Single-phase WiFi eWelink relay rated at 63 A
$19.99
As a reminder, if you want to keep up with new posts on the blog, I recommend following me on Instagram. I’ll post notifications about every new article there, and might also share other interesting things from time to time: link to my account
At first glance, you’ll probably think the only differences between this relay and the previous one are the price and size. But that’s not quite right—in fact, it’s not right at all. Still, I’ll stick to my usual format and start with the packaging and what’s included.
Everything is the same here: a small box, the relay, and its manual.

The manual is all in English, but it has pictures, so it’s pretty easy to understand. As before, I was interested in the specifications. The key ones are:
Input voltage range: 120–280 V
Switching current: up to 63 A
Power consumption: no more than 3 W

Now let’s take a closer look at the relay itself.

It’s all pretty standard: designed for a standard 35 mm DIN rail and one module wide. That’s actually quite unusual for a relay with the stated switching current.
Here it is compared with standard circuit breakers and the previous relay. You can see that the circuit breakers are a little shorter, while the relay from the previous review is the longest of them all.

1. There are two LEDs on the front panel: a service LED (WiFi connection and setup) and an output status indicator.
2. As you can see in the previous photo, there’s a mysterious little wire sticking out of the relay. The thing is, with a case this small, it’s hard to fit a proper terminal for connecting the neutral wire as well (though not impossible). Since the current in this circuit is low, they simply brought a wire out of the case.
3. The terminals are standard, for wires up to 16 mm². I didn’t have any wire that thick, but it holds 6 mm² wire just fine.
4. The output terminal. But what I wanted to show wasn’t the terminal itself, but the hole just below it, clearly intended for the wire routed through the slots in the case.
5. There are connection setup instructions on the side, though they’re not much use since you can’t see them once the relay is in the electrical panel anyway.
6. The clip isn’t spring-loaded; it has two positions: open and closed.

I’m getting ahead of myself again by showing the app before describing it. Sorry about that, especially since this isn’t really what I wanted to talk about.
The point is that they got the output status LED right here: it glows red when the output is active and turns off when it isn’t. Also, the app gives haptic feedback when you toggle the relay, and the background changes from dark to white (off/on). Handy.

The wattmeter starts working at around 52–53 V, though it can sometimes behave erratically.

But raise the voltage a little and everything works perfectly. Unlike the previous relay, this one doesn’t work while failing to connect to WiFi.
Voltage measurement accuracy is generally fine, but it’s very difficult to check properly because there’s a long delay before the readings appear.
The relay behaves in an interesting way: if the voltage or current changes significantly, the readings update quickly, sometimes as often as once per second. But if the fluctuations are small, it can easily take half a minute or even longer between updates.

The current readings are more or less the same story: a little worse than the previous relay, but generally fine.

The voltage drop across the terminals is about the same too, so I’d also limit the maximum continuous current to around 40 A.

As with other devices that have a high current measurement range, I tested it using a low-voltage transformer and a clamp meter.

The results are generally fine here too, although the delay in displaying readings meant that the difference between the clamp meter and wattmeter readings sometimes got smaller and sometimes larger. The clamp meter is much faster, after all. The photo shows the average readings.

1. But when I raised the current to the stated maximum, the device started lying: instead of the 61.5 A I was feeding it, it displayed just 54.39 A.
2. I lowered it to just under 50 A, and it reads correctly.
3. I raised it to 63 A to double-check. Nope, now it’s showing 52 A.
4. I found empirically that it displays currents of around 55 A correctly, but underreads anything higher.

If you thought that was all, you’re very much mistaken. The whole point is that I should have been looking not at the current—or rather, not only at the current.
1. I set it to 42 A, and everything looks great: current and power.
2. I raise it to 45 A. The current reading is fine, but for some reason the power drops to 7.7 kW, even though it should have risen to 10 kW.
3. Just out of curiosity, I tried 80 and 90 A. At those levels, both the current and power readings are completely crazy...

I didn’t stop there, and decided to swap the phase and neutral wires on the load. As with the previous wattmeter, it now reads a little high, but not by much.
The key point is that this had no effect on the current/power measurement error, although the threshold for correct operation went up slightly, to around 44 A.
I should also point out that above a certain load, both the current and power readings go “drunk”—that is, at times they have no direct relationship to the actual current.

Of course, I could say I was just “lucky,” but I came across a review comment from someone who also complained about power measurement problems—only things were even worse for him.
The measured values are not accurate; according to the calculation, 230 V * 0.58 A = 133.4 W, but it reports 6.3 W

I checked the temperature with a load current of about 60–62 A, and I have no complaints here.

Now let's move on to the software.
It's simple: find the QR code in the manual and follow it to the manufacturer's website, where you can download the software. You can also get it directly from the app store; it makes no difference. The software's rating is still very low, just 3.3 stars.
Next, open the app, register using your phone number or email, agree that the manufacturer takes security seriously and will later share our data with everyone, and move on to the setup.

It's more or less the same process as with the previous wattmeter, except that for some reason the connection took a long time, and the success message appeared just as I disconnected power from the wattmeter. I thought it had simply frozen...
Speaking of "frozen," the software takes ages to cold-start—about 15 seconds, compared with just 5 for Tuya.

The settings are more or less the same too: timers and schedules. There are a few differences, though. For example, the main screen has a black background when the output is off and a white one when it's on.
The tiny power/current/voltage readouts have made their way here too. To be fair, they're a bit larger than in Tuya/Smart house, but the interface is still just as lousy—which is no surprise when everyone copies everyone else.

The settings are also largely similar, but there are some interesting exceptions, for example:
1. Network activity indicator: you can turn off the LEDs on the panel, although for some reason the relay then stops responding to on/off commands...
2. State after power-off: I usually set this to "last state".
3. Off delay: set the interval after which the relay will turn off once switched on—in other words, this simulates a momentary switch, or push button.

There are various other options, but some are only available if you pay.

The software can keep logs and also lets you save consumption data to a CSV file, although the information there is very brief.

But what interested me more was the protection feature. First, it warns you that you should probably read the manual, and only then lets you access the settings.
You can set minimum and maximum limits for power, voltage, and current. For the experiment, I set the voltage limit to 242 V.

I apply voltage from a Variac. The output is active; I raise the voltage to 260 V, and after about 1–3 seconds the output switches off. I lower the voltage again, but the output remains inactive. There may be a setting somewhere to reactivate the output automatically afterward, but I couldn't find it.
If you try to turn the output on manually, nothing happens: it stays inactive, and the on/off command is recorded in the logs.

To test it further, I closed the app on my phone. The relay switched off normally, so I was already pleased. Then I tried turning off Wi-Fi altogether, but that was a letdown: the protection doesn't work in that case.

Of course, I was curious to look inside, especially after the odd current and power measurements. Unfortunately, there's no easy way in, so I drill out the rivets and take it apart this way.

Inside are a small circuit board and a relay. Judging by its markings, the relay is rated for 60 A, so with some headroom it would have made sense to use a 40–45 A relay—which is about what it can handle at most during measurement.
The input terminal is connected to the relay by a wire with a current transformer mounted on it. That's the key difference from the previous devices.
The relay's output terminal is designed to also serve as the device's output terminal.

To remove the board, you first have to take out the button and the indicator light guides. The board is also crooked in the housing from the start.
The current transformer and relay are connected via plugs. A small touch, but a nice one.

The power supply section is on the top of the board. Credit to the manufacturer for not using a capacitive dropper. On the left is a 1 A fuse.

There is also a 470 V (330 V RMS) varistor on the input, but here’s the problem: not only is the filter capacitor rated for just 400 V, the varistor is also connected before the fuse. That’s in case the capacitor somehow fails to trip when the voltage gets too high...

At the bottom are the WiFi module and a dedicated CSE7759B chip for measuring electricity consumption. What would it take to make it perform poorly?

The relay’s construction was a surprise. It’s actually an integral part of the device’s enclosure—or, more precisely, the device enclosure is the relay enclosure.
It’s an interesting solution, and I think it’s perfectly reasonable. Of course, someone will ask, “How are you supposed to repair it now?” The same way as the others: you don’t.
The relay looks perfectly decent, by the way. I think it can switch its rated 60 A, though as I wrote, 40 A is probably a more realistic figure for continuous use.

Conclusions.
Overall, this relay is pretty questionable. To start with, I have a strong suspicion that it was originally a 40 A relay, but then the manufacturer decided it wasn’t selling well enough and bumped up the current rating... by changing the label.
So, with a claimed rating of 63 A, we get:
1. A relay that can switch 60 A
2. But current can only be measured up to 50–55 A
3. And if you want the power measurement to be accurate too, the limit is 40–42 A
In practice, this is a device rated for up to 40 A; everything else is marketing. But even that isn’t the problem. Not having a fuse before the varistor could make things exciting if the varistor fails, and fixing it means drilling out the sleeves. A real mess.
Also, how hard would it be to let the protection thresholds work without a cloud connection? That alone would make it much more useful.
All in all, this could have been a decent compact 40 A relay if not for the marketers who listed a current rating 1.5 times higher, and the engineers who forgot that the fuse goes before the varistor, not after it.
I’ll point out two advantages:
1. The price: in my opinion, $20 is more or less reasonable.
2. The size: it takes up just one module in the panel, half the space occupied by most similar devices.
The store provided coupon BGb51e5f, which brings the price down to $19.99 through March 18. That’s all from me; I hope this was useful, and as always, I’ll be glad to answer your questions and read your comments.
The same relay is available on AliExpress for a little less - link.
$19.99
This page was found when searching for:
2, https://www.kirich.blog/obzory/izmeritelnoe/1092-odnofaznoe-wifi-rele-ewelink-s-zayavlennym-tokom-63a, 63a ewelink manual, wifi реле smartmcb макс 63a ном 40а ewelink
2, https://www.kirich.blog/obzory/izmeritelnoe/1092-odnofaznoe-wifi-rele-ewelink-s-zayavlennym-tokom-63a, 63a ewelink manual, wifi реле smartmcb макс 63a ном 40а ewelink
The figures from this review are in the measurement database: Other devices
Discounted products
You may be interested in
Discounted products
Support the project
All the content on the site and the advice are free, but we’d be grateful for your support of the project and the channel!
Discuss on the forumHave a question or some experience to share? Discuss the article with the author and readers: the forum is a convenient place to share schematics and photos.
Go to the forumOnly registered readers can comment. It takes a minute to sign up, and you can log in with the same account you use on the forum.
Log inRegister



