RD6018, a new converter from RuiDeng

Just under four months ago, I published a review of RuiDeng's new RD6012 converter. Now I have the next model in this lineup in my hands: the RD6018, with an output voltage of up to 60 V and a current of up to 18 A.
This review will be relatively short, but I'll still try to make it as informative as possible.


So, today I'll try to keep it “short and to the point,” since it turns out there aren't that many differences from the previous model. That was to be expected, of course, and I already mentioned it in the announcement. For obvious reasons, I also don't see any point in repeating the information about the controls for a third time, and I won't cover installation in an enclosure either.

The packaging is the familiar kind: a large polystyrene box, with the converter packed inside a separate box made of thick cardboard. As before, I'll emphasize that the packaging deserves top marks.
The bottom of the converter's box lists brief specifications and the necessary links.


The store page has a handy table listing all six models, although there are actually only three models; the other three differ only in that they include a WiFi module. Even here, you can see that the specifications of the 6012 and 6018 differ very little.



What's included
1. Converter module
2. WiFi module (optional)
3. Temperature sensor
4. Fuse
5. Fork terminals.

As before, the battery is not included.


I don't see any point in going over the optional accessories. The only difference from the 6012 version is that its spare fuse was rated at 20 A, while this one is 25 A.


Back to the converter: the enclosure is the same panel as before. Since the dimensions haven't changed, you can upgrade simply by swapping one converter for the other. However, this also requires replacing the power supply, and in practice that doesn't always make sense, since an output current of 18 A is needed more often than the full 1080 W.


The display, buttons, controls, and functions are almost 100% identical to those of the 6012, although there's at least one small difference, which I'll mention later.


The three terminals for connecting a load are still there, too. In the announcement, I complained that the manufacturer hadn't replaced them with higher-current ones. But judging by the information I found on sellers' pages for these terminals, their maximum current is 30 A, so the 6024 version will most likely use the same ones as this model.
Of course, it would be interesting to see something more robust, but terminals like that are usually a different size, which would mean redesigning the front panel and the molds for the injection-molding machines. So I don't think anything will change.


Let's take a closer look at the circuit board. Since it's largely similar to the 6012 version, I'll focus mostly on the differences.


1. The input common-mode choke is indeed large, and judging by the wire gauge, it's clearly designed for high current.
2. The input reverse-polarity protection transistor is the same as in the 6012: 110N8F
3. The input fuse is soldered in place. There's a footprint next to it for installing a replacement, but if you do replace it, I'd recommend soldering the new one in as well. Above the fuse, you can see the socket for connecting the temperature sensor.
4. There are also two input capacitors, but while the 6012 had 330 µF capacitors, this one has 470 µF ones.


1. The output capacitance has also been increased. The 6012 had three 220 µF capacitors plus two 330 µF ones. Here, the first group is still three 220 µF capacitors, while the second now has two 470 µF capacitors.
2. The output fuse, which protects the converter if a battery is connected with reversed polarity, is rated at 25 A, just like the input fuse.
3. Three 15 mΩ current-sense shunts connected in parallel, for a total of 5 mΩ. That's a little odd, since the 6015 had two 10 mΩ shunts, also totaling 5 mΩ. I expected the manufacturer to simply add another 10 mΩ resistor in parallel here, bringing the resistance down to 3.3 mΩ.
4. The charge cutoff relay is clearly larger than the one in the previous model: the 6006 had a 10 A relay, the 6012 a 16 A one, and this one has a 30 A relay. Something tells me the 6024 will use one too.


Another change is the converter's output inductor. In my review of the 6012, I was surprised to find that it was the same as the one in the 6006, even though it also determines the output power. This one is clearly larger and wound with two thicker wires. I'll compare it with the previous models later.


The auxiliary voltage converters and regulators are unchanged, apart from a slight rearrangement of the components. The resettable fuse is still there too.


The PWM controller and synchronous rectifier section, on the other hand, are exactly as before. I couldn't spot any differences in the PCB layout. By the way, when you receive it, I recommend checking that the temperature sensor is pressed firmly against the heatsink.


The converter is still awkward to disassemble. To remove the board, you'll first have to desolder the power sockets, which you can do with a relatively thin, high-power soldering iron.


The front-panel and control board is unchanged.


The same microcontroller, memory, and USB interface chip are used.


A couple of views of the power board outside the case.



1. The underside is almost bare, with just a few components.
2. Somewhat unexpectedly, the converter's power transistors are still the same N-channel HY3210s, rated for up to 120 A and 100 V, with an on-resistance of 6.8 mΩ.
3, 4. There is one difference, though. As far as I can tell, it has nothing to do with the increased power and is more about improving reliability: the auxiliary voltage converter chip is cooled by a small heatsink glued to the underside of the PCB. The 6012 didn't have one. But looking at a photo of the 6012 PCB, I noticed that the area for the heatsink is outlined there too, so it seems it was planned from the start.


And now for a few comparison photos. First, the exterior of the 6006, 6012, and 6018. As you can see, the only difference between them is the model name.


The internals differ much more. The 6006's power board is shorter, so part of the control board is visible. The 6012 and 6018 have power boards of the same size.


The cooling system is identical. It's worth noting that the fan thresholds are also the same for the 6012 and 6018: it turns on at 8 A and off at 7.8–7.9 A, to prevent it from constantly switching on and off around the current threshold. The fan also runs when the system temperature exceeds 45 °C.


Here's the inductor I mentioned earlier. The previous models' inductors are not only smaller, but identical in the 6006 and 6012. In all versions, the converter's heatsink is positioned so that this inductor gets some airflow too.


For my first test, I connected the converter to my old power supply, built around a ZXY6005 board.


A little about the menu and settings.
1. On startup, we see the familiar screen. The only difference from the previous versions is the maximum current limit of 18.1 A.
2–4. The settings menu is very similar too. I immediately turned off the awful buzzer, which is just as irritating as the one in the lower-power models, and set the brightness to maximum.
5, 6. An important difference: this model has an option to set an output power limit. This was probably added because power supplies with this much power are much less common, and you don't always need the full output. You may only need up to 18 A or 60 V, while 500–800 W is enough for power, as in my case. The maximum setting is 1110 W and the minimum is 10 W. The rated power is listed as 1080 W, while 1110 W comes from the maximum current of 18.1 A and maximum voltage of 61 V: 61 × 18.1 = 1104 W, plus a little headroom. In my opinion, this feature would have been useful on the previous versions too.
7–9. The operating screen and memory functions work just as before. The latest firmware version at the time of this review is 1.33.


I'll spend a little more time describing the power limit function. It works very simply.
1, 2. If you set the limit to, say, 10 or 20 W, you can adjust the voltage across its full range, but the current can only be set to a level that won’t exceed the power limit—not the other way around. In other words, only the current is limited.
3–5. For example, with the power limit set to 20 W, raise the output voltage and the current setting will automatically decrease as the voltage rises.
6. The function works even when the output is inactive, though I should add a small caveat here. Overall, it works great and is genuinely convenient to use, but sometimes it would be useful to be able to set everything to maximum while still limiting the power. I’d like the current to start being limited during operation when the power limit is exceeded. This could be useful when charging batteries: the voltage is low at first, but the current can be high; as the voltage rises, so does the power, and once it reaches the limit, the current would then be reduced. But now I’m just nitpicking.



I ran into some difficulties powering the converter. I have a power supply from telecom equipment that one of my readers sent me, but it can provide a maximum of 55 V. With it, I could get up to 50 V at 18 A on the output, or 900 W total, which isn’t much more than the 750 W available with the power supply from this review. In the end, I simply removed the 6012 converter from the finished unit and installed the one under review in its place. Thanks to the standardized design, the swap is extremely simple: disconnect the power input connector and temperature sensor, remove the converter, install the new one, and reconnect everything. The longest part was unscrewing the screws holding the case cover in place.
I set the power limit to 750 W to avoid burning out the primary power supply.


A few tests. To make comparisons easier, I’ll run them in the same order as in my review of the 6012 model.

Test one: input voltage measurement accuracy. This is actually the least critical test, and in general, measuring the input voltage is probably only useful when powering the converter from a conventional transformer-based power supply, or when several switching power supplies are connected in series and you want to check whether one of them has gone into protection mode.

At low voltages, the meter reads slightly high, but above 30 V everything is great: the readings match the multimeter exactly. At an input of 60 V, though, the converter’s meter read 30 mV low.


Test two: output voltage setting accuracy and accuracy of the built-in meter.
During the test, I set the voltage to 1, 2, 5, 10, 20, 30, 40, 50, and 61 V in sequence. In every mode, the setting and measurement accuracy were excellent, exceeding the resolution of the converter’s voltmeter.


Test three: output current setting and measurement accuracy.
I set the output current in sequence to 10, 20, 50, 100, and 200 mA; 500 mA; and 1, 2, 5, 7.5, 10, 12, 15, and 18.1 A.
Surprisingly, the converter did quite well even at low currents, especially considering that I set it to the lowest values it supports: 10 and 20 mA.
The error is naturally higher here than when setting the voltage, but it’s still pretty good. To check currents above 12 A, I had to dig out my ancient, nearly indestructible M890G. For comparison, I measured 12 A with both multimeters, and you can see that the difference from the Uni-T isn’t all that large. The only problem with the M890G is that its shunt heats up quickly at high currents, causing the readings to drift, so I had to pause and let the shunt cool down.


Test four: output voltage regulation across a load current range of 50 mA–18 A at 3.3, 12, 24, and 36 V.
Again, no complaints. The difference across the full load current range was just 3–5 mV. Voltage was measured at the converter’s output terminals.


Test five: checking the minimum required input-to-output voltage difference at 10 A, with input voltages of 36 and 57 V. I also checked the difference at an input voltage of 36 V and an output current of 18 A.
1, 2. I found that at 36 V and 10 A, the difference was 3 V; at 57 V, it was already 4.5 V.
3, 4. With a 36 V input and 18 A current, the difference was 3.25 V, not much higher than at 10 A. On the left is the maximum value at which regulation still works; on the right, the output voltage is already limited.


Test six: measuring output ripple amplitude in different operating modes.
The output is set to 30 V, with an input of 70 V.

1. No load
2. At 6 A
3. At 12 A
4. At 18 A

At 18 A, the ripple reaches 200 mV, which is less than the specified 250 mV at 6 A, but there are spikes in the form of “needles.”



I decided to run the next two tests with a pair of capacitors, 1 + 0.1 µF, connected in parallel with the oscilloscope probe. I once saw a mention that such a small capacitance couldn’t affect the test results. Well, there is a significant difference: the “needles” are completely clipped. The same thing should happen at the load input, since the wires from the power supply and the load’s input capacitance will act as a filter.


Test under the same conditions as the previous one.


Here, the load current is 18 A in all cases, while the output voltage is 5, 12, 24, and 36 V.


Test seven: transient response.
1, 2. Applying and removing 5 V at the converter output with no load.
3, 4. Response to connecting/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: voltage set to 15 V, load current 4.5 A, current limit set to 3 A.
6. Same load and settings, but in reverse: transition from CC to CV.

In my review of the 6012 model, I complained about the odd transition from CC to CV mode. There’s no such problem here; it works correctly.


As a supplement, here are a couple of oscilloscope traces with a slower time base.
1. Output voltage being switched off with no load. You can see the “fuzz” from the converter switching, followed by a flat line as the output is powered only by the capacitors, and then a gradual drop.
2. Overshoot when the load is reduced from 4 to 0.4 A. In my opinion, the converter handles this transition quite well. The output was 12 V, and the overshoot was 2 V, but it’s barely noticeable even at 500 ns per division.


“Soldering iron test.” A TS100 soldering iron is connected as the load to check how the power supply responds to continuous changes in load current.
Overall, the results are a little better than with the 6012, most likely because of the higher output capacitor capacitance.


Test eight: measuring converter efficiency.
The input voltage was 70 V. A multimeter was connected in series with the power supply, so as the load increased, the voltage at the converter input dropped slightly due to voltage drop in the wires. I monitored it using the converter’s own voltmeter.

The test was run at a load current of 18 A and output voltages of 5, 12, 24, and 36 V.
Output voltage — efficiency — power loss (results for the 6012 model at 12 A)
5 V — 83.3% — 18.07 W (80.6% — 14.47 W)
12 V — 92.0% — 18.6 W (90.7% — 14.7 W)
24 V — 95.8% — 19.13 W (94.8% — 15.57 W)
36 V — 97.2% — 18.12 W (96.8% — 14.36 W)

As expected, the losses are higher than with the 6012, but overall efficiency is slightly higher, thanks to the more powerful output inductor with lower winding resistance.



Battery charging mode works just like on the 6012 model, but with charging current up to 18 A. Charging stops when the current drops below 100 mA. And once again, I wish there were an option to adjust the cutoff current, because 100 mA is quite a lot when charging at 18 A.



Software.
First, the software has been updated. If you downloaded an older version, a notification with a link will pop up. I didn’t notice any major changes. They say support for the 6018 module has been added, the connection process has been optimized, and the interface has been tweaked a little.



Not long ago, one of my readers got in touch to complain that the software couldn’t plot battery charging graphs properly: it showed only a section of the cycle, not the whole thing. I don’t remember how it worked before. I think the graph really did just keep “scrolling,” so you couldn’t view the entire cycle. Since I was already tinkering with the new power supply, I decided to check this again while I was at it.
It turns out that, at least in the new software, the graph is plotted correctly, from zero to the current point. If you right-click the graph, you’ll see some additional options:
1. Reset values, so the graph starts counting from the moment you reset it.
2. Save the graph as an image. Unfortunately, it just crops out a section of the screen, so the graph remains tiny, and the display selection menu covers a good chunk of it.
3. Saving to an Excel file


I exported the data to a file; this is what its contents look like. It's strange they didn't include CSV export as well. Ideally, they would also make software for plotting graphs from this data, like ZKEtech loads, but “that's science fiction.”


I also got some questions about connecting to the module over Wi-Fi.
It's not entirely straightforward: sometimes connecting feels like a ritual, and sometimes everything connects “at the drop of a hat.”
First, the power supply and the computer with Wi-Fi need to be within range of the same access point, with a stable connection.
Then launch the software and turn on the converter (red dashes will appear where the IP address should be on its screen). Select WiFi, then WiFi Network. If everything is OK, the IP address of the computer running the software will appear. If not, click Initalization (the same address will appear on the converter's screen). The Next button will then light up; click it.


A window will appear where you can enter your access point's name and password. There's also a field for the module's network number; it must be the same in both the software and the module. Naturally, WiFi must be enabled on the module. Then click Distribution.
If everything went smoothly, close the window. If not, try again from the beginning. If the IP address was saved in the power supply but nothing works for some reason, you can reset it by pressing the left arrow on the front panel and then Enter.


This takes us back to the software's main screen. The module's IP address is shown on the left; it's just there for reference.


After you click Connect, the module should connect to the software.
It all seems fine and even works, but there are a lot of limitations. For example, the software must be running when you turn on the power supply, or the power supply won't establish its initial connection to the software.
Even though the power supply's IP address is known, you can't connect to it directly. In other words, it won't work if the computer is connected by cable and the module over WiFi.
It's odd that the software doesn't have an address field, so you could connect without going through the whole Wi-Fi → access point → WiFi → software chain.


I also tried connecting over WiFi using a smartphone. The initial setup went perfectly: the power supply entered connection standby mode and the connection icon turned white, but when I tried to connect, the icon on the power supply briefly turned yellow (on the right) and then went white again.


It turned out I just needed to update the software on my smartphone. After downloading and installing it, a brief connection guide even appeared, which was very handy.


In the software, it looks like this:
1. Launch the app, open the settings menu in the upper left, then select Distribution network.
2. The software detects the connected access point automatically; you only need to enter its password. Note that the access point must operate at 2.4 GHz.
3. Start initialization.
4. When your phone's IP address appears on the converter's screen (the digits will be yellow), check the box confirming that the device displays it, then tap “confirm.”


These steps look like this on the converter's screen:
1. Waiting for the software to search.
2. The software detects your converter and waits for you to enter your access point's password.
3. Everything has been entered correctly; the software detects the access point and the app. This is the window you'll see when you turn on the power supply.
4. If the software isn't running on your smartphone or PC, an error will appear at step three. To connect, you'll need to launch the software and restart the converter.


At the final setup step, the software will report that everything was successful and display the converter's IP address.
After that, it's basically the same as on a PC: tap Connect at the top of the screen, the software connects to the converter, and you're ready to go.



Conclusions
There isn't much to say in conclusion. The converter performed very well in use. I had a few minor complaints about current-setting accuracy, but that's just nitpicking. Overall, this is simply a higher-power version of the 6012 model, with the same voltage-setting resolution and operating modes. The output current and maximum power have been increased, and a power limit function has been added so it can be used with lower-power power supplies.
There haven't been many hardware changes either: the filter and output inductors have been upgraded, as has the capacitor capacity; there are now three current-sense shunts instead of two; the relay has been replaced; a heatsink has been added for the auxiliary converter; and the fuses are now rated at 25 A instead of 20 A.

Overall, I'm glad the manufacturer managed to increase the current by 50% without making anything worse, and even improved the efficiency a little.
I didn't test the thermal performance during my tests because I couldn't provide the full rated power, but I did run an indirect test. The thing is, our heating still hasn't been turned on, and it's already noticeably cold outside. So I did what any radio amateur would do in that situation: turned on the power supply, connected two loads, set the output to 40 V, and drew 18 A. The result was a heater rated at about 700 W, and I used it to warm up for a couple of hours. Nothing failed, and the room got noticeably warmer :)

In short, if you were planning to buy the 6012 but are willing to pay a little extra, I'd recommend taking a look at the 6018. I haven't found anything particularly concerning about the new model. Essentially, all my complaints come down to the clumsy implementation of Wi-Fi: why not allow a direct connection using an IP address on the local network? It's hard to understand.