Power Module for a 30 V, 10 A Linear Power Supply

Recently, I reviewed a kit for building an adjustable linear power supply with an output voltage of up to 60 V and current of up to 20 A (at least in theory). Today I’m reviewing a product from the same seller and the same series, but this time it’s not a kit, just a separate module that can also be used with the control board from the previous kit.
I’ll go into more detail in the review, and I’ll also include the schematic for the relay module from the previous review.


First, why did I buy this module at all, when last time I showed a perfectly functional kit with more interesting specs?
Actually, it’s very simple:
1. When buying from one seller, it’s a little cheaper to buy several items.
2. I’m planning to build not one, but two power supplies, with different specs and controls.

But let’s get to the board.
The packaging is very basic: a bit of soft foam and a clear plastic bag. The latter was added by the forwarding agent, not the seller.


The board is fairly hefty, weighing about 260 g. I haven’t seen this board on any other marketplaces, and probably most of the products from this little shop on Taobao aren’t listed elsewhere either.



The board measures 100 × 82 mm, or 100 × 104 mm including the protruding parts. It’s 60 mm tall, measured to the tops of the input capacitors; without them, it’s about 35 mm tall.


The package couldn’t be simpler: the board and a cable for connecting it to the control board.
I should probably add a small caveat here. In my previous review, I wrote that the kit included plenty of mica pads and mounting screws. While writing this review, I realized the seller had probably put the parts for both orders into one bag. Even so, there was more than enough.


Unfortunately, the board was damaged in transit. I don’t know when it happened—while it was being shipped across China to the forwarding agent, or on its way to me by mail—but given the quality of the packaging, I suspect the former.
The power transistors and the bridge rectifier had been bent out of place. It wasn’t difficult to bend them back, but it still made me a little nervous.


One of the input capacitors was also scratched, but here I’m more inclined to blame the seller. I’ll explain why a little later.


As with the previous board, this one is designed to be mounted directly on a heatsink, as shown in the product listing photos. That’s quite convenient, but I’d add standoffs near the transistors too. Getting the height right would be noticeably trickier, though. I’m used to components not bearing any load—in other words, not serving as mounting points.


Here, the standoff height and placement are chosen so that the board is supported partly by the standoffs and partly by the components.


The power wires connect via the same type of terminal blocks as on the previous kit. The connector for the control cable and relay power is visible to the left.


There’s a fuse on the board, and, just like last time, it’s rated for too low a current, which may limit the output current. I’ll probably swap in the fuse from the previous kit and buy a higher-current one for that board.
All the terminal block contacts are labeled, though some of the markings are hard to read. In practice, you only need to figure them out once.


A pair of rectifier filter capacitors, rated at 12,000 µF each, from a fairly well-known manufacturer.
The relays are the same type as in the previous kit, but with one difference: the coils here are rated for 24 V, while the previous ones were 12 V. Their current rating is a much better fit here, though: the output current is specified as 10 A, not 20 A.


Now, back to those scratched capacitors. Just out of curiosity, I measured their capacitance. The meter read 18,760 µF, which works out to 9,380 µF each instead of the specified 12,000 µF. Whether the capacitors are used or simply counterfeit is anyone’s guess, but the datasheet says that:
1. The LA5 series has no 63 V capacitors measuring 30 × 50 mm.
2. The series does include a 12,000 µF, 63 V capacitor, but it measures 35 × 45 mm.


Since this board’s output current is half that of the previous one, it has just three TIP35C transistors installed, exactly the same as on the previous board. But it does have a KBJ2510 bridge rectifier, rated for up to 25 A continuously with a heatsink and 350 A peak.
Essentially, this board is almost a “half” of the previous one, with a combined transistor current rating of up to 75 A continuously or 120 A for short periods. But keep in mind that these current surges are quite realistic in an adjustable power supply.


As before, the transistors are driven by another transistor, in this case a TIP41C, mounted on a separate heatsink.
The power transistor emitter circuits also have 0.1 Ω resistors, but the driver circuit uses a higher value: 2.2 kΩ.


You saw a pair of relays earlier. Since this board is essentially a hybrid, it also has a comparator to control them. There’s a conspicuously empty spot to the left of the white connector. Judging by the contact layout, a fan was supposed to connect there, but the connector itself is missing. I think you could easily solder one on and install a 12 V fan. It just won’t be speed-controlled, which is a shame.


The underside of the board is a bit dirty, with some dust and flux residue, but the power traces are thoroughly tinned, so no complaints there.
As I turned the board over in my hands, I couldn’t shake the feeling that the routing was a bit “rough”—both in the width and shape of the traces and in the overall topology.


Last time, I redrew the schematics for the regulator’s control board and power stage, but didn’t include the relay board schematic. That wasn’t because I forgot; I was already planning to review this board at that point and wanted to “kill two birds with one stone.”
The thing is, this board is a simplified version of an entire kit, and it also includes a winding-switching circuit and a comparator.


First, as promised, here’s the schematic for the winding-switching board from the previous review. You can use this board on its own to upgrade a linear power supply. Choose the number of channels to suit your needs.


And this is the relay board from that review. You can see that the two boards are quite similar in many ways, but they also differ.
The difference is that the board under review has no voltage regulator and doesn’t need another transformer winding. The comparator and relays are powered from the control board, using both 12 V rails at once (positive and negative).
The output voltage monitoring circuit is also implemented somewhat differently. Essentially, the comparator circuit is turned “inside out” because it has no separate power supply.
I also figured out how the hysteresis is implemented. It turns out that resistor R6 in this schematic and R1 in the previous one are responsible. The logic is simple: the more relays are switched on, the more current they draw, and therefore the more voltage drops across the resistor. But on the board under review, this shifts the comparator’s measured voltage slightly, while on the previous board it shifts the reference voltage supplied by the trimmers (which are also powered from downstream of the resistor).



So, how do we test the board...?
It’s simple. Last time I called the kit a DIY project, and that’s essentially what it is. You can use this board together with the control board; all you need to do is replace the connector that links the control board to the power board.
The board now needs 12 V, -12 V, ground, and a control signal, and all of these are brought out in exactly the required order.
The seller also offers the control board separately for about $11.5. There’s a link in the previous review, which takes you to the page where everything is sold separately.


For now, I’m mounting the board on the same heatsink used in the previous review. I had to put a piece of plastic underneath because the board turned out to be longer than the heatsink. But in the end, I still managed to short the 36 V AC input to the DC output (I hadn’t insulated the transistors from the heatsinks). No big deal—the stress test went fine :)



From here on, everything is very straightforward—even much easier than last time.
First, connect the transformer leads. The board labels everything: 0/12/24/36 V.


Then connect the + and - outputs of the power board to the inputs of the control board.


Connect the control and power cable from the power board to the control board. Everything is straightforward here too: if the connector has been replaced correctly (as shown in the photo above), just connect the five-wire cable, and that's it!
Power is already applied in the photo.


I couldn't resist and connected a load after all, but so far I haven't been able to get more than 12.


And without adjustment, there's no way to get more than 12 V. I'll describe the process, especially since it's exactly the same as adjusting the relay board from the previous review, only that one has more steps.
1. Connect a tester and set the output to about 8 V.
2. Find the pair of trimmer resistors on the board. The upper one is marked PR1, the lower one PR2. If relay 1 is off, turn the trimmer counterclockwise; if it's on, turn it clockwise. Turn it until the relay changes state, then set the trimmer so the relay is on. The relay housing is transparent, so you can see it.
3. Since the first relay has connected an additional winding, you can set a higher voltage. Set it to 21 V and repeat step 2, this time using the second trimmer and the second relay.
In the second case, my relay switched on too early, and in the first, too late, so I adjusted them as described above.

On the board from the previous review, adjust resistors 3 and 4 while monitoring the corresponding relays; the thresholds are 35 and 48 V.

There was a surprise, too: if you try to raise the output voltage even further, the relays start to "go crazy". This happens at 38–40 V, but in any case, the spec was up to 30 V, so no complaints there.


While adjusting it, I noticed a faint smell of something hot. I poked around with a stick thermal camera and found two resistors heating up: one in parallel with the input capacitor (on the right in the first photo), and one in parallel with the output (on the left).
I also checked the temperature of the control board; almost everything here is cool (photo on the right).


Last time, we discussed what the current would be before the bridge rectifier and what it would be after. But as they say, I'm a simple guy: if I have questions, I take measurements :)
Actually, I just don't like simulators much and prefer "hardware solutions", so I took two multimeters and measured the current before the bridge rectifier (AC) and after it (DC) at the same time. Just to be safe, I used a UT 181A before the bridge rectifier, which I trust a little more than its "colleague" the UT 61E when it comes to True RMS accuracy.


So, I varied the load current from 1 to 6 A in 1 A increments. The difference between the current readings was:
1.93
1.72
1.66
1.60
1.57
1.53

So the difference decreases as the load current increases. I think that at the maximum values it will be closer to 1.42–1.45, so for calculating the maximum current, you can simply multiply the output current by about 1.5.


I wasn't particularly surprised by the results of this test, and I only did it because an electronic load was sitting on the table next to me.
I loaded the assembly to 6.3 A, with the output voltage at about 20 V—close to the upper limit, but without the additional winding connected. The output is nice and clean.
I'm planning to run pulsed load tests for the next review, but perhaps someone can suggest the best way to do that and under what conditions.


Video version of the review


The inspection and brief test are complete; time to sum things up.
As last time, the fuse has clearly been sized incorrectly, but the relay situation is a little simpler: its contacts are rated for 16 A AC, which is much closer to the 10 A DC output rating.
The capacitors are most likely counterfeit, although their specs are quite good. The output stage has been scaled down proportionally compared with the previous board, using 3 transistors instead of 7, but the power dissipation is lower here too.

Otherwise, it's a pretty decent little board. Basically, all you need is a control board, and you've got a complete set of electronics for a fairly powerful power supply. The seller has various boards available, so I recommend checking out the store page. I'm thinking of buying something else from him myself; it's a shame shipping things like heatsinks would be expensive :(

That's all for now. As usual, I'd be glad to hear your comments and questions. I still have a review of a controller for digital control coming up, which I plan to use instead of the control board.

I placed the order through the agent yoybuy.com. The link is a referral link; you get a 10-off-50 coupon, and I might get some kind of bonus too :)
The board costs about 25 dollars including shipping, but the final price depends heavily on various factors.