Sanmim SM-PLG06A power supplies: when power isn't the main thing
Inspection, tests, schematic, conclusions.
I won't list all the Sanmim power supplies I've reviewed before, but I can say that while they're no Mean Well, they're still interesting and fairly well made. Unfortunately, it's hard to call them inexpensive.
This unit belongs to the compact PLG06A series, which includes six models with different output voltages and currents.
The following details are taken from the datasheet.

I ordered a lot of five. It works out cheaper that way; buying them individually is quite a bit more expensive.
They came simply packed in zip-lock bags. As it turned out, one of the five wasn't in the usual antistatic bag, but in a regular one.

Also, the one that came in a regular bag has a few minor differences: there's no marking on the input capacitor, and the board marking is different. The board's development date is presumably different too—older than the other four.

Specifications. Sorry for the long image; that's just how it turned out.
There are six models with output voltages of 3.3, 5, 9, 12, 15, and 24 V. This review covers the 15 V version.
The 3.3 V and 5 V models have output power ratings of 4 and 5 W, respectively. The other models are rated at 6 W, as reflected in the series name.

It looks neat, though Sanmim products rarely give me anything to complain about in that regard.

The board is compact, about one and a half times smaller than a matchbox.

The full dimensions are in the datasheet, so there's no point measuring them.

1. The input has a thermistor and a varistor, but since it's rated for 560 V, I honestly don't see much point in it.
2. The input capacitor is 6.8 µF. That's fine for our mains supply, but it may be too small for operation at reduced voltage.
3. There's also a common-mode choke and a 1 nF Y capacitor.
4. On the output, they installed a pair of 470 µF, 25 V capacitors and an inductor to reduce ripple.

The power supply is built around the AP8012H PWM controller. This controller also incorporates a 700 V high-voltage transistor and is designed for power supplies rated from 5 W (in this case) up to 13 W.

The board is double-sided, with components mounted on both sides. I wouldn't call every part of the layout pretty, but overall it's decent.

I sketched out the power supply schematic, though there's nothing unusual about it.

During the inspection, I noticed the “fuse” is an SMD jumper. Interestingly, according to the board markings, it's on the neutral line rather than the live line. Obviously, in this case, hardly anyone will bother to observe polarity, but still.

I was about to complain—and quite rightly so. Using a jumper, especially such a tiny one, in the input circuit is completely wrong. But then I found an explanation in the datasheet.
The manufacturer recommends installing a separate fuse at the power supply input, and also provides an example of adding components on the output. Several capacitors and a TVS diode are recommended.

There’s also a more comprehensive version with an added input filter. In any case, you’ll need to install at least a fuse yourself.

Now let’s move on to the practical testing.
The power supply started almost instantly when connected, with exactly 15 V at the output. Its no-load power consumption was so low that my wattmeter didn’t even register it.

It’s a low-power unit, so testing it wasn’t very convenient. I ended up using an EBC-A10H electronic load with a four-wire connection.

I ran the load test in several sessions because the power supply stubbornly refused to shut down. But even after I increased the load current further, it turned out that above 800 mA the PSU simply entered current limiting instead of shutting off the output.
Still, you can already see that the PSU regulates the voltage very well.

Measuring efficiency was difficult because the wattmeter’s measurement error is so large at these power levels, but I gave it a try anyway.
I tested the PSU at 200, 400 and 600 mA, corresponding to 3, 6 and 9 W. Input power was 4.0, 8.3 and 12.0 W, respectively, giving an estimated efficiency of about 75%.

The high-frequency ripple looked very good. There’s nothing to show with no load, so here are the results at 200, 400, 600 and 800 mA. Even under overload, the ripple amplitude stays below 10 mV.

At a slower timebase, though, things looked worse. Noticeable ripple appears at 600 mA, and it gets quite severe at 800 mA. That’s also why the high-frequency waveform was “jumping” during the test; you can see it shifted in the screenshot above. But I had no complaints at load currents up to 400–500 mA, so it’s all good.
Interestingly, the ripple frequency was 200 Hz, not the usual 100 Hz you get after a bridge rectifier.

The manual includes graphs showing how the load power depends on the input voltage (AC and DC) and temperature. They show that the power needs to be derated once the ambient temperature exceeds 55 degrees.

Of course, I was interested in testing it not only at its rated currents, but at higher ones too.
Testing was done at normal mains voltage and room temperature, in 20-minute stages, with loads of 200, 400, 600 and 700 mA. The 400 mA stage ran longer than planned by accident, but that was actually a good thing.
After I increased the load current to 700 mA, the PSU shut down after about 5 minutes, so it does have thermal protection, and it works.
The output voltage also dropped slightly right at the end, by about 0.1 V.

I didn’t take spot temperature measurements; I decided to stick with thermal imaging.
The hottest component was the PWM controller:
1. At 200 mA, it reached 63 degrees
2. At 400 mA, it reached 88
3. At 600 mA, it was already at 126 degrees, but everything was still working normally
4. At 700 mA, I managed to catch the PSU shutting down: the case temperature reached 178 degrees, and a few seconds later the PSU switched off the output for a while.

While checking the temperature rise, I also monitored the output voltage. The difference between the “cold” and overheated states was just 14 mV.

Now for the verdict.
Yes, this power supply doesn’t have much output power, which made it less convenient to test than more powerful units. But it’s reasonably well built and can run continuously at both its rated 6 W and 9 W, which is a 1.5× overload. It can handle even more for short periods.
I was also impressed by the very low level of high-frequency output ripple. Under overload, though, low-frequency ripple starts to appear; at rated load, everything is fine.
My only significant complaint is the lack of a fuse. I think they could have included one.
In short, I liked these little power supplies, though I wish they cost a bit less.
By the way, the store provided a coupon, BG349d53, which brings the price down to $13.99 through October 10.
That’s all for today. I hope you found this useful, and as always, I’d be glad to hear your comments.
https://www.kirich.blog/obzory/bloki-pitaniya/1045-bloki-pitaniya-sanmim-sm-plg06a-kogda-moschnost-ne-glavnoe, sanmim, блоки питания sanmim, plg06a, 2, sm plg06a 5v 1a, sm plg06a, sm plg06a схема, подключение блока питания sm plg06a sanmim на сколько вольт и какое питание
Questions and answers
How much power can the Sanmim SM-PLG06A 15 V supply handle?
It ran normally at its rated 6 W and sustained 9 W for an extended test. At 700 mA, it shut down after about five minutes as it heated up.
What is the output ripple on the Sanmim SM-PLG06A?
High-frequency ripple stayed at or below 10 mV, even under overload. Strong low-frequency ripple became visible at 600–800 mA.
Does the Sanmim SM-PLG06A have an input fuse?
No. The board uses an SMD jumper, and the manufacturer’s documentation recommends adding a separate fuse at the power-supply input.
This model is included in the summary table: Comparison of power supplies: ripple, efficiency, and temperature rise
The figures from this review are in the measurement database: Power supplies
Discounted products
You may be interested in
Discounted products
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!
Only 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



