Compact converter with QC support and two outputs

I got my hands on a dual-port converter with QC support for a little while, so I decided to test it a bit. Maybe someone will find this useful.

As a reminder, to stay up to date with new posts on the blog, I recommend following me on Instagram. I’ll post notifications about all new posts there, and maybe write about something interesting too: link to my account

The seller offers three versions of these converters, with one, two, or four outputs. But essentially, they’re all the same model, just with a different number of converters on a single board.
Also, it looks like I’ve had the single-port version before, though it gave slightly different results. If you’re interested, you can read that review too.


It comes in a regular antistatic bag, with a small, neatly made board inside.


1. You can easily make out the two independent converters on top.
2. The list of supported protocols is printed on the bottom, but I’ll say up front that not all of them are actually supported.
3. It’s built around a dedicated DC-DC chip. The marking has been carefully erased, but as far as I know, it’s an IP6505.
4. Like the single-port converter, it has 12 V TVS diodes on the output, and a reverse-polarity protection diode in the center of the board.

As you can see, there are already some differences between the single-port and 2/4-port versions. The latter have a protection diode. Unfortunately, it’s simply connected in parallel with the input, so a fuse in front of it is essential. There are small fuses on the board, but each one is installed on an individual converter, so they won’t blow if the input polarity is reversed.


1, 2. You can supply power either through the board’s contact pads or via a standard 5.5×2.5 mm connector. That’s another difference, by the way: the single-port board doesn’t have a connector.
3, 4. The board is about the size of a matchbox, maybe a couple of millimeters longer.


For testing, I used a power supply based on the RD6006P. One thing to note right away: the blue LEDs are very bright, especially when the output voltage is 12 V.


As it turned out, the board’s no-load power consumption is very low. At a 5 V output, it was 4 mA with a 12 V input and 3 mA with a 24 V input.


As I mentioned, the manufacturer is being a bit liberal with the protocol list. It looks like only QC2/3, FCP, and AFC are actually supported, plus Apple and DCP. Just to be sure, I checked with two different testers.


A closer look at the protocols and outputs.
1, 2. QC works as expected, but only up to 12 V, and from 3.76 V in QC3 mode.
3. FCP is supported and works.
4. SCP isn’t supported, and the tester even throws an error.
5. AFC 9 V is supported and works.
6–9. None of the other protocols I tested are supported.


Testing operation in different modes.
First, Quick charge: 24 V input, with 5, 9, and 12 V output.

5 V: maximum 3.3–3.4 A. It compensates for voltage drop in the cable.


9 V: maximum 2.3 A. It also compensates for voltage drop, but interestingly, as the current rises to a certain limit, the voltage drops before the output shuts off completely.


12 V: maximum around 1.5–1.6 A. It also compensates for voltage drop, and the voltage gradually falls as the current rises, but this time the drop is more gradual.
As you can see, at both 9 V and 12 V, the output shuts off at 3.4 A. Below that point, the output voltage simply drops.


12 V with a 14 V input: it behaves just as before, but I’ll note that in this mode the converter maintains 12 V at the output even at maximum current.


FCP and AFC modes


Ripple was measured by connecting the probe directly to the contacts of the USB connector on the board.
1, 2. Input: 14 and 24 V; output: 5 V, current: 1.5 A
3, 4. Input: 14 and 24 V; output: 5 V, current: 3 A
5, 6. Input: 14 and 24 V; output: 12 V, current: 1.5 A

Well, what can I say? First, the higher the input voltage, the greater the output ripple. Second, at a 5 V output, the ripple amplitude was 350–550 mV; at 12 V, the range was wider, from 200 to 700 mV.
As a percentage, ripple is significantly higher at 5 V than at 12 V, reaching about 10% and 6%, respectively, in the worst case.


Heating: the situation is better here than with the previous converter. That one went into protection after 7 minutes at maximum load current, but this one didn't.
1. Ran for 20 minutes with a 24 V input, 5 V output, and a 3.3 A load.
2. Increased the output voltage to 12 V, reduced the current to 1.4 A, and ran it for another 20 minutes.
3. Set the input to 14 V, the output to 5 V, and the current to 3.3 A, then ran it for another 20 minutes.

As you can see, the inductor is the hottest component. The chip reached 95–97°C, which is rather high, but it still didn't go into protection.


I measured efficiency just for the sake of it.
As in the previous review, I tested at different input voltages (14 and 24 V in this case) and different output voltages (5, 9, and 12 V). I set the load current so the converter was operating at about 20, 40, 60, 80, and 100% of its capacity.


And here are the results as graphs. The horizontal-axis values are in 20% increments; 14 V input is on the left, 24 V on the right.



For a more practical test, I connected a couple of phones to the board. First, my own “Chinese” phone, which drew the expected 2 A at 5 V, then my wife's Samsung, which made the converter switch to AFC 9V mode. The current was a bit low in the second case, but that may have been because the phone's battery was 90% charged.


I can't say with 100% certainty, but for some reason I got the impression this is a different chip. The single-port version in the previous review delivered up to 2.5 A at 9 V and up to 2.3 A at 12 V, and went into protection immediately rather than gradually lowering the voltage.

What the previous converter and this one definitely have in common is a fairly high level of output ripple. Of course, it will drop significantly after the connector, and even more after the cable, so it won't harm your device. Still, I'd like to see a lot less of it.

Can I recommend this converter? Generally, yes. I know a lot of people use these, and there are very few complaints about them in actual use. I use one myself—the single-port version—and have no complaints either.
Personally, though, I'd recommend that anyone using it in a car add at least a basic input inductor to reduce spikes from the vehicle's electrical system.

That's about it. I hope this was useful :)