Multichannel Voltmeter

Today I'm reviewing a somewhat unusual device. To be honest, I'm not even sure what the right title for this review would be, since this device isn't just a voltmeter, though that's probably the main thing I'm interested in here.
This won't be a very long review, partly because the device is rather specialized and partly because there isn't all that much to say about it.


Not long ago, I reviewed a 80 V adjustable power supply board and an electronic load. This voltmeter will also be used with them as part of a multichannel battery tester.

I actually already reviewed a product from this seller/manufacturer, and even from the same listing. That time it was a Sousim electronic load controller. It works fine, and I use it from time to time. This time I needed a multichannel voltmeter, so I remembered this listing.

Oddly enough, it arrived from the forwarding service without the usual bag.


But the packaging was something I'd never seen before. Of course, thanks to the forwarding service for taking care to protect the item, but a box inside a box is a bit... unusual, especially considering that shipping is charged by parcel weight.
I even filmed an unboxing video, but since it's only tangentially related to the review, I've put it under the spoiler.


Besides the meter board itself, I needed four more things:
1. Enclosure
2. Temperature sensor
3. Current-measuring shunt
4. Bluetooth module

I didn't buy the last two items on Taobao, since shunts are cheaper here, and I already had a Bluetooth module—or rather, more than one.
However, because of the way orders from a single product page work with the agent I used, I had to place orders from different product pages. Sometimes you can't order several lots from one page because the totals are calculated incorrectly—a quirk of yoybue.

In the end, the seller understood without any questions that I planned to use them together, so I received the meter in its enclosure with the temperature sensor already soldered on :)


But there were a few snags: it turned out that the leads and connectors for hooking up the batteries were either not included or had been forgotten. :(


A little about the kit and its components.
First, the enclosure. It costs about $2.20 — link.
The board is fitted very tightly. After removing the four screws, I could barely pry it out.


It's a panel-mount enclosure, meaning it installs in an instrument panel. I didn't buy one like this last time, but this time it's exactly what I need—and it's much easier to mount this way.
The enclosure's cutout dimensions are 92 × 64 mm


There's a fairly large backlit dot-matrix LCD on the front.


There are three control buttons on the right, just like last time, but there's no LED, even though the enclosure has a spot for one.
The display is straight, but it looks like its mounting holes were made incorrectly—the pins are somehow at an angle.


The second add-on is the temperature sensor. It costs $1.76 — link
I like the temperature sensor. It's fairly large, has a metal housing, and comes with a flexible cable. The cable is two meters long.


Last time I complained that the board was bare; this time there are clearly more components. Since I ordered the 24-channel version of the meter, it has three 8-pin connectors, three multiplexers, and three DACs soldered on.


The seller offers two power supply options: 35–200 V and 6–35 V. This time I planned to power the meter from the same power supply as the adjustable DC-DC converter, i.e. at about 90–95 V, so I chose the 35–200 V version.
As I noted last time, a big advantage is that the instrument's power supply is galvanically isolated, which makes it easier to connect it to the shunt.


The 24 channels are divided into three groups of eight. Each group is handled by a 74HC4051 multiplexer and an 18-bit ADC, the MCP3421. This was probably done to speed up measurements, since the ADC isn't particularly fast.


1. There's another ADC next to the power connector. The current-shunt connection terminals are brought out there too, and there's also an EEPROM chip.
2. Apparently, this is a universal board designed to connect various peripherals, including two temperature sensors, GPS, a TFT display, and so on.
3. Pads for connecting additional inputs, a computer, and an external keyboard are brought out to one end of the board.
4. Since I planned to connect a Bluetooth module, I soldered on a four-pin header before installing the board in the enclosure.


In addition, I had to buy:
1. One meter of 10-wire ribbon cable. Eight wires would have been more appropriate, of course, but that wasn't available. — link
2. A few connectors and their contacts — link
3. A 10 A, 75 mV shunt. — link


1, 2. As I mentioned, you can buy the shunt from the same seller on Taobao, but it turned out to be even cheaper locally.
I have no complaints about the quality, but the resistance is a little high. To produce 75 mV at 10 A, the shunt should have a resistance of 7.5 mΩ. In reality, it's noticeably higher at 7.72 mΩ—a 2.9% difference, even though Class 0.5 implies 0.5%.
3, 4. I also messed up with the ribbon cable: instead of the usual stranded wire, I bought solid-core wire. It does have a slightly larger cross-section (26 AWG versus 28 AWG), and it's quite stiff. But part of the reason I bought it was that its pitch is 2.54 mm, so in theory it would be easy to strip the wires and fit them into a 2.5 mm connector.


But let's move on to the meter itself.
1. When you switch it on, this rather nice splash screen appears for a while. It also shows the development date of this version.
2. The device then switches to measurement mode.
3. An alternative screen view, which I'll talk about later. The middle button controls the backlight; holding it down resets the Ah counter.
4. I found an option to switch to English in the menu, so the rest of the description will use the English-language display.


First, the control menu and a brief description of some of the options I could figure out. To enter the menu, hold the top button for about 8–10 seconds. Release it to enter settings mode.
1. Minimum voltage threshold. As far as I can tell, this just turns on an LED that isn't there.
2. Same as item 1, but for the maximum threshold.
3. Data transfer speed when connected to a computer, from 2400 to 115200.
4, 5. Shunt type selection. Unfortunately, the adjustment on this device is very coarse, in 1 A increments, and there's no fine calibration like there was on the electronic load controller.
6. Set the timeout for switching something off (?)—not used here.
7, 8. Select the maximum voltage measurement range.
9. Another timer setting; I couldn't figure out what it's for either.


1, 2. A strange setting for the computer output: 485 or TXT. I can't quite see how those two go together.
3. Select the port number for the 485 interface.
4. Set the capacity of the connected battery to display the charge indicator. Don't touch this under any circumstances! I tried enabling it, and now I can't set it back to 0 Ah. The values cycle like this: 13.1, 13.0, 12.9, 7.2, 7.1, 7.0, 6.9, 6.8, #6722 Ah.
5, 6. Select the number of inputs to measure: 1/8/16/24/32, which suggests you can add more inputs yourself.


For the next experiments, I need one of the test batteries. In this case, it's a Samsung 10S2P.


Since the battery only has two external wires, I remove the heat-shrink tubing.


Inside is a protection board with a terminal block. I plan to make a universal adapter for different battery types to make connecting them easier.


The connector has 11 contacts. The first is the black common wire, followed by the positive terminal of the first cell, then the second, and so on up to the tenth, whose positive wire is marked red. It's very clear and convenient.


Before the tests, a few words about the display. It has four modes:
1. Voltage shown cumulatively. This is the default: first it shows the voltage of the first cell, then 1+2, 1+2+3, and so on.
2. Voltage of each cell individually. Press the bottom button to switch to this mode.
3. Press the top button to change the display format. This view shows the total pack voltage, current, temperature, and capacity.
4. Press the bottom button in this mode to show the minimum and maximum pack voltage on the bottom line.


Changing the number of inputs also changes the display in the first screen view.
1. 8 cells
2. 16 cells. In this mode, the voltage readings are small, but it also shows information from the second screen: voltage, current, capacity, and power. I couldn't find the last item in the other display modes.
3. 32 cells. Everything gets pretty tiny here, unfortunately.
4. Here I set the battery capacity to a value above zero. The battery icon now shows the capacity percentage and a graphical fill level. And now I can't turn the icon off.


The tricky part of connecting it this way is avoiding an accidental short circuit. The short-circuit current will be substantial, and there's no protection—the connection is made upstream of it. I powered the whole setup from my adjustable power supply.


I wired it according to the diagram on the product page. There’s also a version with 32 inputs on the same page; the only difference is an additional connector.


I connected it up and didn’t even burn anything out. In fact, everything works :)


The meter starts up at voltages above 33–34 V, drawing about 15 mA. As the voltage rises, the current drops, reaching just 5 mA at 60 V.


The current meter works with either connection polarity. I initially wired it according to the diagram on the product page, and when I tried to charge a battery, the current reading was negative. I simply swapped the wires to the shunt, and the reading became positive.
Accordingly, the capacity reading in Ah can also be positive or negative.


1. With no load, the current reads −10–20 mA when the power supply is connected, but if its voltage is below the battery voltage, the reading is around −30 mA.
2. If you raise the power supply’s output voltage, charging starts and the current reading becomes positive, though it’s higher than the actual current.
3, 4. The current reading is more than 1% high: instead of 1.01 and 2.02, I got 1.03 and 2.07. The difference is partly due to the shunt, whose resistance is higher, and partly due to the rather coarse adjustment and lack of calibration. At least, I couldn’t find a mode for calibrating the current meter, so the only way to do it is by modifying the shunt itself.


The voltage readings, on the other hand, are excellent.
1, 2. When measuring each battery, the difference fluctuated by about +1–2 least significant digits. The left image shows the best result, the right the worst.
3, 4. The overall voltage measurements were much the same: the difference ranged from +4 to +7 digits.


Aside from a few software shortcomings, everything was going pretty well until I decided to try connecting it to a computer. I mentioned above that the seller offers a Bluetooth adapter; the product page also lists an RS-485 interface adapter, but I planned to connect them myself.
I expected the program to show a window like this. You can download it here.


The program also lets you export data to Excel and plot the voltage of each battery in the pack.


I connected it through an RS232 TTL-to-USB adapter, but all I got was this window.
It supposedly connected, but no data was being transmitted. Also, double-clicking any of the five buttons on the right (the first two are for connecting/disconnecting) caused the software to freeze. A single click followed by connecting or disconnecting did nothing at all.
The window also looked a bit odd because Chinese text wasn’t displayed properly.


After the wired connection failed, I got a Bluetooth adapter and tried that instead. I also downloaded the APK from the corresponding manufacturer’s page.
The page shows what the connection should look like. Mine was more or less the same, except the adapter name was different.
Tap one of the buttons, and a list of Bluetooth devices appears.


After that, it connects and should immediately display data from the device.


I got a similar window too, but it showed 32 batteries instead of 24, and the values had nothing to do with reality. It felt like I was looking at some kind of "placeholder".
I tried various settings: RS-485 and TXT modes, baud rates of 115200, 9600, 4800, and 2400, and swapping RX and TX (just in case they’d wired them differently). I even tried two different adapters. One wouldn’t work properly, but the other clearly showed that it had connected when I connected through the software.


But unfortunately, none of that helped. The software says everything is fine and that there’s a connection, but apparently it means a connection to the Bluetooth adapter, not to the device itself.


Video review


I think that’s probably enough for now. I can say that I like the hardware: the circuitry is simple and provides decent voltage measurement accuracy, which is exactly what I needed. There are some issues with the current, but adjusting the shunt can improve things a little.

From a software standpoint, though, things are far from rosy. To start with, there are problems configuring the device: sometimes it’s hard to select the value you need. For example, when changing the battery voltage threshold, I could set it to 3.71, 3.72, or 3.73 V, but the next step was 4.56, 4.57, and so on. And to make matters worse, I could only get it back down to the original 3.7 after setting the voltage well below the required value and then raising it again; otherwise, after 4.56 I got 3.65.

In short, the software still needs a lot of work. But honestly, even that isn’t the main issue for me. Much worse is that I couldn’t connect it to a computer. I can live with the software being in Chinese; there aren’t many menu items. But it simply doesn’t work. I’ll try contacting the seller; maybe they can help or explain how to get the connection working. I’ll cover that in one of the next reviews about assembling a battery tester.


This review was sponsored by the purchasing agent yoybuy.com, who covered the shipping costs.
The total cost of the kit, including shipping to the agent, came to $34.22. The kit weighs 159 grams. Shipping from the agent depends on various factors, such as the quantity and whether there are other items in the order.