A decent ZXY6005S DC-DC module, or a DIY bench power supply.
At first I wanted to buy a ready-made replacement, but “that’s not how we do things,” so I decided to build one myself again.
While looking for a suitable DC-DC converter, I sifted through a pretty large pile of different devices on AliExpress. But either the voltage was wrong, or the current, or the way the output was adjusted.
At first I wanted to buy a basic one with LED displays, but its power rating wasn’t enough. Then I considered one without displays, with adjustment via trimmer resistors (I even bought a voltmeter and ammeter for it), but its voltage range wasn’t right. I moved on to a more advanced model, with an LCD and digital controls, and even contacted the seller to ask about the specs (the seller explained everything clearly, by the way, and I almost felt bad for not buying the power supply from him). But none of these was quite right.
Then I spotted the converter reviewed here. I couldn’t resist and ordered it, even though it was well over my budget. It cost me 48.19 (thanks to a seller discount and a $5 coupon).
There have been several reviews of bench power supplies on Maisku, but this one is for those who like making things themselves.
And I did mention in the comments on one of my reviews that I was planning to review a bench power supply, so here it is—I said I’d post it, and I am.
Part one: the converter itself.
The seller took a long time to ship it, dragging things out until the last minute. I’d already messaged him several times, and he kept replying, “Don’t worry, I’ll ship it.” He did. After that, it crawled toward me at a snail’s pace through the depths of China Post, then Ukraine Post. But it made it. And even arrived in one piece.
I was beginning to think it would never arrive.
It arrived in a cardboard box, wrapped inside in a relatively small amount of bubble wrap. The box was taped up in such a way that you could easily pull the converter out, tape the box back up, and send it on its way. If you order from this seller, it’s best to ask for better packaging.
The board arrived with the display installed on metal standoffs; matching standoffs are fitted at the corners of the board.
The power components are mounted on heatsinks. One thing to note: the power inductor isn’t secured in place and is held only by its leads. It would be better to secure it separately.
The back of the board.
The device itself is a digitally controlled converter.
It can be controlled either from the keypad or from a computer. For this, the board has an RS232 connector with 3.3 V logic levels. The settings can be adjusted using either the buttons or the rotary encoder.
The central microcontroller is installed under the display: STM8S105K4T6C
The logic and processor power supply is regulated by an XL7005 switching regulator; the power switch (the main control PWM) is driven by an XL1509
First power-up
This is the smallest board in this converter series.
It provides:
Output voltage: 0–60 V (adjustable up to 62 V)
Output current: 0–5 A (adjustable up to 5.1 A).
Voltage adjustment step: 0.01 V
Current adjustment step: 0.001 A.
Input voltage up to 62 V (I'll explain this below).
Calculates and displays output power.
Calculates and displays the energy delivered to the load in amp-hours (resettable)
Operating timer (resettable)
CC and CV indicators, plus an indicator showing that the board is operating in regulation mode. In other words, if you try to get an output voltage higher than the input can provide, the normal-operation indicator goes out, showing that the board cannot maintain the set parameters.
Decoding the board model numbers
ZXY6005 — 60 V, 5 A
ZXY6010 — 60 V, 10 A
ZXY6020 — 60 V, 20 A
S suffix — computer control; D suffix — has a rotary encoder instead of potentiometers, but no computer control; no suffix — board with trimmers for adjusting voltage and current.
After startup, the display shows the default settings
The menu lets you calibrate the voltage and current readings, set the data transfer speed, and disable the buzzer on the board.
List of all board settings.
(1) SET U-CAL Calibrate voltage measurement.
(2) SET I-CAL Calibrate current measurement.
(3) SET OVP(E1) Set the upper voltage limit
(4) SET OCP(E2) Set the upper current limit
(5) SET OPP(E3) Set the overpower protection threshold
(6) SET OAH(E4) Set the overcharge capacity protection threshold
(7) SET OFFTIME(E5) Set the overtime protection threshold
(8) ---SAVE DATA!--- Save the settings
(9) Start up:OFF Set whether the output is on or off at startup
(10) System Recover Restore the system
(11) Sound Enable:ON Set the sound
(12) Save Parameter: Save the settings
(13) Set Addr. Code Set the address code
(14) Set BaudRates Set the baud rate
(15) Charge Mode:OFF Enable or disable charging mode
There is one minor flaw: the sound setting only applies for the current session, so it isn't saved. Combined with the very loud buzzer, this is pretty annoying. I ended up covering the buzzer with tape, which made things much better.
One more thing: parameters should be saved to a specific memory slot, such as M0, rather than to the general memory that stores calibration, the address, and the baud rate.
The board's firmware lets you set a maximum of 62 V and 5.1 A.
In operating mode, with the output enabled, the screen looks like this—
When a load is connected, the screen shows the output voltage (measured), output current (measured), output power, the amount delivered to the load in mAh or Ah (if the counter goes over 1000), as well as the operating mode and the voltage or current limit.
By the way, when you enable the output (OUT button), the voltage rises gradually, taking about 1.2 sec depending on the set voltage (the higher the voltage, the longer it takes to rise). A very useful feature, in my opinion.
Naturally, I tried comparing the readings on the power supply's meter with those on the connected instrument. Measurements were taken with no load.
At this point, I decided to pause my examination of the board and move on to some hands-on work. Since there are a lot of photos, I’ve hidden this part of the review under a spoiler.
I made a few mistakes while assembling it, both big and small, but let's take things in order.
The power supply circuit is a classic design with an oscillator and an IR2153 driver. The driver is powered through a resistor from the 300 V bus, and also self-powered from the half-bridge midpoint through a 47 pF capacitor. The frequency is 30 kHz; the power switches are IRF740. I also added transistors to the MOSFET drive circuit to speed up turn-off of the power switches. The output rectifiers are six UF5404 diodes on the 60 V channel and two on the 25 V channel. The fan control circuit uses a 34063 PWM converter. The input filter capacitors are two 470 µF × 200 V units connected in series (a bit undersized; better to use a pair of 680 µF capacitors). The output capacitors are two 1000 µF × 100 V units plus one 680 µF × 100 V unit; the 25 V channel has a 1000 µF × 35 V and a 220 µF × 35 V capacitor.
I started by making the PCBs for the whole project.
Here are the power section boards and the front panel boards
I etched the boards, cut them, and tinned the traces.
I gathered almost all the components. For the photo, I laid everything out at once; normally, I look for parts as I go during assembly.
The board is assembled. There are a few minor flaws, but they're perfectly acceptable for a one-off device.
When I was deciding which transformer to use, I even considered installing several ready-made transformers from computer power supplies. But I changed my mind and decided to make one myself. It's not difficult: all you need is a core (by the way, if anyone in Ukraine is looking for one, I can tell you where to buy it), enamelled wire, and insulating fabric tape (you can insulate it with other materials too, but I find the tape more convenient).
The power supply board is fully assembled and tested.
I killed two transistors in the process: I accidentally soldered an IRF740 alongside an IRF3205. The second transistor couldn't take that kind of abuse and died, taking its partner with it. No one else was harmed.
A small capacitor ended up on the back of the board, though it wasn't part of the original design. It turned out that the 34063 regulator IC doesn't work well with a fan as the load, so I had to add the capacitor in parallel with the upper feedback-divider resistor. After assembly and testing, I coated the board with two layers of Plastik 70, just to be safe.)
During assembly, I discovered a few design flaws (and cost-cutting measures) by the manufacturer.
The board's input has a 1000 µF, 63 V capacitor. It would have been dangerous to leave it in place, so I replaced it with a 470 µF, 100 V one. I also replaced the 100 µF, 50 V capacitor at the input of the regulator that powers the board's logic. As you can imagine, it wouldn't have lasted long there. That's a serious design flaw by the manufacturer; I replaced it with a 100 µF, 63 V capacitor.
I used a 100×63 board rather than a 100×100 one, which would have made more sense, because I decided to make a few improvements to our Chinese friends’ design. The converter board consists of two sections: the power section, with a 100 V transistor, and the low-voltage section, with a PWM regulator at the input. The regulator’s maximum input voltage is 65 V, and since I’m feeding it roughly that same voltage (so I can make full use of the board), this could cause problems. So I decided to improve the circuit.
The low-voltage circuitry regulator was connected through a 1.5KE13 TVS diode to reduce the voltage supplied to the small PWM controller to 50–55 V. But while studying the circuit, I also found that the board monitors a threshold voltage of about 12 V (to prevent operation at low voltage) and measures the input voltage. To avoid interfering with its operation, the 12 V monitoring point needs to be moved to the output of the TVS diode, while the input voltage measurement should stay where it is.
I bought some terminals. There are quite a few because I decided to add several fixed voltage outputs to the power supply. For now, they’re 5, 9, and 12 V.
A separate board is used to mount the extra terminals (it’s included in the overall drawing). The board fits over the terminal screws and is held in place with nuts. It’s the only board designed to be installed with its traces facing the front panel. It has 0.9 A resettable fuses, with a resistor-and-LED circuit connected in parallel with each fuse to indicate when it trips. The fixed voltages are provided by DC-DC converter boards (covered in one of my reviews), powered from the main power supply’s 25 V output.
Another front-panel board (with two holes) fits over the main output terminals. It holds a 100 µF, 100 V capacitor and a film capacitor rated for at least 100 V; the higher the capacitance, the better.
I drew up a simple front panel. The layout is very compact, but it turned out to be quite convenient in use. I also changed the positions of the first two buttons. I’m used to the right button increasing the value and the left one decreasing it, but the board has them the other way around.
By the way, if you use an encoder with a push button, its contacts are already routed on the board, so you can use it either as the OK or SEL button. There’s a backlight LED under the encoder, powered through a 180 Ω resistor from 5 V (available on the outer pins of the display connector). These are the only extra components I added.
Review of the encoder and backlight here
All the ribbon cables are wired for direct connection: the pin order and pin 1 location on each board connector match those on the converter board.
I decided to fit the power supply into a small enclosure. In fact, that’s the main reason I went to the trouble of building a custom power supply (the power section) for this converter. It would have been much easier to buy a ready-made 60 V power supply instead of building one myself.
The boards are mounted inside with an offset so they don’t touch each other and fit within the available height.
The converter board is mounted on the bottom of the enclosure, since I attached the front and rear panels to it, and all the main wires run from those panels. The power supply board is mounted on top. The ventilation openings are at the front, and the fan blows air through this “sandwich.”
The fan is covered by a grille at the back.
The layout inside is very compact. The bench power supply is fully assembled and tested.
And here are some photos of the finished result.
Yes, your eyes aren't playing tricks on you. The indicator is green in the last photos.
After using the indicator that came with the kit, I realized it wasn't very convenient for me.
It looked worse in the photos than it did in real life, but when I came across a green indicator lying in a drawer at home, I tried it and decided to keep it. It fit with almost no issues: I had to add a diode in series with the indicator's 5 V supply because the contrast was too high, and add a 10 Ω resistor to the backlight supply (the original one was already on the indicator board).
It's worth noting that not all boards come with a blue indicator. A friend bought the 6010 boards, and they had green indicators. I do plan to replace it with an indicator made using either VATN or OLED technology; the former is a little cheaper. But I'd like to find them on AliExpress—has anyone come across one?
I also decided to go to the trouble of adding a power indicator for the device. In the last photos you can see that it's built into the encoder knob. There wasn't much room on the front panel, and while making it I noticed that the encoder shaft is hollow inside. I put a bright 3 mm green LED under the shaft, but that's a matter of taste.
Speaking of LEDs, our Chinese friends used a very elaborate circuit to drive the CC and CV LEDs. As a result, they run at just 1–1.5 mA, and there's no way to increase it. So if you're adding extra LEDs to the front panel, it's best to use high-brightness ones; they'll be visible enough. Otherwise, regular LEDs will barely be visible.
So what did I end up with? A lot of (mostly) pleasant memories of the build, and a decent bench power supply that provides:
Main output voltage up to 62 V at up to 5.1 A, i.e. over 300 W of power.
Convenient parameter control.
10 memory slots for presets.
The ability to control it from a computer (via UART or a USB converter; if you really want, you can even add a 485 interface and control it from a kilometer away. And the more adventurous can install a UART-Ethernet module and control it from anywhere).
I simply connected mine using a USB-RS232 TTL converter with galvanic isolation — link
The program's working window (the program has been tested and is guaranteed to work)
Front panel connectors.
Note: On the LED connector, + is +5 V and — is the board ground; you can use it to indicate that power is present.
On the button connector, the second pin is the board ground (the buttons are wired in a rather clever way), but the first pin is the common connection for the buttons. I swapped the first two buttons around for convenience.
I may have mixed up left and right on the encoder connector; I drew it from memory.
3 channels with fixed voltages of 5, 9 and 12 V, delivering up to 1 A per channel.
Smooth fan speed control based on temperature.
Very compact for a device like this.
So, to sum up.
I really liked the board, despite its shortcomings. I think this review should make things easier for anyone considering it.
If I were buying one now, I’d probably go for the 10 A version; 20 A seems like overkill for home use. I spoke with a friend who also bought similar boards. Many thanks to him for finding the control software. He said he plans to write his own program
Pros.
Fully featured programmable DC-DC converter
Can be controlled from a computer.
Pretty decent specs.
You can get a good bench power supply with minimal investment.
Essentially a complete device: the board has a display, buttons and a rotary encoder. It even comes with a knob for the encoder (some packages also include a UART-USB module and cables for connecting the front panel).
Cons.
The input capacitor has virtually no headroom, and neither does the capacitor before the secondary power supply’s PWM circuit (though this may be an issue with the board from this particular seller).
Shipping from this particular seller takes a very long time.
I wasn’t too impressed with the included display, though apparently some packages come with a different one.
The output for external LEDs is very weak; installing high-brightness LEDs solves this.
The PWM supply for the control circuit has little headroom, so take precautions when using the maximum input voltage.
The power inductor wasn’t secured, so I had to do it myself.
Would I recommend this board? Absolutely—yes. Would I buy it again? Same answer—yes.
This review turned out very long and complicated, and I’m sure it has mistakes and omissions. If you have any questions or easy suggestions, let me know; I’ll add or correct whatever I can.
Thanks in advance, everyone.
Update.
Someone from Portugal contacted me (I think), having even registered on Muska specifically for this. He found the schematic for this power supply and asked me to add it to the review.
Here it is. diogoc, thank you very much.
регулируемый источник питания на 5-24в на 18650 своими руками, модули бп из китая с цефровым индекатором, 4848, лабораторник из ибп и модулей из китая, мини лабораторный блок питания своими руками, сетевой блок питания китайский модуль замена, самоделки на китайских модулях, китайский модульный бп с цыфровой, лбп на qc, самоделки дсдс модулей, модули бп из китая с индекацией, самодельныйлабораторный блкпитания из иодулей, лабораторный site:kirich.blog, лабораторный блок питания на аккумуляторах 18650, https://www.kirich.blog/obzory/15-neplohoy-modul-dc-dc-zxy6005s-ili-laboratornyy-blok-pitaniya-svoimi-rukami, бп из китайских модулей своими руками, лабораторный блок питания на литиевых аккумуляторах, модуль из китая для лабораторный блок питания обзор, источник питания 0 - 300 вольт с регулировкой напряжения и тока, лабораторный блок питания из запчастей с алиэкспресс, схема китайского модуля dc dc 300вт 20a, обзор китайских модулей силовых, самодельный лбп на wz5020, двухканальный лбп из китайских модулей, самодельный лабораторный блок питания 24вольта 20ампер на китайских модулях
Questions and answers
What voltage and current ranges does the ZXY6005S offer?
The stated ranges are 0–60 V and 0–5 A; the article says adjustment can reach 62 V and 5.1 A. The adjustment steps are 0.01 V and 0.001 A.
How can the ZXY6005S be controlled?
Settings can be changed with the buttons and rotary encoder, or from a computer via RS232 at a 3.3 V level.
What problems did the author find with the ZXY6005S?
The author reports a loud buzzer whose mute setting is not saved, little headroom in some capacitors, and a weak output for external LEDs. The power inductor was unsecured, so the author fixed it in place.
This model is included in the summary table: Comparison of adjustable converters for a bench power supply
The figures from this review are in the measurement database: Converters
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!
Comments8
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
















































Comments are published in their original language.