|
On AliExpress, this linear power supply kit is on offer for just € 41.37. That seems like a bargain for a power supply which, according to the supplier, can deliver 28 V at 2 A. Unfortunately, it turns out that this kit doesn't meet the specifications at all. |
An introduction to the TJ-56-549
The end result
The end result after assembling this kit looks a little unusual. A fully transparent casing measuring 14.5 cm by 10.0 cm by 7.5 cm, with a switch, a display, two rotary encoders and two 4 mm female banana sockets on the front. The whole unit weighs 705 grams. On the right-hand side is a temperature-controlled fan; the mains power is supplied on the left-hand side.
 |
| The end result of this kit. (© AliExpress) |
One encoder allows you to adjust the output voltage in steps, whilst the other sets the maximum current. A handy feature is that the step sizes depend on how fast you turn the knobs. If you turn them quickly, the step sizes are large; if you turn them slowly, they are small. Pressing the encoder button saves the values to memory. These will become the start-up values the next time you switch on the power supply.
The voltage measurement has a resolution of 100 mV, the current measurement 10 mA.
The first alarm bells are ringing!
Forty euros for a linear power supply capable of delivering 56 W of power is, of course, a real bargain. But... a quick glance at the previous photo triggers our alarm bells. Comparing the dimensions of the transformer – clearly visible on the left of the photo – with those of the 4 mm banana plug sockets, it’s clear that the transformer is truly tiny.
Have the Chinese developed a new, revolutionary technology to extract more power from a iron core transformer? Unfortunately, a search via Google’s Gemini for the specifications of such a transformer, with the brand name Yanglin and model number 48*24, reveals that this transformer has an EI-48 core with a stack height of just 24 mm. Such a transformer can deliver a maximum power of 10 VA. Even if a slightly wider tolerance is applied for the maximum heat dissipation, such a transformer will not be able to deliver more than 15 W!
In short, this is the first evidence that the specifications of this linear power supply are greatly exaggerated. We are curious to see what this design can actually deliver in practice.
 |
| The transformer supplied is far too small. (© AliExpress) |
Manufacturer, supplier and price
It is unclear who designed this power supply and who put the kit together. The supplier offering this set via AliExpress is the 'Electronic DIY Tools Store'. A complete set, including the transformer and housing, is on offer for € 41.37, including postage.
Delivery of the kit
All the parts are packaged in two plastic bags. One contains all the electronic components, the display and the two circuit boards; the other contains all the parts for the housing. The transformer is supplied separately. Also included is a very brief assembly guide.
 |
| The contents of the kit. (© 2026 Jos Verstraten) |
The components supplied
We have neatly laid out all the supplied parts. In the photograph below, you can see the two circuit boards at the top left, with the standard 1602 LCD display beneath them. The fan is situated below this display. Just how small the supplied transformer is – located beneath the two test leads – becomes clear when you compare its dimensions with those of the two 4 mm banana plugs on the test leads.
The housing is assembled by ingeniously connecting six pre-milled Perspex plates using bolts and nuts.
 |
All the parts supplied. (© AliExpress) |
The assembly instructions
The instructions included in the package are very brief, but fortunately there is a comprehensive and highly detailed English-language guide available online. Well done! It describes every step of the assembly process, illustrated with a photo. We have uploaded these instructions to our own ‘cloud’ on Google Drive; you can download them using the link below:
The circuit boards for the circuit
The electronics are housed on two circuit boards. One contains the actual linear power supply with rectification, smoothing and current limiting. The other houses the digital section, comprising the processor, the display and the two encoders. Both circuit boards are designed with adequate space, and soldering the components should not pose any challenges even for a novice hobbyist.
 |
| The two PCBs of the power supply. (© AliExpress) |
The specifications
In fact, hardly any specifications are given:
- Output voltage: 0.0 Vdc ~ 28.0 Vdc
- Output current: 0.01 A ~ 2.00 A
- Current setting resolution: 0.01 A or 0.05 A
- Voltage setting resolution: 0.1 V or 0.5 V.
Hum? Noise? Ripple? Output stability? Never heard of them, sir!
The electronics in the TJ-56-549 power supply
The linear power supply
The complete circuit diagram of the actual power supply is shown in the figure below. The circuit is rudimentary and the very simplicity itself; there are absolutely no frills.
A transistor D1047 is used as Q1. It can handle a collector current of 12 A at a maximum power of 100 W. The collector/emitter-voltage must not exceed 140 V. A microcontroller output cannot supply sufficient current to drive the base of this semiconductor directly. Hence the three transistors Q2, Q3 and Q4, which form a current amplifier to supply sufficient base current to Q1. This circuit is driven via the base of Q3 from the Vctrl output of the controller PCB.
The resistors R3 and Rp form the voltage divider which feeds back a portion of the output voltage via the Vsens wire of the ribbon cable to the microcontroller. By adjusting Rp, you can ensure that the output voltage of the power supply equals the value shown on the display.
R1 is the current-sensing resistor which feeds back to the microcontroller the value of the current supplied via the Isens wire of the ribbon cable. Resistor R2 and the zener diode ZD1 form a protection circuit, ensuring that the voltage on Isens can never exceed approximately 5 V.
The analogue circuit board also contains two voltage regulators, U1 and U2, which supply +5 V and +12 V to the microcontroller and the fan respectively.
The block labelled J4 is the schematic representation of the KSD9700 thermal switch. This switches on when the heat sink of transistor Q1 exceeds 50 °C. It connects the fan (J5) to the +12 V supply voltage.
Diodes D5 and D6 protect the power supply against reverse-polarity voltages or excessive voltages that you might accidentally connect to the J2 outputs.
It should be noted that no large capacitor is connected across the power supply’s output terminals. This is not permitted with a power supply featuring adjustable current limiting, as the charge stored in such a large electrolytic capacitor could cause the power supply to deliver more current than you have set.
 |
| The circuit diagram of the linear power supply. (© AliExpress) |
The control electronics
There is actually very little to say about this. On the left-hand side of the diagram below, you can see an R-2R digital-to-analogue converter. This generates an analogue voltage from the microcontroller’s outputs PD0 to PC2, which is undoubtedly used to control the linear power supply. On the right is the control section, featuring the RW1B and RW2B switches for the rotary encoders and the RW1A and RW2A switches, which close when you press the encoder buttons. The PR1 adjustment potentiometer is used to control the brightness of the display.
 |
| The circuit diagram of the control electronics. (© AliExpress) |
Assembling the power supply
Assembly of the control circuit board
Start by soldering the 25 resistors. Next, fit the 100 nF capacitor. Then solder the IC socket. Solder the black 16-pin connector to the circuit board. Now it is time to fit the 22 nF capacitor and the 10 kΩ adjustment potentiometer.
Turn the circuit board over and fit the 5-pin connector for the ribbon cable in the correct position on the solder side of the board. Turn the circuit board back over and solder the five pins of this connector to the component side of the board. The correct orientation of this connector is shown in the photograph below.
 |
| The correct position of the connector. (© 2026 Jos Verstraten) |
Now solder the two rotary encoders in place and, finally, fit the six 10 mm copper spacer sleeves to the component side of the circuit board. Fit the IC into its socket.
You can see the result of all this work in the photograph below.
 |
| The fully assembled control circuit board. (© 2026 Jos Verstraten) |
Assembling the analogue circuit board
The manual suggests a rather unusual assembly order. It is more practical to start with the very smallest components and then move on to the larger ones.
Start by soldering the small 5V1 ZD1 zener diode above the position where the 7805 is to be fitted. Pay attention to the polarity! Then continue by soldering the five resistors. Solder the two 1N4007 diodes; here too, you must pay close attention to the positions of the anode and cathode. Soldering the single light-brown capacitor C2 completes the assembly of the smaller components.
Press the two transistors, 9015 and 9014, into the circuit board and solder them in place. Prepare the rectifier for operation by fitting the four 1N5408 diodes onto the circuit board and soldering them. Ensure that these components are raised a few millimetres above the surface of the circuit board. Next, fit the small two-pin connector for connecting the fan in the top left-hand corner of the circuit board. Pay attention to the position!
Now it’s time for the five-pin connector for the ribbon cable; pay close attention to the position! Then solder the two terminal blocks to the circuit board, with the holes facing outwards. You can now fit the adjustment potentiometer R7 in place and solder it. Pay attention to the position of the adjustment screw! Next, press the transistor D882 into the circuit board and solder it. Again, pay attention to the position; the metal back should face inwards.
Fit the 0.47 Ω wire-wound resistor onto the circuit board, ensuring there is a half-centimetre gap between the board’s surface and the underside of the resistor. Solder the two 100 μF electrolytic capacitors, taking care to respect the polarity.
Screw the two voltage regulators, U1 and U2, onto the heat sinks and mount them on the circuit board. Mount the large 2,200 μF smoothing capacitor on the circuit board and solder it in place. Pay attention to the positive and negative terminals! Mount the power transistor D1047 on its heat sink and solder it in place.
There is now still an empty space in the top left-hand corner for the thermal switch. Bend the connecting wires of this component so that it is clamped between the transistor Q1 and the left-hand flange of the heat sink.
 |
The fully assembled analogue circuit board. (© 2026 Jos Verstraten) |
Assembly inside the housing
This is clearly described in the manual; if you follow all the steps, it’s guaranteed to go well. The photo below shows the complete wiring that you need to connect between the various circuit boards and components.
 |
The power supply wiring. (© Electronic DIY Tools Store) |
Afterwards, you must screw the six Perspex panels together to form a complete housing. This is a fiddly job, not only because of the wiring running between three of the panels, but also because of the awkward way in which you have to fit the panels onto and into one another. In total, you’ll need to make 36 of the bolt-and-nut connections shown below!
 |
| Construction of the housing. (© 2026 Jos Verstraten) |
Adjustment
You can adjust the brightness of the display using the RP1 adjustment potentiometer on the controller circuit board.
Connect a multimeter set to DC voltage to the outputs and turn the encoder to adjust the voltage until a value of 20.0 V appears on the display. Then turn the adjustment potentiometer RP on the analogue circuit board until the multimeter reads the same value.
Working with the TJ-56-549 power supply
Setting the voltage and current
This power supply is very easy to operate. Use the left-hand rotary encoder to set the output voltage with a resolution of 0.5 V or 0.1 V. Use the right-hand encoder to set the maximum current the power supply may deliver, with a resolution of 50 mA or 10 mA. This depends on the speed at which you turn the knobs. Pressing the encoder saves the set value to memory, and this becomes the value the power supply will use the next time it starts up.
The display shows both the current voltage and current values (bottom) and the ‘SET’ data stored in memory (top).
 |
| The data on the display. (© Electronic DIY Tools Store) |
No standby mode!
The TJ-56-549 does not have a standby switch; the set voltage is applied to the outputs immediately after the power supply is switched on.
Testing the TJ-56-549 power supply
Test conditions
All the tests listed below are carried out with the current setting at maximum (2.00 A), so that the current limit cannot affect the measurements.
We use the following measuring equipment:
- Fluke 8842A multimeter for measuring the output voltage.
- Philips PM2454B mV meter for measuring noise and ripple.
- Alientek EL15 current sink for setting the load current.
- OWON XDS2102A oscilloscope for monitoring the output voltage.
Accuracy of the voltage indication
We measure this under no-load conditions. At the output, the 8842A and the PM2454B measure the output voltage and the noise and ripple on this voltage respectively. As the table below shows, the results are not bad at all for such an inexpensive power supply.
 |
| Accuracy of the voltage indication. (© 2026 Jos Verstraten) |
We next measured the noise and ripple at the 12 V output under no-load conditions. As can be seen from the oscillogram below, the results are not disappointing either. There is virtually no 100 Hz ripple and only a little noise. The narrow, needle-shaped interference pulses so often feared in switched-mode power supplies are completely absent here.
 |
| Noise and ripple at 12 V and no load. (© 2026 Jos Verstraten) |
Accuracy of the current reading
We set the output voltage to 5.0 V and connect our Alientek EL15 current sink to the output. We apply precisely set currents to the power supply and record the current reading on the TJ-56-549. At a load current of 700 mA, things go wrong. The TJ-56-549 starts behaving erratically and, moments later, the power supply is as dead as a doornail. Only by switching off the load current and restarting the power supply do we get things up and running again. We’ll come back to this later. For now, we’re only interested in the accuracy of the current reading, which is summarised in the table below.
 |
| Accuracy of the current reading. (© 2026 Jos Verstraten) |
Output stability at 5.0 V output voltage
Output stability defines how constant the output voltage remains as a function of the load current. This parameter is not necessarily the same at every output voltage, so it must be tested at various voltages. We start at 5.0 V; the results are shown in the table below. The power supply is again connected to our EL15 adjustable load. Connected to the output are the Fluke 8842A, the Philips PM2454B mV meter and our OWON XDS2102A oscilloscope.
Up to 500 mA, the TJ-56-549 delivers a usable output voltage, although the stabilisation leaves much to be desired. The voltage drops by no less than 0.3 V, which is completely unacceptable for a stabilised power supply. When loaded with a current of 600 mA, all sorts of strange things happen. The power supply becomes completely unstable and the needle on the PM2454B swings right to the end of its range.
So, even at an output voltage of just 5.0 V – where the power supply retains more than enough voltage to regulate – little remains of the promised 2.0 A current.
 |
| Output stability at 5.0 V output voltage. (© 2026 Jos Verstraten) |
In the oscillogram below, you can see the ripple in the output voltage at a load of 600 mA. More than 1 V of 100 Hz ripple on a DC voltage of approximately 4.5 V! This is, of course, completely unacceptable!
 |
| Ripple at the output at 5.0 V and 600 mA. (© 2026 Jos Verstraten) |
Output stability at 12.0 V output voltage
We repeat the measurements at 12.0 V. Now things start to go wrong at over 400 mA! At 500 mA, after a few seconds, the same instabilities occur as we observed at 5.0 V. And here too, the voltage drops significantly at 400 mA, by more than 0.269 V.
 |
| Output stability at 12.0 V output voltage. (© 2026 Jos Verstraten) |
To find out exactly what happens in the power supply when the load current increases, we gradually increase the current drawn by the EL15. The results are shown in the oscillogram below. The blue trace represents the zero level. At each step, the output voltage drops by approximately two volts. At a load current of 800 mA, the output voltage even drops to zero!
 |
| Output voltage at currents greater than 400 mA. (© 2026 Jos Verstraten) |
Output stability at 24.0 V output voltage
As might be expected, things go completely wrong at an output voltage of 24.0 V. The power supply can now deliver a maximum of 200 mA, and the voltage collapses when the current is higher.
Switch-on and switch-off behaviour of the power supply at 12 V and 300 mA
We have summarised this by combining two oscillograms. Unfortunately, we are anything but enthusiastic about the switch-on behaviour. The output voltage suddenly jumps to around 20 V and then creeps back to the set value of 12.0 V over approximately 280 ms. Completely unacceptable!
 |
Switching the power supply on and off at 12 V and 300 mA. (© 2026 Jos Verstraten) |
Effect of load variation on the output voltage
When you power a circuit, it draws a certain amount of current. If that current suddenly increases or decreases – for example, when a relay switches on or off – the output voltage of the power supply is expected to remain constant and not exhibit any unusual transient behaviour.
We test this by setting the TJ-56-549 to 12.0 V and supplying a load that draws approximately 100 mA. We then suddenly switch an additional load in parallel for a short time, causing the load current to rise to approximately 300 mA. The response of the power supply’s output voltage can be seen in the oscillogram below. Both when increasing and when reducing the load current, a brief peak of more than 300 mV appears in the output voltage. This, too, is, once again, unacceptable for a reliable power supply.
 |
Effect of load variation on the output voltage. (© 2026 Jos Verstraten) |
The remarkable thing is that these spikes arise counter-intuitively. When increasing the load current, we would expect a negative spike that the control system compensates for as quickly as possible; conversely, when reducing the load, we would expect a positive spike.
Our opinion on the TJ-56-549 power supply
We cannot recall ever having tested a kit that performed as poorly as this TJ-56-549 power supply in virtually all the tests we carried out. The results are frankly disappointing and give the impression that fundamental errors were made in the design of this power supply.
A significant part of the poor performance appears to be attributable to the transformer supplied with the kit. It is far too small for the specified output current of 2.0 A. The specification under which this power supply is offered by the supplier, Electronic DIY Tools Store, therefore does not correspond to what the assembled unit can actually deliver in practice.
However, even if you were to connect a more powerful transformer to the circuit board, we do not believe that the performance of this power supply would improve significantly. After all, our tests clearly show that the regulation system reacts far too slowly to changes at the output.
Our advice: however tempting the price may be, never add this kit to your shopping basket! For an extra twenty euros, you can buy a power supply that actually works properly.
Laboratory power supplies
