New RD6006 (W) Voltage Converter Module from Rui Deng

The new product from Rui Deng has finally made its way to me: an adjustable voltage converter for building a bench power supply. I originally planned a more extensive review, but one of the suppliers really let me down and shipped the power supply much later than I’d hoped, so I had to split the review into two parts: the main review and a follow-up. The follow-up will also answer questions that came up after reading the main review.


The new product was a pleasant surprise for me. I’ve reviewed various converter boards, including three models from Rui Deng, which were enough to form a pretty good impression of the whole range. But at the time, I wasn’t too keen on them because they looked a bit... toy-like, even though their specs were quite “grown-up.”
But it seems the manufacturer heard my thoughts, which I shared in previous reviews, and has finally released the first model in an updated product line.

I ordered the converter together with a case, but for some reason the seller shipped them separately, with two tracking numbers. When I received everything from the Meest Express courier, I understood why: each item has its own packaging, and it’s quite bulky. The larger box had clearly been opened, judging by the carelessly applied tape. I assume this happened at customs, though they usually put on a corresponding sticker when they do that.
Five stars for the packaging, although even so, the seller warns that the case may be slightly deformed. Apparently, he knows the postal services better than I do :)


Of course, I’ll start with the converter.
In addition to the outer packaging, there’s a sturdy cardboard box with color printing. The box also lists the main specs and notes that this is the WiFi version (the W in the model name). There’s also an RD6006 version without WiFi; the price difference is $4. Unlike the previous models, all models now have a USB interface for connecting to a computer.

Full specifications
Model: RD6006
Display: 2.4” color LCD
Input voltage range: 6–70.00 V
Input voltage measurement resolution: 0.01 V
Output voltage range: 0–60.00 V
Output voltage measurement resolution: 0.01 V
Output current range: 0–6.000 A
Current measurement resolution: 0.001 A
Output power range: 0–360.0 W
Battery voltage measurement resolution: 0.01 V
Output voltage accuracy: ±(0.3% + 3 digits)
Output current accuracy: ±(0.5% + 5 digits)
Input voltage accuracy: ±(1% + 5 digits)
Battery voltage measurement accuracy: ±(0.5% + 3 digits)
Typical output voltage ripple: 100 mV VPP
Operating temperature range: -10°C to 40°C
Response time in constant-voltage mode: 2 ms (load 0.1 A–5 A)
External sensor temperature measurement range: -10°C to 100°C / 0°F to 200°F
Load regulation in constant-voltage mode: ±(0.1% + 2 digits)
External sensor temperature measurement accuracy: ±3°C, ±6°F
Load regulation in constant-current mode: ±(0.1% + 3 digits)
Charge capacity measurement range: 0–9999.99 Ah
Energy capacity measurement range: 0–9999.99 Wh
Display brightness settings: 0–5, 6 levels in total
Gross weight: 607 g
Statistical measurement error for charge capacity and energy: ±2%
Dimensions:
167 × 81 × 65 mm
Operating mode: step-down
Voltage drop >1 V and >10%




First impression: it’s clearly larger than all the previous models, and the controls are much more sophisticated. They also included a few extras.


The extras included:
1. A pair of fork terminals, probably for connecting multimeter probes or a load.
2. A 10 A fuse, though just one. I’d have liked at least a couple.
3. A temperature sensor, 10 kΩ resistance, with a 1 m cable.
4. A WiFi module, included only with models bearing the W suffix.


The design is essentially the same as on the previous models: a module intended for panel mounting, the so-called “panel-mount” version. But this one is certainly larger: 166 mm long and 80 mm high. It’s almost white, with a slight gray tint, but looks fine.


Full dimensions.


There are clips on the sides to secure the module in the case.


One of the key new features is the keypad. It lets you set the output parameters directly and access the memory slots. The encoder is still there, and it works just as before: select what you want to adjust, then turn the encoder to set it. The encoder is also push-button, but pressing it essentially works like a “back” or “cancel” button, rather than switching the adjustment range.


A. Power button
B. Shift — additional control modes
C. MEM — access to preset memory slots
D. Set output current (with Shift, set overcurrent protection)
E. Set output voltage (with Shift, set overvoltage protection)
F. microUSB port for connecting to a computer
G. Negative terminal
H. Positive terminal for charging batteries
I. Positive terminal for the main output
J. Turn on output voltage
K. Encoder: adjust current and voltage, select menu options; press to exit the menu
L. Up/down and left/right buttons: navigate the menu and select the adjustment step for current/voltage
M. Enter: apply the selected settings
N. Numeric keypad: enter current/voltage values directly, select memory slots, open the menu, and lock the keypad.
O. Display



The second feature is the larger display. It measures about 50 × 37 mm, or, to put it simply, it’s exactly the size of a matchbox.
My eyesight isn’t great, so in my opinion the larger display is a real plus, although I wouldn’t mind an even bigger one.


Not only are the load terminals now mounted on the case, there are three of them. The outer terminals are for connecting a load when the unit is operating as a power supply; the middle one is for use in charger mode. I’ll show you this feature later.
The clamping nut is plastic, and it doesn’t come off the contact when unscrewed. The second photo shows how far it can be unscrewed. There’s a hole for

There are a lot of components for a model with relatively modest power. In fact, this converter is similar to the 8005 (5005) model, which I reviewed earlier, but that one was much more compact.
The unit is built around two boards: a power board and a control board. The control board also serves as the keypad board.


P, Q — fuses, for the input and output respectively. They’re said to be made by Little Fuse.
R — power input
S — connector for an external temperature sensor.
V — fan connector
U — connector for the WiFi module, and the module itself (T)


Power is supplied through a pluggable terminal block. The input has a common-mode choke and a pair of 100 nF capacitors. Above it, you can see one of the fuse holders; there are two in total, one on the converter input and one on the output. It’s nice that the fuses aren’t soldered to the board. I’ve already managed to blow a fuse in one of the converters.

There’s a connector for a temperature sensor next to the input terminal block, and I think this is a design flaw. No, the connector itself is certainly handy. A temperature sensor is useful for measuring battery temperature while charging, and is useful in general, but the sensor is external while the connector is inside. That doesn’t make much sense.


The converter that powers the control circuitry is based on the XL7015, which supports an input voltage of up to 80 V.


1. The temperature sensor is next to the heatsink. The fan turns on in response to the sensor, as well as when the output power exceeds a certain level. You can also see a power resistor here, rated at 5 W or more by the looks of it. I’m not sure what it does yet, but it gets noticeably hot during operation.
2. Another voltage regulator, presumably for powering the “brains”—the microcontroller and its associated circuitry.
3. The main converter itself is controlled by the rather old TL594C PWM controller. This manufacturer is quite conservative in that respect; some of its previous models used the same approach.
4. There are two current-sense shunts, each rated at 30 mΩ, connected in parallel. Their voltage is fed to the GS8332 op-amp. 350 kHz Zero-Drift CMOS Rail-to-Rail IO Opamp with RF Filter.
5. The converter’s input and output capacitors are rated at 100 V and 330 µF each. They’re budget components, of course, but with plenty of voltage headroom. The second fuse is visible to the left and below.
6. The converter’s power inductor; the relay to the right controls power to the battery connection output. Overall, it's a pretty decent solution, but personally I'd prefer electronic switching alongside software control of the output voltage, since a relay at these voltages can cause problems in certain situations.


Control is handled by a fairly powerful STM32F103 microcontroller. To its right is a 25Q32 flash memory chip, presumably for storing the firmware. There's also a second op-amp nearby.
A 512-byte FRAM MB85RC04V chip is installed to the left.


More chips are visible deeper inside. I suspect they're DACs/ADCs, but taking the whole thing apart proved extremely difficult, both because of the many clips and because the three power terminals pass through the lower board and are soldered to the upper one. I still plan to take it all apart, but that'll be in part two.
For now, I can only show a photo from the product page, and it really does look like an actual photo rather than a 3D model.


The converter has a real-time clock. It doesn't really need one, but it's still best to install a battery. Since one isn't included, I picked one up at the market. It uses a CR1220; I'd probably have gone with the more common 2032, especially since there's room for it. I didn't buy the battery right away, though, so some of the photos below show a random time and date.


The WiFi module plugs into its dedicated connector and is held in place by one of the case clips—a pretty thoughtful design.


First power-up: nothing burned out, and the converter started automatically as soon as power was applied. The power button essentially puts it into standby; the primary power supply keeps running. In operating mode, the button stays lit. In standby, its backlight slowly brightens and dims.
A splash screen appears at startup. If you find it annoying, you can turn it off.

I've already been asked how good the screen's viewing angles are. I can confidently say they're excellent: the readings are clear from any angle, whether viewed from the side or from above or below. The image doesn't invert and always remains sharp and high-contrast.


The specs list an operating voltage range of 6–70 V. In reality, it starts working at 5–5.2 V and keeps running as the voltage drops to 4–4.2 V, but the readings start to drift as the reference voltage shifts.
At an input voltage of 9 V, current consumption is around 90–100 mA; at 62 V, it drops to 17–20 mA.


The screen is informative. It shows not only the main parameters—output voltage, current, and power—but also additional ones: input voltage, set voltage/current, and the voltage/current cutoff thresholds below. The thresholds act as a trigger-based protection: if the measured values exceed the set limits, the output is shut off. The main setpoints work in the usual CC/CV mode.
At the very top are the current time, the sound notification setting, and the computer connection mode.
At the bottom are the active preset memory slot number, CC or CV operating mode, the protection trigger indicator, and battery operating mode. This line also cycles through capacity in Ah, energy in Wh, and temperature. If no sensor is connected, dashes are displayed. Temperature is shown in both Celsius and Fahrenheit.


W - current date and time; works correctly only when a battery is installed.
X - sound notifications on/off
Y - keypad lock
Z - connection type: USB/WiFi/UART
AA, AB, AC - output voltage, current, and power
AM - input voltage
AL, AK - set output voltage and current
AJ, AI - voltage and current thresholds at which the output is automatically shut off (trigger-based protection).
AD - active memory slot number
AE - CC/CV operating mode
AF - status: normal, overheating, fault, etc.
AG - battery charging mode; when active, it lights up red.
AH - amp-hours, watt-hours, temperature. The Ah/Wh counters reset when power is switched off and back on.


To enter output parameters directly, press the corresponding I-set or V-set button, enter the desired value, and press Enter. You can enter it however you like—for example, to set the voltage to 60 V, you can enter 60, 60.0, or 60.00; it makes no difference. This is genuinely convenient and implemented correctly.
Pressing Shift first lets you set the overcurrent/overvoltage protection thresholds in the same way, i.e. in trigger mode.


The encoder adjustment, on the other hand, has some issues. When you select parameter adjustment mode, it always starts with the finest step, which you have to change using the left/right buttons. Since you usually need to make coarser adjustments, you end up changing the step size all the time. I’d either have it remember the last selected step, or always start at 1 V and let you switch to a finer or coarser step with the left or right button.
If you set a parameter higher than the board can handle, an error message appears briefly. You can’t set the output voltage higher than the input voltage, but if, for example, the input is 50 V and you set the output to 48 V, then lower the input to 40 V, no error message appears; the output voltage setting simply changes.


I also have a complaint about the temperature display. For testing, I connected a high-power resistor to the output and inserted the temperature sensor into it. Once the temperature passed 99 degrees, it displayed 00, 01, 02 instead of 100, 101, 102. I think the manufacturer would be better off adding a menu option to choose Celsius or Fahrenheit, rather than displaying both at once.


The Shift button gives you access to the protection trigger settings, memory slots, menu, and keypad lock.


The control menu has four screens:
1. General settings
2. Display options: numeric values or a graph
3. Preset configuration
4. Firmware information


The graph view didn’t really grab me. It’s interesting and looks nice, but only for specific applications; for everyday use, numbers are still more convenient.


There aren’t actually that many settings. They include:
Call ok - confirmation prompt when switching to presets
Call out - automatically activate the output when switching to a preset.
Power on - automatically turn on the output when power is applied
Beeper - turn off the buzzer. I did this right away because it beeps not only when you press the buttons, but also when you turn the encoder. Annoying.
Logo - turn off the logo shown at power-up

1. Language selection: English or Chinese
2, 3. Screen brightness adjustment, from 0 to 5. At 0, the image is almost invisible, so I took the photos at brightness level 1. I immediately set mine to 5; the default was 4.
4, 5, 6. Connection options for a computer: USB (on the front panel), WiFi (using a module), and TTL (using an optional module).
7, 8, 9. Voltage and current update (measurement) speed. The default is slow, with medium and fast options also available. To me, all three modes work equally well, so there’s no real need to change it.


The manufacturer says you can use a UART-RS485 module, and the option to set the device address suggests the same.


Now for the tests. I’ll follow the order used in the 5005 board review, since it has the closest output specifications.

Test 1: checking the accuracy of the input voltage measurement.
A clear 5 out of 5. If the readings differed from the multimeter, it was by just 0.01 V or one count in the last digit.


Test 2: output voltage setting accuracy.
I initially planned to add a table, but after looking at the results, I realized there was no point: the accuracy is excellent, with room to spare.
I set the following values: 0.5, 1.0, 2.0, 3.3, 5.0, 10.0, 15.0, 20.0, 30.0, 35.0, 40.0, 45.0, 50.0, 55.0, and 60 V. In almost every mode, there was no error, even in the last digit. The only exception was at 60 V, where the reading was 59.98.
The board’s own measurement accuracy is also very high: an error of one least-significant digit at 0.5, 20, 35, and 40 V, and two digits at higher voltages.
Bottom line: excellent.


Test 3: output current setting accuracy.
In almost every mode, the results are just as good as for voltage setting—in other words, excellent—except at very low currents. There is some over-reading in the 1–5 mA range, which approaches a minimum at currents of 10 mA and above.
Below are the results for currents of 1, 5, 10, 50, 100, and 500 mA, and 1, 2, 3, 4, 5, and 6.1 A.


Test 4: how well the output voltage holds under load.
I tested this in two modes, with no load and at 6 A, at four voltages: 3.3, 5, 12, and 24 V.
I measured the voltage at the output terminals. The difference between no-load operation and operation at nearly the maximum current was about 5–7 mV, which is excellent in my opinion.


Test 5: minimum input/output voltage difference.
I tested this at 6 A with two input voltage settings. The test was simple: first I set a certain voltage at the input, then set a similar voltage at the converter output. I lowered the input voltage and watched for the point at which the converter's output voltage began to drop—in other words, when it stopped regulating.
1, 2. With an input of 37 V, the minimum difference was 1.0–1.1 V.
3, 4. With an input of 60 V, the difference was higher, about 2 V.

Based on the test results, I can say that for the converter to work properly, the input voltage must be 3–4 V higher than the required output voltage. Put simply, to get the specified 60 V at the output, you need to supply at least 63 V.


Test 6: output voltage ripple.
The manufacturer specifies ripple of up to 100 mV. I tested loads from 0 to 6 A; the results below are for 0, 1.25, 2.5, 3.75, 5, and 6 A, respectively. The input voltage was 62 V and the output voltage was 30 V; ripple is usually close to its maximum in this mode.
The main ripple component reached 95 mV at maximum current and was lower in the other modes. However, there were short spikes at the output—over 160 mV—when the switch changed state. These spikes are very brief and can easily be suppressed with a simple filter or even wires, but they are there. I couldn't get to the power stage circuitry, but I suspect it uses a synchronous rectifier, since the waveform looks very much like the one on the DPS5020.


Here's a second version of the test: the current is still 6 A, and the input voltage is again 62 V, but the output voltage is 3.3, 5.0, 12, 25, 40, and 60 V. The vertical scale is 20 mV per division.


I hadn't run this test before. Here I connected an electronic load operating in CV mode to the converter, so the converter was operating in current-limit mode, in this case at 6.1 A, its maximum. Unfortunately, the load's maximum power is only 150 W, so I tested only up to 24 V.


Results at 5, 10, 15, and 24 V; same scale, 20 mV/div.


Test 7: measuring the board's efficiency.
The input voltage was around 60–62 V in all cases. I monitored the input current with a multimeter; the input voltage and output power are shown on the converter's display. The load current was around 6 A in all cases.
Output voltage – efficiency – power loss
5 V – 73.7% – 11.1 W
12 V – 86.7% – 11 W
24 V – 93.1% – 10.6 W
36 V – 95.4% – 10.3 W
48 V – 96.4% – 10.6 W
59 V – 95.6% – 16.1 W
As you can see, the efficiency is quite respectable. What's interesting is that power losses are roughly the same across almost the entire range, and only start to rise at maximum output power.


After that, I ran the converter at maximum power for a while longer and checked the temperatures of various components. Here's what I found:
1. The power resistor gets the hottest. I showed it at the very beginning of the teardown.
2. The output fuse also gets noticeably hot. I expect the input fuse does too, but its heating is masked by the high temperature of the power resistor.
3. Most of the other components barely get warm.


A few words about the battery charging function.
You can charge a battery the same way as before, by connecting it to the converter's output terminals, but it's better to connect it as the manufacturer intended—that is, to the separate terminal.


In this configuration, battery charging mode is activated, and the battery voltage is shown in the lower field.
The difference from normal mode is that this output is connected through a relay, which switches off when the charging current drops below 10 mA.
And here's my immediate concern: 10 mA is too low a charge termination current for some types of batteries. In my opinion, it would be better either to add a menu option to set it, or to set it to 1/10 of the specified charging current. But in any case, progress is clear, and this is a useful feature.


Software.
I downloaded the program from the product page via this link. To get ahead of myself, the link includes Windows software and a well-translated Russian installation guide for fitting it into the enclosure, but there’s no Android app, so I can’t test the WiFi connection. I think this is only temporary and will be fixed by the second part of the review.

I downloaded the program. I already had the CH340 driver installed, so there were no connection problems; the software found the right port automatically.
A useful tip appears when the program starts, but it gets in the way more than it helps. My antivirus also flagged the program.


The program did start, connect, and display information about the converter: its model name, firmware version, serial number, and even the temperature from the built-in temperature sensor.
There’s nothing unusual here; it’s all quite simple and convenient.


There’s a calibration option, but you need a password to access it. I think it’s 168168, but I could be wrong. I don’t even want to poke around in there.


You can update the board’s firmware. In my case, the latest version was already installed.


There’s also an unusual option: you can set your own logo. You need a 320x240-pixel image file.


During the upload, the converter’s screen shows the progress, and when it’s done, the startup logo appears—now with my image.
The image isn’t great, but the bigger problem is that I can’t restore the old logo now. I’ll have to ask the manufacturer to make that one available too.


As with the previous models, an enclosure is available for this one too, and it’s really big.


There are two enclosure versions, short and long, but the manager said they don’t plan to promote this model with the short enclosure.
The long one costs $32.5, the short one $21.5 — link.

Personally, I think something in between would be a better fit. At first, I was thinking of getting the short one and building my own power supply to fit the enclosure. But on the other hand, since the manufacturer plans to release more powerful models, it probably makes sense to get an enclosure with room to grow.


Dimensions of the large enclosure.


The design is simple: two U-shaped pieces of metal, about 0.9 mm thick by the look of it. The enclosure itself is very rigid.
A bag of fasteners and other small parts is taped inside.


The price difference between the enclosures is due not only to their size, but also to what’s included. The large one comes with a fan control board and the fan itself, as well as more fasteners.


There are ventilation holes on the bottom and sides, and several sets of holes on the bottom for mounting the primary power supply.


Assembly is very simple. First, install the mains inlet (which includes a fuse), the power switch, and the fan.


Then you can install the feet. They’re rubber, by the way, with a metal washer inside to keep the screw from pushing through. The power supply doesn’t slide around on the table.


And here’s the important difference: the fan control board. On the older models, it had only a power converter; this one also has a temperature sensor.


Power is handled by the XL7005A, with an output current of up to 0.5 A and an input voltage of up to 80 V.
A microcontroller controls the fan. When power is applied, the fan runs at full speed for a few seconds, then its speed depends on the temperature:
40℃–50℃: low speed
50℃–60℃: medium speed
Above 60℃: full speed


The board mounts on the standoffs provided for it and can only be installed in one orientation. Unfortunately, with such a large enclosure, installing the board takes up a lot of useful space, and if you install a wide power supply, you’ll have to figure out where to put it.


All connections are made without soldering, so four wires with terminals are included.


The next step is to install the converter itself. Its opening is a very tight fit; you can’t even insert it at a slight angle. It has to go in parallel to the front panel.


The connection here is also solderless: a wire with terminals is included, and it connects to the converter via its built-in terminal block.


Next, I wanted to see what power supply could be installed here. It turned out that with passively cooled units, there’s no way to mount one properly in the perforated enclosure without drilling holes for the fasteners.
Power supplies rated at around 360–480 W, on the other hand, fit perfectly—their mounting holes line up with the holes in the enclosure.


One minor point: the power supply casing sits close to the switch terminals. For added safety, you can either bend this part of the PSU casing out of the way or add insulation.
Also, the power supply mounting screws are quite long and press against the PSU circuit board. There’s usually a sheet of plastic in between, of course, but I wouldn’t take any chances—I’d add some plastic or fiberglass washers between the enclosure and the power supply.


The connections are very straightforward. Everything is also explained in great detail in the manual, which is written in proper Russian—special thanks to the manufacturer for that.


For now, I’ve temporarily installed the PSU shown above, but I ordered a more powerful one some time ago. I’ll show it in the second part of the review.
It looks pretty good inside. I plan to remove its cover and take out the fan; the enclosure fan will do the job instead.


The unit fully assembled.


Compared with my two main power supplies.
1. ZXY6005, 62 V, 5.1 A, 315 W—in other words, almost the same as the unit under review, but clearly more compact.
2. ZXY6020, 62 V, 20 A, 720 W—twice as powerful as the unit under review, but noticeably larger. The reviewed unit’s enclosure is also a little longer, which isn’t always convenient.


Conclusions.
The specifications match the claims. I’d also highlight the very convenient controls, especially direct parameter entry, and the lack of encoder “slippage”: even when turned quickly, it adjusts the output parameters without any trouble and responds very well. The screen is large, the feature set is decent, and it can charge batteries properly instead of simply switching to CV mode and staying there until you turn it off yourself. I have to praise the excellent voltage and current setting and regulation accuracy, as well as the good efficiency. I was particularly impressed by the calibration accuracy, which is why I think it uses separate DACs and ADCs.
The output ripple is within the limits specified by the manufacturer, but personally, I’d like to see it lower—by a factor of 5, say. For example, the DPS8005 has noticeably less ripple, despite having comparable power specifications.
They even improved the enclosure fan control board: now the fan can not only switch off, but also change speed, albeit in discrete steps.
Compared with the previous models, the new one is technically very similar. In some places, it even uses the same circuit designs and components, based on what I’ve looked at. But in terms of features and ease of use, the new model is in a league of its own—and that’s no exaggeration.

Still, there are a few shortcomings and “growing pains.”
1. The connector for the external temperature sensor is inside the enclosure, which doesn’t make much sense. It should be on the outside.
2. The external sensor can monitor the temperature of, for example, a battery being charged, but there’s no option—or at least I couldn’t find one—to set a temperature threshold for emergency shutdown.
3. Temperatures above 100 degrees are displayed as 00, 01, 02, and so on. I think it would be better to move the Celsius/Fahrenheit selection to the menu and display only one unit at a time. That would free up space and make the readings clearer.
4. When you switch to adjusting current or voltage, the minimum adjustment step is selected by default. That’s inconvenient. I’d prefer either a 1 V or 100 mA step, or a memory for the last-used setting.
5. Standby control. I’d like to have a logic output for controlling the main power supply’s on/off state. Then you could use a small PSU to power the controller, which would start the high-power PSU in normal mode when needed.
6. Charge termination current. Here it is set to 10 mA. It would be better either to make it adjustable or set it to 1/10 of the charging current, but I prefer the first option.

In short, I really liked the converter, and I can already guess the question: would I buy it? The answer is an unequivocal yes.
At the moment, I'm waiting for the manufacturer to release more powerful models, rated for 10 or 20 A. Then it might be worth considering designing a relatively simple primary power supply rated at around 800–1000 W and fitting the whole thing into the enclosure shown in the review. I'd really like to see a model with an output voltage of 80–90 V and the same 10 or 20 A, but I doubt that's likely.



That's all from me. I hope you found the information useful. As usual, I'm looking forward to your questions. I may be able to answer some of them now; if not, the answers will most likely be in part two.