Fnirsi DC-6006L Adjustable Voltage Converter

Some time ago, I posted information about a new model of adjustable converter that I’d stumbled across completely by chance. But that very same day, a company representative contacted me on Skype and offered to write a review of the model. I received it not long ago, tested it, and am ready to tell you more about it.


Since I already have quite a few adjustable power supplies of various power ratings, I can’t say I was eager to take apart yet another one. The original goal of the article was simply to tell you about the new product I’d found. But when they offered to send one for testing, I thought, why not? It turned out quite a few readers were interested, too.
The company representative wrote that the unit shown in my article was essentially a pre-production sample, sent to one of the bloggers and featured in his review. I was to be sent the final version, though it was still being fine-tuned and would ship once production began. They sent it about three weeks later. The package took around a month to arrive, which is actually unusually fast for Fnirsi, as it normally takes much longer. I picked up a fairly large polystyrene box at the post office and, for some reason, immediately thought of converters from another well-known company that packages its products the same way.


Inside was a hard case, exactly like the one included with their 1C15 oscilloscope. While unpacking it, I even briefly wondered if they’d accidentally sent me a new oscilloscope model instead of a converter. But no, inside was the DC-6006L.


The store offers four package options in total. Mine is the basic one, though they could have sent me the full kit.
1. DC-6006L converter
2. USB-to-microUSB cable
3. Rubber feet
4. Case.

The expanded packages may also include a USB-C cable and one or two sets of power leads.


The converter comes without feet; they’re simply included in the box. Since it slides around on the desk without them, I stuck them on right away.
The cable is very flexible, which is probably its only advantage. Otherwise, it’s a standard microUSB cable.


Specifications
Input voltage: 6 V–70 V
Output voltage: 0–60 V (Note: Make sure the input voltage is at least 1.1 times the output voltage)
Output current: 0–6 A
Output power: 0–360 W
Voltage resolution: 0.01 V, accuracy: ±0.3% + 3 counts
Current resolution: 0.001 A, accuracy: ±0.5% + 5 counts
Output ripple: 100 mV p-p

When I picked up the converter, two thoughts came to mind:
It’s so small!
Everything is put together so neatly.

After a closer inspection, my second impression changed a little, as I spotted a few minor blemishes. Still, it looks quite nice in its own way.


On the product page, I saw an interesting comparison with other converter models.


It reminded me of similar ads from well-known car manufacturers :)
You can’t help asking: your move, RuiDeng. But actually, that move had already been made several years ago with the DPS8005 model. It has a maximum current of only 5 A, but it’s more compact and can output up to 80 V.



And it really is small: 40 mm high, 87 mm wide, and 103 (118) mm deep.



Unlike the version I showed in the announcement, the front panel looks different. They removed the memory access buttons, and I think it’s actually better this way.
The front panel has:
1. Display
2. Encoder
3. Navigation and output on/off buttons
4. microUSB port for connecting to a PC
5. Output terminals.

The front panel is made of thin plastic. Also, during disassembly—or, more precisely, reassembly—it’s easy to scratch the paint.



The rear panel is made of fiberglass laminate, with the fan grille simply milled into it.
The power supply input connectors are also on the back. There are three options:
1. Two banana jacks, the same as on the front panel
2. 5.5/2.1 mm jack
3. USB-C port

I can already hear the question: what if... Yes, if you apply power to the banana jacks or the 5.5 mm socket, voltage will also appear at the USB-C input, so only one connection can be used at a time.
One more thing: using banana jacks for the power input isn't a great idea. The wires can be pulled out and accidentally shorted together. This would be especially spectacular if they were battery leads.


I was going to take it apart next, but first I checked that the converter actually worked. I connected it to a power supply with a USB-C output, and it worked right away. The screen is bright and sharp, with decent viewing angles.


The power supply isn't particularly difficult to take apart, though I wouldn't call it convenient either. First, remove the four screws at the front and the four at the back. Interestingly, the front screws are yellow, while the ones at the back are white.
But that only lets you remove the top cover.


The layout is fairly spacious. The fan blows air into the case, but the only ventilation holes are near the heatsink, and there aren't enough of them. I'd add some more to improve cooling.


The first thing I noticed was the poor soldering on the power connectors. You can see that they were cleaned before soldering, which is good, but they were clearly soldered with an underpowered iron, which is not. The connectors also have some kind of washers that keep them from sitting flush against the panel. I'd fix this, or at least recommend checking it after purchase.


To remove the board, you'll also need to unscrew the fan. Then you can slide the board out of the case.


There are three capacitors in the input and output circuits: two 330 µF and one 150 µF, all rated for 100 V. I also spotted ceramic bypass capacitors on the board, which is very useful.


Nothing interesting on the underside. You can see the chip that handles Power Delivery, but its markings have been erased.
The unit consists of two boards: the converter itself, and a board with the display, buttons, and other controls.


To get a better look at the board and its components, you can move the wires out of the way. The connectors on the board are labeled: one ribbon cable connects the display, and the other handles control.


Overall view of the board


A quick look at the components
1. Input regulator and, presumably, a transistor for reverse-polarity protection at the input
2. Output transistor, presumably used as a current regulator to discharge the output capacitors faster. The 3.3 V regulator is next to it.
3. Current-sense shunt. Under it are a fuse and an LED, which, judging by how it is connected, is probably there to indicate a blown fuse.
4. A pair of op-amps: an LM358 and a precision GS8332
5. STM32F103CBT6 microcontroller
6. CH340N UART-to-USB converter. Yes, you read that right: there's no galvanic isolation. As someone used to converters from RuiDeng, I found that a complete surprise.


1. Power stage. The inductor is quite large, at least for a claimed 360 W. The power components are mounted on the heatsink next to it.
2. TL594 PWM controller, with some kind of auxiliary converter and an SOT23-6 chip I don't recognize nearby
3, 4. Since both components mounted on the heatsink are transistors, this appears to use synchronous rectification. Both transistors are marked SKD502T, and both are electrically isolated from the heatsink.


The display board is soldered to the front-panel board via a pin header. The display is soldered on slightly crooked, and it won't be easy to straighten it out.



Now let's move on to the operating modes and tests.

1. When powered on, the manufacturer's logo appears on the screen
2. The main screen then appears, showing the set voltage and current, output voltage, current and power, temperature, a protection indicator, CC/CV operating mode, how long the output has been active since power was applied, and an output activation icon.
3. Turn the encoder to select other display modes, such as the delivered capacity in Wh and Ah, and the input voltage
4. There's also a screen showing the input and output voltage, as well as output current and power.
5. Next is the settings menu
Automatic output activation when power is applied
Language
Display timeout
Image rotation
Temperature units: degrees Celsius or Fahrenheit
6. Second settings screen
Maximum voltage
Maximum current
Maximum power
Temperature at which the emergency shutdown occurs
Output shutdown timer.

In my opinion, the controls aren't very convenient or intuitive. For example, to set the voltage, you press the V/A button, use the left/right arrows to select the adjustment increment, then turn the encoder to set the desired value. But there's another complication. You can turn the encoder quickly, but it only registers about five steps per second. For example, I'm used to being able to spin the encoder very quickly on my power supplies, and in some situations I don't even switch to a coarser increment, preferring to make a few extra turns of the encoder. You can do that here too, but it takes a long time.

Changing settings in the menu is no more convenient. To change something, you press the left/right button, then use those same buttons to move through the menu. To activate an item, press the encoder, turn it to adjust the setting, then press it again to exit. I got confused at times, even though I'm used to working with different power supplies.


The manufacturer provides an explanatory diagram


Settings menu options
1. Auto-start — on/off
2. Language — Chinese/English
3. Display timer — on/off and time setting; activate it later by pressing and holding the encoder
4. Image rotation — 0, 90, 180, and 270 degrees
5. Maximum output voltage
6. Maximum output current
7. Maximum output power
8. Overtemperature protection — on/off (off by default) and limit, up to 100 degrees
9. Output shutdown time — on/off and limit.


Later, I found a description of how the controls work in the manual.


Testing. For convenience, I'll cover it in the same order as I did for the other converter models.
Test 1: checking the input voltage measurement accuracy.

The converter starts up with 5 V at the input and continues to work at 4 V, although the screen goes blank. No complaints about voltage measurement accuracy across the entire tested range.


During testing, I found an issue. While switching the input power on and off, I noticed that the multimeter's autoranging changed its range when power was applied. I enabled peak hold and saw that the output briefly spiked to 11 V at startup. Apparently, the pulse is very short, since in normal mode all I could see was the range changing.


Test 2: output voltage setting accuracy and the accuracy of the converter's own meter.
Overall, I could say that everything is excellent, but there's a minor issue: at voltages below 0.1–0.2 V, the converter correctly sets the configured value, but its own voltmeter displays nothing.
Otherwise, the calibration is spot on. I checked at 0.1, 0.2, 0.5, 1, 5, 10, 20, 30, 40, 50, and 55.07 V.
Yes, that last value is rather odd. It's the maximum I could set with 61 V at the input, but more on that later.


Test 3: output current setting and measurement accuracy.
First, apologies for the “lots of pictures,” but it would be quite difficult to describe what I saw any other way.
This test is usually performed with the converter loaded by the multimeter inputs—in other words, it's effectively operating in short-circuit mode. This lets you assess both the current setting accuracy and the current measurement accuracy at once. I did it the usual way here too, but ended up with some rather unexpected results.

Output current limit set to: 1, 10, 50, 80, 60, 50, 100, 200, 500 mA.
I deliberately increased the current first and then decreased it. You can see that the converter responds differently in each direction. It starts working more or less properly at 200–500 mA.


But when I looked at the multimeter's “fast bar graph,” most of the glitches started to make sense: its CC mode doesn't work correctly—or, more precisely, the feedback isn't handled correctly. The video didn’t turn out very well. I spotted this during the test, switched the camera to video mode, and quickly filmed a short clip to demonstrate it.


I wanted to see what was happening in peak measurement mode, but at 500 and even 400 mA the meter went off the scale. I had to set it to 300 mA, and then I could see the actual current peaks.


Measurements at current settings of 1, 2, 4, 5, 5.5, and 6.1 A
Everything went very well up to 5 A, but then the glitches started again. In 6.1 A mode, I got 7.3 A!


But it got even more interesting when I switched the multimeter back to peak measurement mode. Here are the results with the current set to 4 and 6.1 A.


After all the tests, I was testing the converter with an electronic load and noticed that, generally speaking, everything was working normally. The current was measured correctly in CV mode, and I checked the same thing with a load rather than in short-circuit mode.
In my opinion, it really does work noticeably better here: the multimeter shows the actual current, while the converter display shows the measured value.


Test four: output voltage regulation across load currents from 50 mA to 6 A and voltages of 3.3, 12, 24, 36, 48, and 54 V.
Voltage was measured directly at the socket contacts where they are soldered to the board, to rule out their influence—in other words, this tested the board itself. In all modes, there was a consistent 0.1 V drop at 6 A.


Test five: checking the minimum required difference between input and output voltage.
The specifications said the input voltage had to be 10% higher than the output. It turned out to be a software limit: you simply can’t set the output voltage higher than the input voltage minus 10%.
For example, with no load, an input of 61 V let me set the output to 55 V. Under load, the input voltage dropped, so the output setting was adjusted accordingly.


In reality, the converter could regulate the voltage at 6 A with just 2.5 V between input and output. The drop from the set value here was caused by the voltage changing under load, not by insufficient input voltage.


Test six: measuring output ripple amplitude under different operating conditions.
The output was set to 30 V, with an input of 61 V.

1. No load
2. At 2 A
3. At 4 A
4. At 6 A

The specifications claimed ripple of up to 100 mV, but it exceeded that figure at just 4 A. At 6 A, the ripple amplitude was 150 mV.
The ripple was also quite high with no load, which is a little strange.


All the oscilloscope traces were taken with the probe connected directly to the board output. I could have connected capacitors in parallel with the probe, as I did in the previous review, but this converter model has no output pins, so the difference in the measurement results would be practically zero.
For comparison, here’s an oscilloscope trace with capacitors connected in parallel with the probe.


In all these cases, the load current is 6 A, while the output voltage is 5, 12, 24, 36, 48, and 54 V. You can see that ripple amplitude is greatest around the middle of the output voltage range, which is why I showed the traces at 30 V above.


Test seven: transient response.
1, 2. Applying and removing 5 V at the converter output with no load.
3, 4. Response to connecting and disconnecting a load. The output was set to 12 V, with a 30 Ω resistor connected continuously and a 3.3 Ω resistor connected in parallel with it (or disconnected). So this shows the response to a current step from 0.4 to 4 A and from 4 to 0.4 A.
5. Transition from CV to CC mode: 15 V, load current 4.5 A, current limit set to 3 A.
6. The same load and settings, but in reverse: transition from CC to CV.

The first four tests raised no particular concerns. The results were roughly like those for the RD60xx series converters, but the last two...
You can see that in CC mode, the converter output has an absolutely wild ripple amplitude. This is exactly what we saw in the current-setting accuracy test. I can barely comment on it—it's awful...


“Soldering iron test.” A TS100 soldering iron is connected as the load, and the power supply’s response to constant current changes is checked.

Based on the previous test results, I expected this to be a disaster, but to my great surprise, the converter handles this kind of load even better than the RD converters, despite clearly having lower capacitor capacity. You could even say it's quite acceptable.


Test eight: measuring converter efficiency.
The test measured efficiency in different operating modes. Input current was monitored using the RD6018 display, input voltage using the meter on the converter under review, output voltage using a multimeter, and output current was set with an electronic load.
The results at an input voltage of 60 V and a load current of 6 A:
Output voltage — efficiency — power loss
5 V — 87.9% — 4.04 W
12 V — 93.1% — 5.26 W
24 V — 95.1% — 7.42 W
36 V — 96.0% — 8.83 W
48 V — 97.0% — 8.81 W

At an input voltage of 20 V and a load current of 6 A:
Output voltage — efficiency — power loss
5 V — 88.5% — 3.84 W
12 V — 94.3% — 4.3 W
17 V — 95.1% — 5.16 W

Overall, I think that's pretty good. The synchronous rectifier helps improve efficiency.


I checked the temperature at a load current of 6 A and output voltages of 20 V and 40 V, running each mode for 20 minutes. The hottest component was the power inductor (60 degrees); the thermal image also shows the path of the power traces, which are hotter in those areas. The heatsink was barely warm. By the way, the fan turns on either when the load power reaches 100 W or more, or when the temperature reaches 40 degrees or more.
I do have some concerns about cooling, though. As I mentioned, the fan blows air in, and there are very few openings in the case.



Testing operation with a PD input.
For the test, I used a power supply with a PD output and a cable that allows it to deliver 5 A, for a maximum power of 100 W.
With no load, the converter draws 70 mA. At a current of 6 A and the highest output voltage I could set, the input current is already around 5.5 A. This means the power supply will be operating in overload, so you need to keep an eye on it.
Still, this operating mode and voltage headroom are quite sufficient, for example, for charging a car battery.



Software.
The program is very simple and straightforward, with minimal service features. It can update the firmware, but there are no firmware files. You can download all Fnirsi software here.

To connect, click "open", then "connection". To disconnect, do the same in reverse order.
As soon as the software starts, a graph begins scrolling. The graph always starts at zero, but runs regardless of whether the converter output is active.
There are two more buttons at the bottom, open and clear. The open button lets you open a bin file, and you can choose the packet size.



One drawback is that the current and voltage graphs are shown on separate tabs, so you can't view them at the same time.


You can also set a time limit for operation here, but note that this is set on the converter itself, so the limit will be saved in the unit's settings.
You can also save data as an XLSX file. It's odd that they didn't include the more familiar CSV format.


Data logging also starts as soon as you connect to the converter, but there's a reset button.
The logging rate is around 4–5 readings per second. The log records the actual time, record number, voltage, current, power, and board temperature.



I'll keep today's conclusions relatively brief, since I've covered most of the points throughout the review, trying to be as objective as possible and backing everything up with test results. Still, although the device is quite rough around the edges, the manufacturer has made a pretty good start. The display is very well implemented, the case is neat and even comes with a carrying pouch, and I have no complaints about the front-panel layout either.

The main concerns are the problems with CC mode and the very high ripple level. Batteries won't mind, but other loads might. It also affects current measurement accuracy.
Ripple in CV mode is higher than specified, but this only occurs in certain operating modes. On the plus side, there are no short spikes (needles), but at maximum current and a voltage of 20–30 V, the total ripple is 150 mV, compared with the specified 100 mV.

There are other issues too, such as the design—or, more precisely, the poorly tuned manufacturing process and quality control, as evidenced by the poor soldering on the power connectors. There's no protection against high voltage reaching the USB-C input from the converter side. The controls aren't very convenient, especially when setting current and voltage. The software also needs work; they could use the software for the RD60xx series converters as a model. And of course, add galvanic isolation to the port used to connect to a computer.

In short, I can say that the manufacturer rushed things a bit. In my opinion, they should have spent more time on the circuit design and software (fortunately, updates can be released here). It can be used, but only with non-critical loads. Quality control during assembly could also use some improvement.


That’s all for today. I hope this was useful. If you have any questions, I’ll be happy to answer them.