ZUBR-D2-63 Voltage Relay from DS Electronics

As usual, a reminder: to keep up with new posts on the blog, I recommend following me on Instagram, where I'll post notifications about every new topic and maybe write about something interesting from time to time — link to my account


Some time ago, we got to talking about manufacturers of protection devices, and I said that I consider products from three companies to be high quality: Novatek-Electro, Meandr, and Zubr. Meandr products aren't readily available to me, and I've shown Novatek products several times, so now it's the turn of another Ukrainian manufacturer, DS Electronics, which makes products under the ZUBR trademark.


Voltage protection relays that disconnect the load when the voltage goes outside set limits seem to be more common in the CIS countries. As a result, the market has a few types of products:
1. Relays from large companies, usually reliable, but often with limited functionality or a higher price.
2. Relays from well-known local manufacturers: Zubr, Novatek, Meandr, and DigiTop. I have some doubts about the last one, both as to its country of origin and its overall quality.
3. And quite a few Chinese-made options, ranging from perfectly decent to rather mediocre.

There are also relays from less common manufacturers, such as the Kharkiv-based DigiCop and Adecs, the presumably Russian HS Electro, AKIP-DON, and so on.

Local devices are generally noticeably better made than most products "from AliExpress," yet offer more features than comparable products from established brands at the same price. That makes them very popular.


A brief note before I start. Try as I might, this review will inevitably turn into a comparison with the Novatek RN-263t, so I'll do my best to keep it as objective as possible. Also, although these devices are very similar, the RN-263t differs in functionality: it can also measure current and therefore work as an overcurrent relay. So a direct, side-by-side comparison wouldn't be entirely fair.


I first came across DS Electronics products several years ago, when I was doing the electrical work in my daughter's apartment. I bought the previous model, the Dt-63, and would have said everything was great if not for one issue. Its default trip threshold was set too low, which caused frequent nuisance trips. The building was new, so the wiring was too, but it wasn't fully occupied yet. During the day, the voltage could easily reach 240 V, and spikes would trip the relay, so I had to raise the upper threshold.


DS Electronics manufactures its products exclusively in Ukraine, but sells them under two trademarks: ZUBR in Ukraine and RBUZ in Russia. Novatek, by contrast, uses one trademark and product naming scheme, but manufactures in both Ukraine and Russia. I like the ZUBR/RBUZ approach a little more, since product quality stays consistent no matter where you buy it.

The link in the title goes to the Ukrainian Rozetka store, but just in case, here's a link to the manufacturer's website, as well as its Russian mirror site.


The ZUBR-D2 series comes with two display colors (red and white) and three maximum current ratings (40, 50, and 63 A). This review covers two 63 A models, which differ only in the display color. They're packed in cardboard boxes labeled with the model, display color, 5-year warranty period, and manufacturing address.


The package also includes a manual and a warranty card.


The manual covers the entire series and is quite detailed, especially since there are many adjustable parameters, all configured through the menu.
As you'd expect, each device has a serial number linked to its warranty card. I'm not sure what happens if you lose the card; I assume they'll simply calculate the warranty period from the date of manufacture.


For some reason, the manufacturer's website only has the manual in Ukrainian, so I had to find an English version to make it accessible to everyone. By the way, the Ukrainian version of the manual is better made.
As usual, I'll quote excerpts from it throughout the review, so sometimes the name will be RBUZ rather than ZUBR.

The technical specifications are very brief, though to be fair, the simplified version will be enough for many users.


But let's move on to the device itself. I have to say, it looks unusual, at least: the front panel has no familiar little indicator window.


There are terminals at the top and bottom. The specified maximum wire cross-section is 16 mm², which, given the stated current, is suitable even for wires with aluminum conductors, let alone copper ones.
The side has laser-etched markings with the device name, brief technical specifications, the manufacturer's address, and the product's serial number.
There's a clip underneath for mounting on a DIN rail. I read a couple of complaints online that it was flimsy, but I can't confirm that; it held securely for me.

The plastic is a little unusual: it's glossy, whereas modular equipment is usually matte. It's not a big deal, though. No concerns about its strength, either, and the plastic is flame-retardant, so that's all good.


The front panel has a three-digit, seven-segment display, which you can't see yet, an LED, and four control buttons.


What the control buttons do.


The case is two standard modules wide and has a trapezoidal cross-section, leaving a small gap at the top. It's not a problem, just a little unusual.


All right, that's it for the exterior. Let's take a look inside.
The two halves of the case are held together by clips and a small self-tapping screw hidden under the mounting clip. Once the bottom half is removed, the neutral busbar and the relay's power contacts are immediately visible.


The circuit board is double-sided, with components mounted on both sides. It looks clean and neat, not just around the SMD components but also around the through-hole parts, which are usually soldered by hand.
A thermistor is visible near the center of the board; it monitors for overheating. It measures the temperature inside the relay's case, and is positioned fairly close to the relay contacts, though on the other side of the board.


1. The power supply is built around a Viper06H PWM controller, which has a high-voltage transistor built in. Here, though, it drives an inductor rather than a transformer, since galvanic isolation isn't needed.
2. There's also a second DC-DC converter on the board, which steps 24 V down to 3.3 V to power the microcontroller. It uses the rather old 34063 PWM controller.

I have to give the manufacturer credit here. Personally, I prefer circuits with switching converters over ones based on a capacitive dropper, though that's just my subjective opinion.


The other side of the board has some fairly large components, including a polarized relay, NRL08B-24D. This relay series is rated for a continuous current of 60 A and a short-term current of 80 A, contact resistance of up to 2 mΩ, and switching times of up to 15 ms.


A couple of words about the relay and its stated specifications.
The datasheet lists the relay model, maximum load power, and load current.


But to keep the description simple, the manufacturer doesn't mention anywhere that these specifications apply only to resistive loads. They can differ significantly with inductive, capacitive, or combined loads.
This isn't stated in the text, either; there's only a brief note, which also doesn't apply to all load types.


The assembly is fairly compact; the display and control board is soldered to the power board.


But I was curious to trace out the schematic, so I had to desolder the assembly.


I had to remove the relay as well.


The input has two 2.2 µF, 400 V capacitors from Suntan connected in series; the low-voltage circuit has a pair of Jamicon 100 µF, 35 V capacitors.
I should point out that the device input has not only a 680 V varistor (or 485 V RMS), but also a proper fuse. That definitely earns the developers at DS Electronics a plus from me.


The additional board carries the display, control buttons, and an STM32F031 microcontroller.


This was clearly soldered on an automated production line too; everything is very neat.


I took the relay apart simply because it was easy to do: the cover isn't glued on, but held in place by clips.
Inside is a conventional power latching relay. You can even switch it manually between its two states.
The contacts are substantial, and the spring section is reinforced as well.


The relay coil resistance is about 190 Ω, which at a 24 V supply gives a coil current of about 123 mA. Power switching is handled by PDTS143XT “digital” transistors, rated for a maximum current of 100 mA. Sure, they only operate for very short periods here and may work just fine, but I would have used something beefier.


I drew up the schematic just out of curiosity. When most of it was done, I remembered that my colleague ksiman had already done the same in his review, so many thanks to him. I did find what may be a change or an error in the process; it doesn't really matter, and the schematic itself is a little different anyway.
I should also mention that the power board has clearly been substantially redesigned.


Before going over the settings and tests, here's an explanation of the wiring diagrams.
The Ukrainian version describes this in more detail and includes not only the basic schematic but also a wiring diagram.
There are two ways to connect it. In one, the power neutral passes through the relay; in the other, the neutral terminal is used only to power the relay and for measurement. I think the second option is more reliable, since it means fewer connections in the power circuit.


Let's connect it and give it a try.
The first thing that catches your eye is the display. Yes, the device looks stylish and unusual, and I think this design is cheaper and easier to manufacture than a case with a tinted window. But personally, I don't find it as convenient.
With the flash on, the readings are practically impossible to make out, so I had to take the rest of the photos without it. But that brings another problem: although the plastic lets light through reasonably well, the display looks noticeably less sharp.
In everyday use this isn't critical, since the readings are still legible. The device also has brightness adjustment, which is useful if, for example, it's installed in a panel with a transparent cover.

By the way, subjectively, the white display is easier to read.

I had to measure the power consumption of two devices at once, because when I tried measuring just one, the wattmeter showed zero.


Let's move on to the settings.
To access the settings, press the button with the three horizontal lines. There are six items; each press selects the next one. If you don't press anything, the relay returns to normal operation after 5 seconds.

1. Turn-on delay. This is needed for compressor equipment and to prevent frequent switching on and off when there are problems with the mains. I recommend setting it to at least one minute.
2. Voltmeter reading correction. This is up to you; I would leave it as is.
3. Standard or professional model: it trips as quickly as possible during a fault, or more slowly for a slight overvoltage and quickly for a very high one. The standard option is selected by default. I recommend switching to professional right away; honestly, I don't understand why anyone would need the “standard” mode.
4. Turn-off delay in case of a voltage drop. This prevents false trips when the voltage dips briefly, which is usually harmless to equipment.
5. Turn-on delay mode: measured from when the voltage returns to normal or from when the relay switches off. I don't think most users will notice much difference; the second option is a little faster, while the first is safer (in my opinion).
6. Display brightness. I tried the different modes, but since the display shines through relatively opaque plastic, it will probably usually be left at 100%. But it’s still nice that if the voltage relay is installed in a panel inside the apartment, its high brightness can be annoying.


Here’s what the menu looks like in action. Some aspects reminded me of caller ID menus, but overall it’s quite convenient and intuitive.
The i button has several functions, including:
1. Hold it for 3 seconds until rSt appears, then release it to clear the fault log.
2. Hold it for 6 seconds to display the firmware version; in my case, d2.1.26.A.


And of course, the measurements.

First, as always, I checked the operating voltage range and voltage measurement accuracy. In the first test, both relays switched on at around 37–38 V. They read slightly low, but that’s normal given the specified minimum of 100 V.
At input voltages of 50 and 75 V, the relays also read 2–3 V low, but that’s still acceptable.


But here I noticed something a little odd. During testing, I noticed that the multimeter display was too bright, so I turned down the brightness and decided to repeat the tests. This time, the relay with the red indicator started up later, and at one point both relays’ indicators were noticeably dimmer than during the first run. Interestingly, the readings also changed: they had been low at the beginning before, but now they were high.

In any case, the specified range starts at 100 V, and both units perform excellently there. Accuracy is quite good up to 250–300 V. Above that, the difference starts to grow, reaching 5 V at input voltages of 400–450 V.
In my opinion, that’s quite respectable. My only minor complaint is about the range above 300 V.


The manufacturer specifies a maximum voltage of 420 V, and you’re unlikely to see anything higher on the mains. Still, I decided to run it for 30 minutes at 450 V.
No complaints here about either operation or heating; the maximum temperature was 44 °C on the case where the two relays were next to each other.


By default, the relays have a lower threshold of 198 V and an upper threshold of 242 V. For some reason, I got 196 and 240 V in practice. The photo shows that because of the small difference in readings, one relay trips slightly before the other, so I’ve identified the thresholds correctly.
And this brings me back to where I started the review: false trips. Yes, the normal mains voltage range is formally considered to be 198–242 V (220 ±10%), but brief spikes or daily fluctuations can easily cause the relay to switch off, even though they’re actually harmless to equipment. That’s exactly what happened when I first tried the Dt-63 relay.
In my opinion, narrowing the range like this only hurts the relay’s reviews, because users’ first experience will be brief, unnecessary shutdowns.
I think a range of 180–250 V would be just as safe but more stable. For example, mine is currently set to 175–260.


Here’s the process in action. In this mode, the readings updated about once per second.


For reference, here’s a typical daily voltage graph from my home.


And this is what I got while preparing the review, with the relay lying next to me.
Default settings: 198/242 V, 3-second turn-on delay. With these parameters, your equipment gets an unforgettable experience—especially the refrigerator...
This behavior sometimes lasted half a minute or longer.


Probably one of the most important tests for protection devices is how they behave in different fault conditions and how quickly they trip.

A brief explanation: the device supports two operating modes, Normal and Professional. Normal prioritizes the fastest possible response, while Professional reduces the number of false trips.
However, the descriptions of the modes differ between the two versions of the manual (links above). The difference concerns the description of the response to a voltage drop in Professional mode, so I’ve included both versions.


Normal mode, default settings:
1. Voltage drops to 190 V; switches off after 1 sec
2. Voltage drops to 160 V; switches off after 1 sec
3. Voltage drops to 130 V; switches off after 1 sec
4. Voltage drops to 110 V; switches off after 0.02 sec
5, 6. Voltage rises to 245 V; switches off after 0.03 sec

In most cases, shutdown occurred either as the sine wave crossed zero or during its final quarter-cycle, with random polarity.


Professional mode:
1. Voltage drops to 190 V; switches off after 10 sec
2. Voltage drops to 170 V; switches off after 1 sec
3. Drop to 130 V, disconnect after 0.02 sec
4. Drop to 110 V, disconnect after 0.02 sec
5. Rise to 245 V, disconnect after 0.5 sec
6. Rise to 270 V, disconnect after 0.03 sec

The key difference from the standard response mode is the longer delay: up to 10 sec for a slight drop in voltage and up to 0.5 sec for a rise. The response time for a drop to 130 V has also been reduced from 1 sec to 0.02 sec.
Otherwise, its behavior is no different from the standard mode.


When the voltage returned to normal after either an undervoltage or an overvoltage, the relay usually switched on close to the zero crossing of the sine wave, and with random polarity as well.


The device stores up to 100 fault events. Pressing the i button activates this mode, and the display shows the voltage recorded when the fault occurred, along with the memory slot number, from newest to oldest.
I wouldn't call it particularly convenient, and 100 events is perhaps a bit much, but I suppose it's nice to have, and certainly can't hurt.


As an extra check, I measured the voltage drop across the relay. The multimeter probes were connected directly to the device terminals, with currents of about 32, 63, and 80 A.
The readings were 15.1, 29.2, and 37.4 mV, respectively, corresponding to power losses of 0.48, 1.84, and 3 W. The test isn't entirely valid because the relay was switching a low voltage, but the results are close to what I'd expect. I don't think it's a good idea to exceed 50 A for extended periods.


And finally, a comparison photo of the ZUBR-D2-63 and RN-263t.



Conclusions.
Subjectively, I really liked this relay. In particular, I'd highlight the good circuit design, quality assembly, component selection, and compliance with the stated specifications.

Of course, there are a few “caveats” too, which I'll go into in a little more detail.
Features. There's nothing to add or take away here: it has everything you need, including adjustable thresholds and delay time, as well as some less common features like voltmeter calibration and adjustable display brightness.
Safety. The enclosure is made of flame-retardant plastic, and the terminals are high quality. They effectively serve as the relay contacts, reducing contact resistance. There's also overheat protection.
Controls. On the one hand, all the necessary settings are there, but in my opinion some could easily be removed without reducing the relay's functionality. Still, “more is better.”
Ease of use. The controls are buttons, which is convenient in some ways, but less intuitive than the “knobs” below. For example, if you install a relay for someone, they can call you if there's a problem—even if they're a complete novice—and tell you what position the controls are in. That's harder with buttons, and they can also accidentally change the settings.
Design. It's an attractive and uncommon design, but the digits are a little hazy. A conventional tinted display filter would have been a better choice, although the display is readable without one. Overall, it's a matter of taste.

The only truly significant criticism is the default settings.
1. Trip thresholds: 198/242 V. That's an overly narrow “window,” and our mains voltage clearly doesn't stay within it, so you'll get a near-100% chance of nuisance trips. They're nuisance trips because voltages below 198 V aren't dangerous for the vast majority of equipment, while the upper limit is also too low because the maximum permitted mains voltage here is now not 220 V +10%, but 230 V +10%, i.e. 253 V.
2. The standard response mode is selected instead of the professional mode. Honestly, I'd remove the “standard” mode altogether. In practice, it causes more problems and only serves to confuse users.
3. The recovery time after a fault. This relay's recovery time is 3 seconds, which, combined with points 1 and 2, creates a downright awful problem: the relay starts tripping unnecessarily even when the mains voltage is still within the normal range, resulting in a “light show” with no music. In the end, instead of protecting your equipment, the relay starts destroying it.

Yes, of course you'll say, “But you can configure all of that.” Absolutely, and you should do it right after installation. I recommend setting the lower threshold to 180 V, the upper threshold to 255 V, the delay to 60–90 sec, and selecting the professional response mode.
But in my opinion, the ideal product is one that needs no configuration after purchase or installation.

And of course, the question is: which relay is better, one from DS Electronics or Novatek-Electro? To me, that's about like asking which bear is tougher, a brown one or a polar bear.
In short, they're roughly equal overall. The Novatek RN-263t offers more features (current monitoring and limiting, and a better fault response mode) and a 10-year warranty. As for the one reviewed here, I liked its circuit design, build quality, and accurate voltage measurement right up to the maximum.
The RN-263t also uses a single response mode, but it's even more sophisticated: it takes three voltage thresholds into account and provides three different disconnection speeds. It also has an important feature—a hysteresis function. This relay either has no hysteresis at all, or very little. And if three levels aren’t that critical, hysteresis is definitely a must.

I also have to point out that after the early version of this relay, the manufacturer completely redesigned the power board and improved the build quality. That means they’re paying attention to feedback and suggestions—kudos to them.

P.S. After I sent the developers a bug report, they replied that they plan to add hysteresis. They’re also considering the trip thresholds and may change them to match the current standard.

That’s all from me. I hope this was useful.