XYH3606, a universal adjustable converter with WiFi and other goodies

I'll start with a quick announcement: 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 maybe write about something interesting too - link to my account


I've reviewed a similar converter before—or at least one that looked very much like it. But in this case, the store calls it the Second generation, so this is attempt number two :)
So why not take a look and see what it's like?


One of my readers sent me this converter to test, and I'm very grateful to him. I'll do my part and see how it performs.
This converter looks a lot like the XYS3580, which I've reviewed before, and is an updated version of it, so they have a lot in common.

It comes in a small box, with everything neatly arranged inside: the converter and all the little bits are packed into compartments.


The specifications are shown in the picture, but I think there's a mistake: it says “Output current limit” instead of “Input current limit.”
The thing is, universal and step-up converters have not only a “maximum output current” rating, but also a maximum input current. In other words, a step-up converter might be able to deliver 1000 W, but that doesn't necessarily mean it can do so with a very low input voltage.
Here's a calculation for this model. It's rated to operate from 6 V, with a maximum input current of 12 A, so without accounting for efficiency, you can get up to 6 × 12 = 72 W. To reach the rated 216 W (again, without accounting for efficiency), you'd need 216/12 = 18 V. Taking efficiency into account, I'd recommend using a power supply rated for 22–24 V and at least 250–300 W for full performance.


The package is quite extensive and includes:
1. Converter
2. WiFi module with connecting wires
3. Remote control
4. Temperature sensor
5. Diode isolation board
6. Fuse



1. The remote is a very basic one, like the ones that used to come with TV tuners, though the button labels are of course different here.
2. 2.4 GHz WiFi module
3. “Drop” thermistor
4. A 15 A fuse and a small board with a pair of SS56 diodes connected in parallel.

There's no information about the diode board. I assume it's installed between the converter output and the load to prevent current from flowing backward if power is cut while charging a battery. But the diodes seem rather small for 6 A.



The case and control layout are typical for devices like this.


I didn't measure the dimensions; they're in the manual, which also gives the dimensions of the mounting cutout.


1. One unusual feature is the small window for the remote's IR receiver.
2. There's a small fan at the back to cool an equally compact heatsink. Given the claimed 200+ W, that seems rather modest. The previous converter, remember, was rated at just 80 W.
3. The terminal block is also typical; I've seen exactly the same one on RuiDeng converters, and I can't say I'm a fan.
4. The terminals are labeled. Note that the input common and output common must not be connected together; they're separate lines.


Compared with the previous version, they made the display larger, which is definitely a plus—the old one was really too small.


1. The components are packed in fairly tightly. You can see the fuse holder on the right; at least they didn't decide to solder the fuse in place.
2. This is also where the fan, temperature sensor, and WiFi module connect.
3. The power inductor is on the other side. The input and output capacitors are identical: 470 µF, 50 V. The quality is so-so.
4. Under the inductor is the PWM controller chip, with its markings removed.


On the back are the current-sense shunt, an op-amp, a reverse-polarity protection transistor (for a permanent connection, it's better to bypass it for higher efficiency), and a regulator for the secondary circuits.

Judging by the single inductor with a single winding on the board, this isn't a SEPIC like the previous model, but a true universal converter. So I assume it uses an LTC3780 chip; I've already reviewed a board based on one here. In short, it’s good—the chip works very well.


I didn’t remove the front-panel board; the main components are visible as it is. In this case, they’re the microcontroller, voltage regulator, and optocoupler. There’s also a buzzer hidden in there.


Since this isn’t my board and the heatsink is glued on, I left it in place and moved on to testing.
The converter was powered by an adjustable power supply based on the RD6018.
When powered on, the screen displays the logo and the word SINI.


The control scheme is unchanged: there are several tabs, and you move between them by turning the encoder.
1. Main screen: voltage, current, power, set values, temperature, output mode, and status indicators for WiFi, the IR receiver, Shift (available from the remote), and control lock (double-click the encoder).
2. Additional screen: delivered capacity in Ah and Wh, and input voltage
3. Graph mode. I find it pretty pointless, but it might come in handy for someone.
4–6. Protection settings for the auto-shutdown modes: current, power, temperature, capacity, etc.
7. Settings menu: brightness (3 by default), auto-shutdown time, temperature sensor calibration, sound and remote control on/off, color scheme, and PC connection.
8. The display has plenty of brightness; even at the lowest setting, the readings are easy to see.
9. When certain functions are turned off (sound, for example), the corresponding indicator goes out.


Here’s the description from the manual. It’s in English, but mostly easy enough to understand as is.


You can change the color scheme, but after trying them out, I found that only two or three are really comfortable to use.


1. When control is locked, the padlock icon changes. At first, I couldn’t figure out what had happened or why the controls had stopped working :)
2. The maximum setting is 36.5 V and 6.1 A. Front-panel control is just as awkward as before: click the encoder to activate adjustment, click again to adjust the hundredths, click again for the tenths. To adjust the current, you have to turn the encoder after the first click. But now there’s a remote, so you can set the voltage directly, just like on the RD60xx. You could even tape it to the panel with electrical tape :)
3. With a 24 V input, the output enabled, and the voltage set to maximum, power consumption is about 2 W with no load.
4. With the output disabled, power consumption is close to zero at both 12 V and 30 V input


I’ll skip checking the input voltage measurement accuracy, since I don’t consider it important—and everything was fine there anyway—and move on to evaluating the accuracy of the output voltage setting and measurement.
The usual test: a 1, 2, 5 sequence starting from the lowest possible value, which in this case is 0.01 V.


Yes, you’re seeing it right: in places, the screen shows some kind of nonsense. I originally wanted to show a few photos, but then thought it would be easier to understand if I showed a GIF instead.


At 1 V, things stabilize, and at 2 V and above, it holds the voltage perfectly.


The current behaves much the same way: it doesn't hold at low values, and you can see this on the converter's own display, too.


Here's the same thing in action. I think it's easier to see this way—and if you put on some upbeat music while you're at it, it's more fun too :)


Here, too, things get more stable at higher current values, and the setting accuracy itself is pretty good overall.


I checked how the output voltage varies across the full load range, with 24 V at the input and 3.3, 5, 12, and 24 V at the output.
Overall, not bad. I think the terminal block made the results worse; without it, they would have been even better.


I measured efficiency at an output current of 1 A and at the maximum current possible for each operating condition. Since heating is affected by operation at maximum values, I'll only include those results.
I tested at input voltages of 12, 24, and 30 V, but focused mainly on 24 V, since power supplies with that voltage are easier to find and it lets the converter operate across its full range at maximum power.
12 V input:
12 V output, 6 A, efficiency 91.8%
24 V output, 5 A, efficiency 84.8%
36 V output, 3.3 A, efficiency 84.1%
24 V input:
5 V output, 6 A, efficiency 91.3%
12 V output, 6 A, efficiency 93.3%
24 V output, 6 A, efficiency 94.8%
36 V output, 6 A, efficiency 91.7%
30 V input
12 V output, 6 A, efficiency 92.8%
36 V output, 6 A, efficiency 94.4%

The actual efficiency was a little higher, since I didn't account for the voltage drop in the wires. You can also improve it by shorting the reverse-polarity protection transistor, but even as it is, I'd say it's pretty good.


The converter can actually handle more than its rated 12 A, but as they say, “trust the protection, but don't exceed the power rating.”


Ripple in different operating modes.
First, with a 12 V input and the following output settings:
1. 5 V, 6 A
2. 12 V, 6 A
3. 24 V, 5 A
4. 36 V, 3.3 A


And with a 24 V input and the following output settings:
1. 5 V, 6 A
2. 12 V, 6 A
3. 24 V, 6 A
4. 36 V, 6 A


Transient response testing
As usual, I tested it in several operating modes.
1, 2. Applying/removing 5 V at the output.
3, 4. Response to connecting/disconnecting a load. The output was set to 12 V, with a 30 Ω resistor connected continuously and a 3.3 Ω resistor connected in parallel with it (or disconnected). So this tests the response to a current step from 0.4 to 4 A and from 4 to 0.4 A.
5, 6. Switching from CV mode to CC mode and back. Voltage: 14 V; load current: 4.5 A; current limit set to 3 A.

The first four tests are all fine, apart from the output voltage dropping very slowly—it looks like there's no discharge circuit. But the CC mode is a disaster, and the converter starts whining, too.


On the other hand, it does fine in the “soldering iron” test.


Thermal performance.
First, I tested it at 12 V output and 6 A, then at 24 V and 6 A. The input was 24 V in both cases.
Overall, it's fine. I wouldn't say it gets very hot, and it feels like more heat is dissipated underneath the board. Judging by the component layout, that's the shunt (0.9 W) and the reverse-polarity protection transistor.
1. First test.
2, 3. Second test, photos from above and from the connector side.


Then, with the same 24 V input, I set the output to practically its maximum: 36 V and 6 A, for an output power of 215 W.
I ran it like this for a while, until the temperature stopped rising according to the internal sensor, leveling off at around 81 degrees.


The picture is obviously worse here. I wouldn't call the heatsink hot, but the shunt, fuse, and protection transistor get seriously hot; in places, the temperature even exceeds 120 degrees.
I don't know the temperature of the converter transistors—I can't get to them—but I suspect it's just as high. The heatsink is glued to their cases, so it obviously won't dissipate heat very well, although the air coming off it is quite hot.


External connections.
Temperature sensor. When an external temperature sensor is connected, the temperature readout alternates between the readings: the internal sensor on the usual background and the external one on a blue background.


Remote control.
No complaints here. In my opinion, it's even convenient—and as I wrote above, you could just tape it to the case. The controls are essentially no different from those on the RD series converters.


WiFi connection.
When you connect the WiFi module to the converter and power it on, the controller detects it and adds the corresponding icon on the right. The menu also gains one additional item.
The new menu item lets you choose the operating mode: Disabled, Touch, or access point.
In theory, after selecting Touch, the device should search for connection options; after selecting AP mode, you can connect directly to the device.


The idea is pretty good. As I understand it, this should be more convenient to use than the RD series converters, and the software looks interesting too.


But, as so often happens, "something went wrong"—specifically, the software requires a username and password, and I couldn't find or guess them.


All right, if the PC software won't work, let's try another way and install the Android app. Fortunately, it's also available as a free download.
I install it and launch it. The software offers to update, so I agree.


The software updated, installed, rebooted, and... asked to update again. I repeated the process; it asked for a third time, and at that point I decided to give up.
The app launched and, once again... asked for a username and password. That's as far as I got.

Maybe I'm looking in the wrong place. I'll try to figure it out, and if I do, I'll update the review. Perhaps someone can tell me what I'm doing wrong; I'd be very grateful.


Conclusions.
While inspecting and testing it, I kept thinking: this is actually a pretty good converter. It has a few minor issues—the voltage and current readings fluctuate at low values—but the price, size, and features are good, and they even included a convenient remote and WiFi.
That was true right up until I saw what was happening at the output in CC mode. This was with a 24 V input, 14 V output, moderate current, and a resistive load—in other words, practically ideal conditions.
The password needed to launch the software was the last straw, after such a promising start.

I think the password and WiFi setup can probably be figured out somehow—perhaps someone can suggest what to do, or maybe it makes sense to contact the developer. But the ripple is a problem.

The price on Banggood is high at the moment, but the converter is available on AliExpress for $32.30 with WiFi .

That's all for now, but I'll try to figure it out. If I do, I'll update the review.