Cheap used QC 3.0 charger

I was browsing Taobao and happened to come across a charger that caught my eye because of its low price and pretty decent appearance, especially considering the price. I ordered it, received it, and was somewhat disappointed—twice, in fact. But more on all that in the review.
Inspection, testing, conclusions.



For various reasons, I won't write much today, and I'll start with what disappointed me right away.
I spotted a charger on Taobao for a laughably low price, about a dollar and a half, plus 1.88 for shipping to the forwarding agent, and then the agent's own shipping fee. I put three in my cart, thinking they might come in handy. Some time passed while I picked out other items, and once the order reached a "critical mass," I placed it.
For some reason, the quantity changed from three to one during the process. I thought I'd messed something up, but when I checked the screenshot, I found that wasn't the case—the seller simply had only one left.
I also found out that the forwarding agent offers 90 days of free storage, and in my case I should have placed the order while the items were still available. Anyway, that ship has sailed, so let's get on with the review.

They wrapped my little board up like crazy: one layer of packaging from the seller, then an envelope and another layer from the forwarding agent. The whole thing ended up about three times bigger.


Inside is a fairly neat little board, not much different from the photos on the seller's page. By the way, the product link no longer works, which is why I used my own photo in the title.


A couple more general photos, in case anyone comes across one of these.




One reason I decided to order it was that the input filter looked decent. It turned out, though, that it doesn't have a thermistor, for example; it's been replaced with an ordinary resistor. At this power level, that's not too critical.


The second reason was the input capacitor, with a capacitance of 47 µF—reasonable for the claimed 40 W output.
There are also quite a lot of noise-suppression capacitors on the board, which are usually the first thing manufacturers skimp on. What's more, there are even two interwinding Y capacitors: one from the pair connected to the input connector, and the other from the negative terminal of the input capacitor.


A reasonably large transformer, along with a heatsink for the high-voltage transistor.


The output consists of six USB ports: two vertical and four horizontal. One of the horizontal ports has a different-colored plastic insert and provides the QC output.


The charger can deliver up to 40 W. When drawing power from the 5 V line, the claimed output is 8 A. A pair of diode assemblies is mounted on the second heatsink.


The second rectifier and the DC-DC converter are mounted separately; the converter is powered by 18 V.


The underside of the board is fairly neat, apart from someone's ham-fisted repair work. Judging by the soldering, the high-voltage transistor was replaced, and the input connector was also desoldered because it was clearly in the way. What surprised me was that only the transistor had been replaced. Usually, when it fails, you also have to replace at least the fuse and current-sense resistor, and often the PWM controller as well—and in some cases, the bridge rectifier.


The second thing that surprised me was the spark gaps.
The first one is more or less fine, though a slot in the board wouldn't have hurt.
The second one is very strange: there's a slot, but the left side of the spark gap isn't connected to anything.


1. PWM controller. Since I only have this one charger, I didn't bother trying to figure out the marking.
2. Feedback uses the familiar TL431, but it's connected to both outputs: 5 V and 18 V.
3. DC-DC converter chip HX114G.
4. QC chip — NT6008


The buck converter's PWM controller has a synchronous rectifier, operates at frequencies up to 500 kHz, and is originally designed for a 5 V output. In this case, though, it works together with the QC controller, so the output voltage can be higher.


The QC controller can control the converter's output voltage, but only up to 12 V.


I'll come back to the QC controller later. For now, let's move on to the tests, starting with a basic functionality check.
We apply power, and the first thing we see is a painfully bright blue LED. And it was so garish that I had to cover it up with something—in short, it was everything you'd expect...


The six ports are configured as follows:
1. Two horizontal and one vertical port following the Samsung standard
2. One horizontal and one vertical port in the Apple configuration
3. One horizontal port is supposed to support a voltage range of 5–12 V, but in reality it's "stuck" at 9 V. So anyone who plugs it in without checking first may get an unpleasant surprise when connecting a regular (non-QC) smartphone or tablet.

A quick inspection showed that the QC controller was faulty. It is powered from the same USB port through a 100 Ω resistor, but while there's almost 9 V before the resistor, there's only about 3 V after it, and the controller doesn't respond to "external stimuli." I checked AliExpress, but they don't have any there. I eventually found some on Taobao for 30 cents each, with free shipping to the forwarding agent, but you have to buy 10. I ordered them, so I think there'll be another review, possibly with a demo of a simple QC charger :)


Anyway, for now we'll test only the 5 V output. I'll check the rest when the QC chips arrive, assuming it works, of course.
Let's connect the electronic load, multimeter, and, of course, the oscilloscope.

With no load, the output voltage is too high: 5.36 V, above the 5.25 V limit.


First, I ran a load-capacity test and checked the overload protection. Almost everything was fine: the power supply delivered the rated 8 A, and even a little more. The voltage stayed stable until shutdown, then the protection kicked in, and once the load was removed, it returned to normal operation.
There is one minor issue, though: in some cases, the protection tripped even at 8 A. It happened when raising the load current from 6 A, even though the power supply could operate normally at full current.


Output voltage at load currents of 2, 4, 6, and 8 A. Aside from the fact that the output voltage only drops to a normal level at 2 A, the results are pretty good.


There is some ripple at the output with no load, but it's not particularly significant.


Under load, though, things are much worse. The lack of an output inductor and the low capacitance of the output capacitors are to blame—or, more likely, their poor quality...
At 2–4 A, the ripple is about 80 mV, but at 8 A it reaches 150–200 mV, which is quite a lot for a 5 V output.


As usual, the load test ran for 20 minutes at 2 A, then another 20 at 6 A, and 20 at 8 A. The total test time was just over an hour.


I tested the power supply without its case, which I don't actually have. It ran at a moderate temperature, but it would certainly get noticeably hotter inside a case. I'd estimate the maximum output current at 6 A in a case and 8 A in the configuration shown.
The output diodes got the hottest, followed by the transformer, and then the high-voltage transistor.


The voltage has a positive temperature coefficient, and the change is fairly small—about 25 mV between cold and hot conditions.


That's all the testing. The power supply left me with the impression that they took a fairly decent design, cut costs a little by using no-name capacitors, and ended up with this "budget version." Overall, though, it works well enough. I think I might install it somewhere after I repair it.

These units are clearly used, and it's also obvious that the Chinese seller tried to repair this one by replacing the high-voltage transistor, which has nothing to do with the missing QC functionality. In any case, I've ordered the chips, and once they arrive I'll check whether my diagnosis was right.

That's all from me. I hope this review was at least interesting, if not useful. In my opinion, you can find some interesting, inexpensive things on Taobao, but you don't always get lucky with the quality.


This review was sponsored by the yoybuy.com forwarding agent, which covered the shipping costs. Affiliate link to register; newcomers get a 10-off-50 coupon.
The power supply cost $3.4 including shipping to the forwarding agent. Shipping from the agent depends on various factors. The power supply weighs 123 grams, according to the order page on the agent's site.