A pair of 60 W power supplies rated for 12 and 24 V

I've reviewed quite a large number of different power supplies by now, but today's review is a little different. No, the power supplies themselves are the same; the difference isn't in them, but in the manufacturer. If I'm not mistaken, this is my first review of Ukrainian-made power supplies.


Of course, familiar devices like power supplies are made by both Ukrainian and Russian manufacturers, but against the backdrop of all the Chinese-made products, they sometimes get lost in the crowd.
So today I'll try to support a domestic manufacturer.

The power supplies are made by a company called Light Route, which was previously unknown to me. As it turns out, the company has been around for about 10 years, making LED lights, and has also been producing power supplies for the last 6 years.

The link in the title goes to the manufacturer's website, but their products are also available from online stores, for example, at Rozetka.

But let's get back to the subject of the review.
They came in a plain cardboard box, although the packing tape had the company's logo and name on it.


I ordered two power supplies. They have the same form factor and power rating, but differ in output voltage and, accordingly, current.


The power supplies are designed for a “narrow” input voltage range of 175–265 V. The maximum case temperature is 80 degrees, or 50 degrees relative humidity.
The inputs and outputs are marked, and the stated protection rating is IP66.
The first power supply has a 12 V output at up to 5 A; the second is rated for 24 V and up to 2.5 A.


Since the power supplies differ essentially only in output voltage, I'll compare them in this review, starting with their weight.
There isn't really any point in weighing them; I did it out of curiosity. It turned out that the 12 V model is slightly heavier.


1. Wires are provided for connecting the power supply and load: white wires for the input, and red and black wires for the output. The wires are 17 cm long and have soft silicone insulation, but the conductors themselves are quite stiff. At first I even thought they were steel, but they're not magnetic.
2. There are traces of sealant around the points where the wires enter the case.
3. The bottom of the case has a plastic cover secured with clips.
4. I assume the covers were fitted before the sealant had fully cured, so a little got onto them too.


Some components are visible inside. For example, you can already see three input capacitors—I'd guess they're probably 22 µF, 400 V, or possibly 33 µF, though the latter seems unlikely—and four output capacitors. The transformer and a long heatsink are also clearly visible.
Since the two power supplies have different amounts of potting compound, the input EMI filter choke is still visible on the left-hand unit.

According to the IP66 protection rating, the power supply is rated for:
6 — Dust cannot enter the device. Complete protection against contact.
6 — Protected against powerful water jets. Water projected against the enclosure in powerful jets from any direction must not have any harmful effects.

Broadly speaking, I agree. Judging by their appearance, the power supplies should pass the test for this protection rating. I do have one small suggestion, though: safety could be improved by using a little more sealant to cover the input capacitors better. The layer there is a bit thin, after all.


At this stage I didn't take anything apart—and taking apart potted power supplies is no fun anyway. But the manufacturer's website has photos of other power supplies, which look like 25, 40, and 60 W models, though they're a slightly different type.
You can see the input filters, a fairly large transformer, a fuse, and apparently even a varistor on the circuit boards.
If the units under review are built along similar lines, that's pretty good.


I'll start the tests by measuring the output voltage and the power consumption with no load.
The output voltage on both power supplies is fixed, which makes sense for this design. And judging by their low power consumption, they have a “green” mode and no output load resistors.

No complaints here: the voltage is stable and power consumption is low.


For the remaining tests I used an EBC-A10H electronic load, connected using four-wire sensing, but on the load side of the wires. This should have some effect on how the output voltage varies with load.


The 12 V power supply did very well in the load test. When cold, it delivered 6 A before the protection kicked in. The output voltage varied by just 90 mV across the entire range.


The 24 V model, however, doesn't look quite as good. No issues with regulation: the difference was the same 90 mV (strangely enough), and the protection worked correctly. But the maximum I got was 2.6 A, against the rated continuous 2.5 A—really cutting it close.


For the efficiency measurement, I decided it would be more convenient to combine the test results for easier comparison.
Unfortunately, this test has the highest measurement error, since it uses two instruments, each measuring two parameters. Also, the wattmeter is less accurate at such low loads.

As expected, the 24 V power supply was more efficient, reaching 90%, although the 12 V model wasn't far behind. The difference is because the 12 V power supply has twice the output current, resulting in greater losses in the output diode and wires.


Measuring output ripple is a must when testing power supplies. If you're using one to power LED strips, it's not that critical, but someone might decide to use it with a more sensitive load.
Measurements were taken at the ends of the output wires. Two capacitors were connected in parallel with the probe: a 1 µF electrolytic and a 0.1 µF ceramic, as shown in the photo from the Power Integrations instructions. Without the capacitors, the amplitude of the main component was lower, but there were "spikes".


Tests at load currents of 0, 2, 4, and 5.8 A, with a sweep rate of 5 µs per division, and at 3 and 5.8 A with a sweep rate of 10 ms.
What can I say? 15 mV is pretty good—I'd even say very good.


The second power supply was tested at currents of 0, 1, 2, and 2.5 A at high frequency, and 1.3 and 2.5 A at low frequency.
Here, ripple reached 20 mV, but keep in mind that the output voltage is twice as high, so the percentage is a little better.


An interesting observation: above I wrote that the 12 V power supply can deliver up to 6 A, but measuring ripple showed that its amplitude increases sharply at that point. The left image shows ripple at 5.8 A, the right at 5.9–6 A.


And, of course, a thermal test.
The test followed my usual routine: several 20-minute stages, with the load current set to 2, 4, and 5.5 A.
Near the end of the test, the power supply shut itself off. As it happened, I took a screenshot, then took a couple of thermal images without turning off the load, and when I looked at the screen again, I saw that the power supply had shut down.


While the power supply was still hot, I ran a load test. It showed that the maximum current of the warmed-up unit was no longer 6 A, but 5.4–5.4 A. It seems the thermal protection works a little differently than usual: in effect, it lowers the overcurrent protection threshold as the temperature rises.

At the same time, the output voltage didn't "drift" at all, which points to decent-quality components.


And in my opinion, the power supply's case temperature wasn't particularly high either, at around 71–72 degrees.
The first photo was taken after 40 minutes of testing; the second and third after an hour.


The 24 V power supply test turned out a little differently. For technical reasons, the first stage, at a load current of 1 A, lasted 40 minutes. Then came a 20-minute stage at 2 A, but when I tried setting it to 2.5 A, the power supply predictably shut down. Predictably, because after looking at the previous one I already understood how the protection works.


The power supply itself wasn't hot, just 52 degrees. The first photo was taken after 40 minutes of testing at 1 A, the second and third after another 20 minutes at 2 A.


A subsequent load test showed that it couldn't deliver more than 2.4 A in this mode, so I set it to 2.4 A and continued the test.


After 20 minutes at 2.4 A, the power supply shut down again. This was also entirely predictable: the overcurrent protection threshold had dropped even lower.


And since this model is a little more efficient, its case temperature was lower too, around 60 degrees (the thermal image was taken a couple of minutes after shutdown).


At this point I could have called it a day, but I got curious about what was hidden under the potting compound. I ran an opener blade around the perimeter and pulled the innards out.
Everything came out cleanly. The compound has a characteristic rubber smell. I also found a cavity it hadn't filled, but since it's deep inside, that's not too critical.
The wire connections are properly sealed, and the wires clearly extend deep inside, so moisture has less chance of getting in.


But as you can imagine, I didn't stop there. In the end, I cut away almost all of the potting compound. It came off relatively easily, sometimes in whole sheets, especially from the underside of the board, though in some places I had to pick it out.


I didn't remove all of the compound; it would take a long time, and there was no point since I'd already seen what I wanted to see.


1. The power supply input has not only a common-mode choke and a fuse, but also a thermistor and, rather unusually, a varistor.
2. There's also an X capacitor, of course, but I was a little surprised to see it on the power-supply side rather than the mains side. In other words, it's intended to suppress interference from the mains reaching the power supply more than the other way around.
3. As I expected, there are three 22 µF, 400 V capacitors in parallel on the input. Their measured capacitance is 70 µF.
4. A P10NK60ZFP transistor in a fully insulated package. Interestingly, the transistor (and the output diode, for that matter) isn't screwed to the heatsink. Apparently they're pressed against it slightly before potting, then the assembly is filled with compound, so most of the heat transfer happens through the compound. A somewhat odd solution; at least, I've never seen this before, and I would probably have done it the conventional way, with a screw. But the manufacturer knows best, so I won't argue.
5. The one sign that this isn't an OEM power supply with a sticker, as "domestic manufacturers" often do, is the marking on the transformer, which includes the company logo and power supply specifications.
6. Four capacitors and an inductor are installed on the output to reduce ripple: three 470 µF, 35 V capacitors before the inductor and one 330 µF, 35 V capacitor after it.



All the other components are on the underside. The soldering is a bit of a mystery: judging by their appearance, some components were placed and soldered by machine, while others show signs of hand soldering, such as some of the resistors and the Zener diode.


The inverter section: you can see both the PWM controller and the current-sense resistors—three 1.5 Ω resistors and one 2 Ω resistor, for a total resistance of 0.4 Ω. The 2 Ω resistor was clearly soldered by hand. If they'd used another 1.5 Ω resistor instead of the 2 Ω one, the total resistance would have been 0.375 Ω, and the protection would trip at 2.66 A instead of 2.5 A. Why they did it this way is a mystery...


The PWM controller is quite well known: it's the NCP1251 from ONsemiconductor.


The only caveat is the overtemperature protection (OTP): the controller itself doesn't have it, and it can be implemented as an option, but in a very awkward way, through the overload and overvoltage protection circuit. But that's beside the point: I couldn't find the thermistor responsible for it on the board. Apparently, the current drop is caused by some other factor, such as the current-sense resistors changing resistance as they heat up.


The output uses the conventional arrangement: an adjustable Zener diode and an optocoupler. I didn't redraw the schematic because I don't see the point; most of it is clear just from looking at the circuit board.


For a sense of scale, here's a comparison photo with the "popular" power supply, which is also used with T12 stations and is noticeably larger.


Conclusions.
Speaking generally about both power supplies, I have no complaints about the build quality, nor any issues with output voltage stability, response to overloads and short circuits, ripple amplitude, or heating.
The 12 V model can actually deliver up to 5.5–6 A, depending on temperature, but the 24 V version failed the load test: after warming up, I couldn't get even 2.4 A continuously, despite its rated 2.5 A. The problem isn't overheating, but incorrectly chosen resistor values in the overload protection circuit.
Otherwise, the components are all fine. Both the input and output capacitors have plenty of headroom; there's an input and output filter, the interwinding capacitor is the correct type, and there's a varistor.

Also, judging by my tests, both power supplies have power headroom, especially the 24 V version, and these limits could easily be increased. This would also help when powering more "demanding" loads that briefly need more current than during normal operation—something that doesn't happen when powering LED strips, which these power supplies were originally designed for. For example, I think I'll use the 24 V power supply for a compact T12 soldering station, where peak power can reach 70 W; as supplied, the power supply would simply trip its protection.

In short, these are perfectly decent power supplies and they even pleasantly surprised me. The circuitry is well designed, with plenty of headroom, but I do have one concern about the protection trip current.

That's all for now. I hope you found this useful.