earlier.

I don't know how the motor is wired internally in the motor.
For example does it mechanically flip a "lock" to allow the motor to rotate the other way around, or is it internally a DC motor, and thus has some electronics in front of it?
So this still does not make it clear to me if the motor itself is operating at AC voltage or not.
Can you operate (or should you...) the motor without the remote?
It is still not clear to me what signal is present on the "UP" and "DOWN" pin.

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#13 Post by miazza » 11 Jan 2022, 14:31

TD-er wrote: 11 Jan 2022, 14:24 Can you operate (or should you...) the motor without the remote?
It is still not clear to me what signal is present on the "UP" and "DOWN" pin.
YES. The motor can be operated without the remote only with push button.
Indeed it was like that at the beginning.
The UP and DOWN are 220V AC and no electronic is inside the motor.

According to my understanding there are two windings inisde the motor with a COMMON in the middle.

If you power one winding the motor is rotation in one direction, if you power the other one the motor rotate in the other direction.

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

Re: ESP32 SonOff Dual R3 Schematics.

#14 Post by TD-er » 11 Jan 2022, 14:36

Ah the drawing you made makes it a lot more clear.

I cannot see how your mistake setup may have burned the cap in the Sonoff, unless you had the neutral connected to the motor.
What could go wrong is when the remote tries to pull either UP or DOWN to the neutral, then a closed relay may cause a short.


About what you consider "Right setup", may still be wrong, if the unused UP or DOWN output is pulled to Neutral.
Then you will for sure blow up stuff and that's the same as if you wired the switches to Neutral instead of Live.

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#15 Post by miazza » 11 Jan 2022, 15:33

TD-er wrote: 11 Jan 2022, 14:36 I cannot see how your mistake setup may have burned the cap in the Sonoff, unless you had the neutral connected to the motor.
What could go wrong is when the remote tries to pull either UP or DOWN to the neutral, then a closed relay may cause a short.
This is a good point.
The motor is made by two windings with the COMMON in the center : UP-----COMMON------DOWN
So when you select UP, the DOWN is connected to COMMON via the its motor winding and viceversa.
What happened with the wrong set-up is that as soon as I connected the "L in" of the sonoff, the motor was making a strange noise but not mooving !
When I restored the "Right setup", the motor was well working via the SonOff but nothing happened from the remote connected to S1 and S2 (may be because of the trevious mistake and blow of C19...),

TD-er wrote: 11 Jan 2022, 14:36 About what you consider "Right setup", may still be wrong, if the unused UP or DOWN output is pulled to Neutral.
Then you will for sure blow up stuff and that's the same as if you wired the switches to Neutral instead of Live.
Yes , I agree but this is not expected.... why should it be ?
At the end there is a relay inside the remote device and there is no need to pull one terminal to Neutral.
I do not know if this helps but the remote hase 3 push buttons:
1) UP
2) STOP
3) DOWN

I guess that the STOP put in floating both the UP adn DOWN.

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

Re: ESP32 SonOff Dual R3 Schematics.

#16 Post by TD-er » 11 Jan 2022, 16:13

There is a reason to pull both the UP and DOWN to the neutral.
It will break the motor and also impose an extra force to hold position.
If both lines are floating, then you can rotate the motor easier.

However, this can only be done if the other one is not active.

So the possible states:
- Moving UP: UP -> L DOWN -> floating
- Moving DOWN: DOWN -> L UP -> floating
- Not connected: UP -> floating DOWN -> floating
- Brake: DOWN -> N UP -> N

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#17 Post by miazza » 11 Jan 2022, 17:37

TD-er wrote: 11 Jan 2022, 16:13 There is a reason to pull both the UP and DOWN to the neutral.
It will break the motor and also impose an extra force to hold position.
If both lines are floating, then you can rotate the motor easier.
OK, if this is the case this shall be done internally to the motor because it is supposed to be used with a simple push button on the phase.
TD-er wrote: 11 Jan 2022, 16:13 However, this can only be done if the other one is not active.
So the possible states:
- Moving UP: UP -> L DOWN -> floating
- Moving DOWN: DOWN -> L UP -> floating
- Not connected: UP -> floating DOWN -> floating
- Brake: DOWN -> N UP -> N
YES. It would be interesting to understand which of the above really apply :)

But at the end I think that the best way forward is to try to change C19, test if the Sonoff works as expected and try again keeping the finger crossed :)

So, the initial question comes again: what is the value of C19 ?

I attach a pic after cleaning with IPA; it seems only C19 is actually damaged.
Attachments
DUAL R3 Burnt_cleaned.jpg
DUAL R3 Burnt_cleaned.jpg (1.62 MiB) Viewed 17605 times

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

Re: ESP32 SonOff Dual R3 Schematics.

#18 Post by TD-er » 11 Jan 2022, 19:10

miazza wrote: 11 Jan 2022, 17:37 [...]
YES. It would be interesting to understand which of the above really apply :)

But at the end I think that the best way forward is to try to change C19, test if the Sonoff works as expected and try again keeping the finger crossed :)
[...]
If the remote connects the COM to the Neutral, then you will fry the Sonoff for sure
So I would suggest to check if this may happen.

You can check it using a voltage finder (or whatever it is called in English, "Spanningszoeker" in Dutch)
- Connect the remote and the motor as you normally would.
- Test whether the COM or UP/DOWN is the Live wire when operating the motor. (write it down)
- Swap Live and Neutral to the remote
- Test whether the COM or UP/DOWN is the Live wire when operating the motor. (write it down)


Most interesting is the situation where COM is connected to Live.
If either UP or DOWN also lights up the voltage finder, then you know the remote does connect to what is intended to be the Neutral.

Also you need to take really proper care to not mix up L and N when connecting the remote to the Sonoff.
If you swap it, you will blow up the Sonoff.

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#19 Post by miazza » 12 Jan 2022, 09:12

OK. I got the point.

I will check it at the first opportunity when I will open again the roller shutter box.

Meanwhile I think I will buy a new Dual R3 because this one is really burnt in that area and also some trucks have disappeared (blowed) around C19.

Thanks for your assistance :) !

luizhenriquemm
New user
Posts: 1
Joined: 07 Apr 2023, 17:33

Re: ESP32 SonOff Dual R3 Schematics.

#20 Post by luizhenriquemm » 07 Apr 2023, 18:33

Hello everyone,

I have burned my R3 switch input as well and after some tries I repaired it. Some information that you guys posted here actually helped me, so here is my contribution.

How I burned it? I used different phases, one phase in the Line In (L In) and another in the switch input (S1). Why the board cannot accept this scenario? I guess that is something in the power supply section, some of the tests that I have made on the board suggested that the output is isolated (somehow), but I not sure of that, in other words, checking the voltage between the N In or L In and the S1 or S2, will always show the mains voltage, but that is only the power supply isolation in work. So if it's isolated, why it burns when using different phases? I really want to know that...

My burned board:
IMG_20230403_143315385_HDR.jpg
IMG_20230403_143315385_HDR.jpg (2.54 MiB) Viewed 6351 times
I must say that after the burn, the board still works, it was able to power on, connect to the wifi, in the eWe Link app I still was able to control it and both of the relays still work as before. The only thing that really was burned was the S1 and S2 inputs. Any voltage that I tried to apply on it, doesn't did anything at all. The reset button (IO0) was working fine as well.

After that I decided to try to fix the switch inputs. I dissoldered some components and starting to understand and draw the circuit schematic. Here is my results:
IMG_20230403_230246087_HDR.jpg
IMG_20230403_230246087_HDR.jpg (2.56 MiB) Viewed 6351 times
IMG_20230405_012149788.jpg
IMG_20230405_012149788.jpg (1.78 MiB) Viewed 6351 times
Notes:
S1 and S2 are the board connection for the switch input.
SW1 and SW2 are the GPIO inputs in the Esp32 auxiliary board.

After some effort trying to understand the logic behind this input, I had understood some important things (I guess):

1. The diode (D6 and D7) is for grounding half of the sine wave of the mains voltage, this is used for signal analysis without grounding from the source (remember of the power supply isolation feature).

2. There are pull up resistors (R29 and R31) from the 3.3v supply.

3. There is a LC filter for each input (L3 with C19 and L4 with C20).

4. There is an attenuator 10k resistor (R30 and R32) in each input.

With the circuit drawned and some understanding of the components functions, I started to test what remained of my board. The results of my tests:

1. The D6 diode was vaporized by the short, same happened with part of the pcb track in this input (S2)

2. The attenuation resistors was opened (both R30 and R32).

After that I started to replace these items. The diode was the most complicated, the one that remained in the board was labeled with "CC". What is a CC diode? I have no idea, I tryed to search but I got nothing, so I replaced both of them with some 1N4007 that I have here, just respecting the polarity of the board. I replaced the attenuation resistors as well, with the same value this time. I also used some wire to rebuild that part of the pcb track that was gone in the S2 input. Results:
IMG_20230405_011949072_HDR.jpg
IMG_20230405_011949072_HDR.jpg (2.09 MiB) Viewed 6351 times
IMG_20230405_012007736_HDR.jpg
IMG_20230405_012007736_HDR.jpg (2.04 MiB) Viewed 6351 times
What happened after all this work? The board started to work properly again! Both S1 and S2 inputs was working fine again, turning the relay on and off when applying the mains voltage at it. My conclusions? I don't know why the diode in the shorted input exploded, I think that the power supply is not fully isolated in someway that I can't understand, but the schematic is simple and that made things easier to repair. I hope this post can help anyone with similar problems.

Thanks!

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

Re: ESP32 SonOff Dual R3 Schematics.

#21 Post by TD-er » 07 Apr 2023, 22:01

The max. voltage between 2 phases is more than 230 V.
More like 400V.
So I guess the components are too small for this higher voltage and thus may have caused sparks.

The diode next to a coil looks like it may be intended to prevent high voltages when the current changes through this coil.

Albert38
New user
Posts: 1
Joined: 08 Apr 2023, 16:56

Re: ESP32 SonOff Dual R3 Schematics.

#22 Post by Albert38 » 08 Apr 2023, 17:35

Hi all,
The schematic of luizhenriquemm is right. But the assumption of the diode used is wrong this diode is a "BV03C Zener diode 12a 3v" The working of the circuit is simple (after bringing one of my R3 lite devices I found it).

The IO lines on the ESP32 are kept high by the pullup resistors R29 & R31 what has been drawn by luizhenriquemm as ground is actually Lin when we connect Lin to S1 or S2 what happens is that the IO pins of the ESP32 are pulled down and it is detecting a change.

Well now it is time to tell how I fried mine. This is telling you guys something more not to do.
I connected S1 to the wire coming from the switch, on that wire I also had a light bulb connected. You might immediately understand now what happened. My switch was initially on so no problem. Then I switched it off and what basically was done, the point S1 was connected to N via the light bulb. My diode D6 did not liked having to conduct the current trough the light bulb so it decided to blow ending up in a shorted diode so my light bulb did not go off anymore.

So what did I learn today: The inputs S1 and S2 have to be floating or connected to L1 otherwise the Sonoff Dual R3 will be fried..

Luckily on aliexpress we can buy plenty of these diodes it is item number 1005005350065104 or just search for BV03C and make sure to select the 3v type.

Hopefully someone can also figure out the value for C19 and C20

miazza
Normal user
Posts: 12
Joined: 10 Jan 2022, 12:28

Re: ESP32 SonOff Dual R3 Schematics.

#23 Post by miazza » 11 Apr 2023, 08:33

As far as C19 is concerned (and C20 also I guess) , from SONOFF Assistance I got the following:
"I am sorry that the Electrical Schematic is not for public. C19 is a 50V100nF"

Thank you for sharing your experience. This might hel me in understanding why I burned my R3 and may be how to fix it.
My Wifi is still working well and only S1 ans S2 are burned.

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

Re: ESP32 SonOff Dual R3 Schematics.

#24 Post by TD-er » 11 Apr 2023, 09:11

Maybe this is the same kind of "blow up": https://www.youtube.com/watch?v=qQ9Pxl4KAGw

Post Reply

Who is online

Users browsing this forum: No registered users and 37 guests