Bob found these ESP8266 based Wi-Fi Relays from china, the “Sonoff – WiFi Wireless Smart Switch For MQTT COAP Smart Home” and being that he is is kind of a geek he bought some, now in typical China fashion the difference between 3 and 5 was not all that much, so he got two extra for me to glance over.
Please remove mains from the unit if you are looking at the unit! this is not a Class II device and 230 VAC is present on both sides of the PCB.
Sonoff is a WiFi wireless switch that can connect to appliance of different types and brands. Sonoff sends data to cloud platform through the WiFi Router, which enables you to remotely control all appliances with the App eWeLink on your smartphone.
The manufactures specification is as such:
- Voltage range: 90-250v AC(50/60Hz)
- Max current: 10A
- Max Wattage: 2200 watts
- Dimensions: (L)88*(W)38*(H)23mm
- Color: White
- Humidity: 5%-95%
- Wireless Frequency: 2.4Ghz
- Working Temp: ﹣20℃-75℃
Now the eagle eyed among you will have spotted the lack of CE or FCC marks anywhere on this! a quick look at the user guide also fails to mention any form of compliance.
So without much more waiting around let us have a look at the product.
The ESP8266 module is on the PCB, however does not seem to have any details of an FCC number, so is may not be a pre approved module, however they do make it easy to get to the UART.
Looking at the main components on the board, the relay shows us that it is part number HRS3FNH-S-DC5V-A
This particular part number does not appear to have an English data sheet available, so lets glance over one in Chinese (using the bits gleaned from Aliexpress for a different model as a reference).
- HRS3FN – Model
- H – High Sensitivity 200mW coil.
- DC5V – Coil Voltage
- A – Contact form (A or C) this claims that the Contact is rated for 10 A @ 250 VAC.
There are two 5mm Pitch mains terminals on the PCB, These should be rated to the same or greater current as the box tells me it is rated for, so these should be safe for use with 10 A @ 230 VAC (the box also states 2.2 kW). Breaking out a bit of random scrap mains wire, I am going to make a test probe for these terminals:
Looking at the screw terminals a bit more, they look a lot like these Screw Terminal (5mm) from SparkFun (and other sellers). These screw terminals fail the strand test above, and the data sheet rates them for 8 A…
The layout is nice and tight, but is it too tight! the gap between Live and Neutral is a bit tight, I think it is safe to say that I feel it is a bit too close, The requirement for creepage distance between Live and Neutral before any fusing is a minimum of 2.4mm; there is no fusing on this PCB at all.
If you read the post about the HLK-PM01 you will know why I am using BS EN 55032 and not BS EN 55022 (CISPA 32 and CISPA 22), if not, its because BS EN 55022:2010 was Withdrawn on 05 March 2017, and replaced with BS EN 55032:2012, and CISPA 32 doesn’t have quite the same get out clauses as CISPA 22 did:
As we can see from the graphs above, its a fail on Conducted Emissions (Live Side has measurements, neutral didn’t. However they are very smiler). however on the other scans its a pass.
All in all, from a unit point of view, it is an interesting unit that I may be tempted to use (Fused at 5A), HOWEVER it does not meet the requirements of relevant standards, it is not electrically safe. It would be illegal to place this unit as is on the market within the European Union.
If you are making a product that users are required to wire it in or plug in, Please, please send CASS Industries an email to ask about testing!
The test results on this page are indicative, and in no way constitute evidence of a result!