/       /       /    Low-Profile S-120-24 Power Supply, 24 V, 120 W
Support the project on Patreon


Low-Profile S-120-24 Power Supply, 24 V, 120 W

$13.89
Go to store
The second power supply Banggood sent me for review—and it’s more powerful than the previous one. The only thing they probably have in common is that they’re both low-profile power supplies. As always, the review goes into more detail, with an inspection, teardown, tests, and, of course, conclusions..


Low-profile power supplies aren’t used all that often, which is probably why reviews of them are much less common. Partly because they’re less common to begin with, and partly because they often cost more than regular models. Since I had a chance to get one for review, I decided to take it apart and tell you what it’s like.

It comes in a cardboard box, just like the “grown-up” models.


It looks neat enough, and there’s a label listing the main specifications.


But wait... I ordered a 12 V, 10 A power supply, not a 24 V, 5 A one. Oh well, the great Chinese wall roulette...
There’s no getting around it: we’ll test what they sent.


And here’s the “low-profile” part: the power supply is just 19 mm tall. For comparison, a matchbox is 12 mm tall.
That’s made it longer, though: 225 mm. The width is typical for power supplies like this, at 53 mm.


There’s a terminal block for the mains input on one end, and an output split across four terminals on the other.


The case is aluminum, with plenty of ventilation holes on top and stamped recesses underneath for components screwed to the case.


The cover comes off very easily: just pry it up with something and pull. It’s held in place by the shape of the extrusion itself, but it won’t come off on its own. I couldn’t even slide it out of place.


Inside, everything looks... well, rather sad, to put it politely.


There’s no input filter—not even Y capacitors connected to the ground terminal. There are two 47 µF input filter capacitors. This type of capacitor is used in modern ultra-thin TVs and monitors.


The output side isn’t much better: four no-name 470 µF, 35 V capacitors, and no output filter.


The transformer, on the other hand, looks nice. It’s wound with either litz wire or, at the very least, a bundle of lots of thin wires—which is exactly how it should be done.


The circuit board is also fairly easy to remove; you just need to undo the two screws holding the high-voltage transistor and diode assemblies.


The high-voltage transistor is a 20N60 (20 A, 600 V, 0.15 Ω). There are two SF1620 diode assemblies, rated at an estimated 16 A and 200 V.
There’s also the input bridge rectifier and the feedback optocoupler.


There’s a lot of unused space on the board. I think this power supply could have been made considerably shorter if they’d wanted to.
But take a look at the middle terminal on the left-hand block: it isn’t connected to the interference-suppression capacitors—or to anything else at all. It’s just left floating. The idea must have been to use some kind of screw to connect it to the case, but there’s nowhere to screw one in here.


While inspecting the board, I noticed some other separate little wires inside the transformer, running to the output terminals.
Speaking of the transformer: it has a six-pin bobbin, but only five pins are actually used. One of the wires simply comes out of the transformer and is soldered separately.


1. High-voltage section. The PWM controller is an OB2269. There’s also a circuit here for turning off the MOSFET faster, visible just above the controller. It’s essentially a simple driver: the PWM controller charges the gate capacitance, and an additional transistor discharges it.
2. The output section is a classic design, with an adjustable TL431 shunt regulator.

The assembly and soldering quality are poor. Some leads are badly soldered, and there’s at least one “narrow” spot where part of the solder was cut away to prevent a short circuit.


Of course, I traced the circuit, and you can see that they cut corners on just about everything: filtering, thermal protection. There aren’t even any RC snubber networks across the output diode assemblies.


Input and output capacitor capacitance. At 1 kHz, they measure 77 and 1546 µF (39 and 386 µF each). A regular multimeter shows slightly higher values, and in this configuration the output capacitors are actually very close to their stated capacitance.


But enough of the inspection; let's move on to the main event—the tests.
As usual, I used an electronic load and a multimeter. With no load, the output voltage is 24.175 V. The first startup took quite a while, but after that everything went normally.


This time I didn't load the power supply up to the protection trip point. Instead, I stopped at 150 W, despite the rated 120 W, and the power supply handled it without any problems. According to the electronic load, the initial voltage was 24.17 V, and at 6.25 A it was 24.09 V—very good indeed.


Power consumption at loads of 0, 30, 60, 90, 120, and 150 W.


When I converted the measured values into efficiency, it turned out things weren't so bad after all: the result was mostly around 86–88%.


As for heating, for some reason I wasn't particularly worried. The case is large and made of aluminum, the output current is only 5 A, so the power dissipated by the diode assemblies shouldn't be high.
After 20 minutes at 33% load, even the hottest components were at 54°C, despite the room being quite warm.
The transformer and input thermistor got the hottest. For various reasons, the output is on the left and the input on the right in the photo. I tried to make it a "long" photo :)


Another 20 minutes at 66% load. The temperature went up, but nothing overheated: the transformer was at 74°C and the thermistor at 77°C.


After 20 minutes at 100% load, the transformer had heated up to 97°C—quite a lot; I expected less. On the high-voltage side, the hottest component was again the thermistor, meaning the other components were cooler, so everything was fine.


But when I tried to continue the test at 150 W, I hit a snag: after 5 minutes, the power supply shut down. It wasn't the thermal protection (which it doesn't have) but the overload protection. The transformer started saturating due to overheating, which caused the primary-side current to rise, and then the protection shut the power supply down. But 1–2 seconds later, the power supply restarted. If this had been a regular load, it would have kept cycling on and off, which isn't very good either.

Component temperatures after the specified run times at loads of 40, 80, 120, and 150 W.


It also turned out that when I set the load to 150 W, my USB hub started disconnecting. It would only start working again after I reduced the load on the power supply and reconnected the hub. But that's not all: with the power supply overloaded, my computer speakers started buzzing loudly, even though the load was connected through galvanic isolation. In other words, the USB cable was acting as an antenna.

The overall picture after the test at 120 and 150 W.


Voltage stability, on the other hand, is excellent. The measurement on the left is with the power supply cold; the one on the right is after it had heated up during the previous test.


Now let's see why my USB hub might have been disconnecting, and what was going on at the output in general. The load is fully galvanically isolated from both the mains and the computer.


Yes, of course, it's not pretty. The oscilloscope traces correspond to loads of 0, 30, 60, 90, 120, and 150 W. At 120 W, the output ripple exceeded 800 mV p-p, and at 150 W, resonant effects started to appear.


If we slow down the timebase, we can see what's happening at 100 Hz too: the total ripple amplitude is 1.2 V at 120 W and 1.6 V at 150 W.


Just out of curiosity, I decided to run a small experiment. I connected a high-quality Panasonic 820 µF 35 V capacitor in parallel with the installed capacitors. I was curious whether anything at all would change.


The traces at 60, 120, and 150 W; the added capacitor is shown on the right. In practice, the ripple amplitude didn’t increase, but resonance also started to occur at 120 W. Interestingly, though, the USB hub stopped disconnecting.


Conclusions.
In this case, they’re not encouraging. The build quality doesn’t just leave a lot to be desired—it falls below even that standard. I found several poor solder joints on the board that could come loose over time. The components are unbranded and unsecured, there are no filters, and the overall noise level is high. And then there’s the “lottery”: you might get a 12 V, 10 A unit or a 24 V, 5 A one.
There is some good news, though: the power supply delivers its rated 120 W without any problems, while maintaining excellent output voltage stability, and its temperature stays within reasonable limits.

Recommendations.
If you just want to buy a power supply and use it, this isn’t the one for you. But if you’re looking for a power supply to tinker with, it can certainly be improved. Here’s what you need to do:
1. Reflow all suspicious solder joints.
2. Install the missing components in the primary circuit.
3. Add a common-mode choke at the input. There’s no room for one, though, so you’ll have to improvise.
4. Replace the output capacitors with higher-quality ones.
5. Add an output inductor.
6. Add an RC snubber across the output diodes.
7. And if you put a piece of thermally conductive rubber under the transformer, you can probably “overclock” it to 150 W.

Whether it’s worth the effort is another question. That’s all from me; I hope you found this useful.
$13.89
Go to store
This page was found when searching for:
схема китайского импульсного бп s120-24, схема блока питания 120 вт 24в, руководство по использованию блока питания s-120-24, s-75-24 электрическая схема, s 120 f 24 схема, power supply model: s-120-24 схема, s-120-12 ремонт, 11, ysd-12300 блок питания схема, s-120ctp-pcbv03, https://www.kirich.blog/obzory/bloki-pitaniya/780-nizkoprofilnyy-blok-pitaniya-s-120-24-24-volta-120-vatt, s-120-12 схема, схема бп на 5в модель с120, power supply s-120-24 схема, oreol s-75-12 схема, nes-200 схема принципиальная, схема ws-s150-24, ts6 v14, схема блока питания mn-240-12, блок питания 03-46 apeyron схема, перечень радиокомпонентов в блоке питания светодиодной ленты model;s120-24v-5a, как из импульсного бп model;s-120-24 24v 5a сделать регулируемый блок питания, s-75-24 схема, crp vn120-12 схема, s-120-24 схема

Questions and answers

How much power can the S-120-24 power supply handle?

It ran without problems at its rated 120 W, with stable output voltage. At 150 W, it shut down after five minutes because the transformer overheated and the overload protection tripped.

What is the efficiency of the S-120-24?

Efficiency was about 86–88% in most measurements.

Is the S-120-24 suitable for everyday use?

The author does not recommend it for someone who simply needs a power supply to use: the review found poor soldering, unbranded components, and no filters. It may suit someone interested in tinkering and improvements.

The figures from this review are in the measurement database: Power supplies

Calculate manuallyAll calculators

Discounted products


You may be interested in


Discounted products


Support the project

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!

Discuss on the forumHave a question or some experience to share? Discuss the article with the author and readers: the forum is a convenient place to share schematics and photos.
Go to the forum

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