12V-10A board (claimed)

Good afternoon, everyone.
I put together a short review of a “new” used 12V-10A board (as advertised). First, apologies for any inaccuracies or mistakes. I’ve never written a review before.
I often browse Chinese sites for power supplies and the like. Kirich got me interested in power supplies—someone at home showed me a site with one of his reviews on it and said they’d register me there.
In the spring, I spotted a new board (12V/10A) and decided to get one to try.
The board showed up on Banggood and AliExpress (with the same picture). The lowest price was on Banggood: 423 RUB, about $7.
The original price was 498 RUB. I tried a few of their coupons, and one of them worked (#d7e979). I’d discovered used power supply boards a long time ago, bought a couple, and liked them. I tipped off some electricians. Later, after Kirich, I got interested in them myself.
The board took a little over 2 months to arrive in Western Siberia, with a postmark from Mongolia (I bought it without tracking). It was packed in a couple of layers of thin foam. But since the ends weren’t taped down, it arrived separated from the foam.
Dimensions: 102 mm (L) × 62 mm (W) × 32 mm (H). The board has the manufacturer’s marking (SL-SangHai-Power) and two dates: 080711 (design date?) and 2012.2.1 (manufacturing date?). The larger components on the board are covered in clear adhesive.
As for the components, all the capacitors are branded (JAMICON). The input capacitor measures 120 µF. On the output (35V), there are two 1200 µF capacitors (d13), then an inductor, followed by a 470 µF capacitor (d10).
The input has a proper filter: X (0.47 µF) and Y (471K) capacitors (the Y capacitors connect N and L through a third Y capacitor to the transistor heatsink and the output negative), plus a common-mode choke. There’s a ferrite ring on the input connector wires to suppress interference. The fuse is rated at 2.5 A (usually the rating is higher, 3.15 A). There’s a 10471 varistor. The bridge rectifier is a 408 (with a hole). The only thing missing is a thermistor (there’s a spot for one). The transistor is an MDF10N65B. There’s thermal paste on the transistor and diode assembly. The only component I couldn’t get to was the output diode assembly. To reach it, you have to desolder everything—the transistor heatsink, then remove the transformer, and only then can you get to the diode (the mounting screw is on the transformer side, with no clearance between them).
These components are mounted properly, using special spring washers. The transistor and diode are in plastic packages. The PWM controller is a 16D06 (PMVX).
Ferrite beads are fitted on the output Y capacitors, the transformer winding, and the diode assembly (?). The heatsinks are 3 mm thick.
I really like the look of the board. I couldn’t find any signs of use. All the component holes are plated through. The board is coated with solder mask on both sides—it looks expensive.
The components are factory-soldered; everything is neat and straight. “Chinese hands” haven’t touched anything except one spot—
they soldered their own SMD resistor upside down in parallel with the “current-sense” resistor (0.27 Ω, 2 W). Maybe they wanted it to blend in with the white stripe on the board, or maybe they’re protecting their “intellectual property”—the resistor value.
They do the same thing on used 12V/2A boards, but there they solder in a resistor (for the TL431) to adjust the output voltage.
The resulting resistor value is 1R1 (1.1 Ω). This resistor immediately made me think the Chinese had adjusted the current to 10 A. They did something similar
on used 12V/5A boards (with an NE1109F). The marking also hinted at this: (249497 94 V-0).

This is how I interpreted it (I may be wrong): 249497: 24—24V max. version. (94)—current of either 9 A or 4 A. (97)—power of 97 W.
The SMD components are additionally glued down; the resistors are standard types, except for three that control the output voltage. The output voltage is regulated by a TL431.
While inspecting the board, I noticed a few things I couldn’t make sense of. The trace from the live input runs as usual through the common-mode choke and then to the bridge rectifier.
But the trace from neutral stopped at the common-mode choke. The negative terminal of the output capacitor after the filter inductor was floating (not connected).
I looked at it through a magnifying glass and still couldn’t figure it out. Only later did I notice that the section from the common-mode choke to the bridge rectifier, and the output negative, are on the front side of the board. There’s an unpopulated SMD load resistor (1206) on the output. For safety, slots have been cut in the board. Where there’s no slot, the distance between the hot and cold sides at the narrowest point is 10 mm. All the components are marked differently depending on where they’re located (a P is added on the hot side, and an S on the cold side). I wrote down all the SMD component markings.
Before switching it on for the first time, I always check the input and output for shorts with a multimeter, then power it through a Soviet-era light bulb (this is just to check that the power supply works. The bulb is rated 220V-60W).

I don’t have a decent load. I was going to get a few 100W resistors this winter, but then I saw Kirich’s reviews of his electronic loads. I decided to try making a proper load right away (figure out how and build it), but things have stalled for now. The load will be 60 W. I ran it for about 1.5 hours or more (the readings changed after both 15 and 20 minutes, so I left it running for a long time). The no-load voltage is 12.31V.
The results.
Voltage: 12.14V. Transistor: 65°C (heatsink 62°C). Diode assembly: 72°C (heatsink 69°C). Bridge rectifier: 59°C. Transformer: 67°C.
The capacitor by the heatsink (in the corner) reached 69°C; the others were cooler. I’m not sure about the transistor temperature—it seemed higher. I’ll check again this weekend if I can. The output voltage dropped very slowly as the board and resistors warmed up.
I couldn’t make a load out of resistors (they would have burned out), but I did check the current and protection.
With the Chinese SMD resistor soldered in (1R1—1.1 Ω), I cranked it past 11 A. The protection didn’t trip.
I desoldered the SMD resistor (restoring the factory configuration). At 9.3–9.4 A, the protection trips and stays on until you reduce the current.
The voltage stays at 12V up to 8 A. Then it drops sharply; at 9 A, it’s 9.4V.
But with the Chinese SMD resistor (1R1—1.1 Ω), the voltage doesn’t drop until around 11 A. (at 10 A — over 12.0 V). So that’s how the board is configured; it’s not a lack of power from the power supply.
It turns out that with good airflow, you can adjust the power supply’s behavior using the “CURRENT” resistor (by reducing its value).
I checked the power supply with an oscilloscope under 30 W and 60 W loads. 20 mV per division. Using the factory probes (Hantek DSO5102P) and some “improper” probes: just two wires (with a ceramic and an electrolytic capacitor, same values as Kirich’s, plus copper alligator clips).
I also checked a used 12 V/5 A board (30 W).

My conclusion: I like this board. So far, in terms of quality (and class), it’s the best used power supply board I’ve seen. Name-brand components. In the same class as Kirich’s popular power supply.
The relative downsides are that the output diode assembly is hard to access. The transistor is also difficult to reach without the right tool (to fit the ring and washer and hold the nut), so it’s better to desolder it together with the heatsink. No thermistor. The price, perhaps.
I don’t understand why the Chinese sellers resort to this resistor trick. Just sell it as originally designed: 12 V, 8–9 A. Why insist on tweaking it to 10 A?
Chinese SMD resistor.
Comparison with the 12 V-5 A board
SMD components.
Power supply input
Power supply output
Improper probes
Powering up the power supply through a light bulb — CLICK.
Oscilloscope test — CLICK.
I still don’t know how to draw schematics.
Good luck, everyone.