FNIRSI-FNB58 USB Tester: A Metrology-Focused Review

Almost a year after this model came out, it finally made its way to me. That was inevitable: I enjoy tinkering with things like this. But since there are already countless reviews of this model, I decided to look at it as a measuring instrument—or, more simply, check its measurement accuracy.

A reminder: to keep up with new posts on the blog, I recommend following me on Instagram. I’ll post notifications about every new article there, and might also write about something interesting now and then - link to the account

You can also subscribe to updates and new articles on the Telegram channel - https://t.me/KirichBlog

On top of the AliExpress discounts, the store offered an extra $5 off with code ZKT555, bringing the price to $30.73 for the standard model and $33.36 for the Bluetooth version.

The first thing that catches your eye is the packaging. The manufacturer not only decided to ditch the metal box, but also went to the trouble of making a cardboard package. To be honest, I personally prefer the metal box, although the cardboard one looks more presentable.


The tester is the only thing included in the box, so let's get straight to taking a look at it.


It’s certainly large, especially compared with my other testers, but most of that size is down to the screen.


And yes, the screen really is big, taking up almost the entire top panel. To be fair, the wide case can make it awkward to use in some situations. For example, I sometimes have to pull out my “old-timer” from YZXstudio for that reason.
There’s a plastic panel on the back, hiding the Bluetooth module used to connect to a smartphone. A reminder: there’s a Bluetooth model (with a light-colored case) and one without it (in black), and naturally the price differs too.

One important change is the enclosure—and a metal one at that. Remember how for years testers were made as boards stacked like a sandwich? Sure, you could use them like that, but you always had to worry about something getting inside, like a lead from some radio component....



The case specifications and dimensions are listed on the product page.


The top and bottom edges have Type-C ports (input/output), a MircoUSB port (input), a MicroUSB port (for auxiliary power and connecting to a PC), a Back button, and a switch for controlling power to PD devices.

The USB-A ports are on the ends of the case. So far, so good: solid power contacts, plus a fifth contact, but... I’ll pause here and take a breath so I don’t swear... why USB 2.0? The FNB-48 had USB 3.0. What happened? Where did the manufacturer go wrong?
And, though I can’t say for sure, it seems to be the same story with Type-C.



I’ve often seen people in the comments on reviews wondering why it matters that there’s no USB 3.0. It matters because the connection used can make a difference when testing some devices.
For example, I reviewed a flash drive where you can clearly see that its power consumption differs when connected via USB 3.0 and 2.0.

The top shows read and write over USB 3.0; the bottom, over USB 2.0.


The port and control layout is almost identical to the FNB48 tester. The only difference is that on the new model, the larger case lets them sit farther apart, which is definitely more convenient.


When power is applied, a welcome screen appears, then the tester switches to normal operation.


There are four operating screens. The number of screens has been reduced a little, and the functions rearranged. From experience, I can say it’s more convenient this way, though not in every respect.
1, 2. The main screen displays the current voltage, current, power, and, for some reason, the temperature inside the device. The fonts are noticeably neater and larger, which is genuinely convenient. The only function available on this screen is Hold.
3, 4. The second screen is more informative: it shows the previous readings, capacity in Ah and Wh, operating time, and various details about the active protocols. There are some shortcomings too, in my opinion. The additional values are not only shown in gray, but are also very small, so people with poor eyesight will struggle. They’ve also removed the equivalent load resistance display, although it’s still available in the software. It also lets you control the recording function and select a data group.

Yes, the tester controls work the same as before: a short left/right press switches screens and moves through the menu, pressing selects an item, and there are also long presses and long presses to the side that activate additional functions. But since the controls are better optimized here, there are fewer combinations.



The third screen is the most informative. It brings together all the auxiliary modes and settings, which have also been regrouped to make them easier to use compared with the FNB48. Note that in Samsung protocol control and analysis mode, Samsung is displayed as sUmsung. There are also a few minor typos, such as Temperture and Vender.
This screen also has a calculator for estimating the capacity of power bank batteries. And they finally fixed the bug I mentioned in my FNB48 review: capacity is now calculated more accurately.


1–6. Settings. There aren't many, and they're already set up so that most users won't need to change anything. The only thing I changed was the brightness in standby mode; I don't like it when the screen goes dark.
7–9. Voltage/current graph for the power line, voltage on the data lines, and an oscilloscope. I'll definitely come back to the oscilloscope later.


Before testing, though, I decided to update the firmware. It's easy to do from a dedicated tab in the PC software. It came with version 0.63; the latest is 0.68.
While pressing buttons to enter update mode, the tester froze. This happened twice during all my testing: the first time was almost right at the start, while I was exploring its features, and it froze with a continuous beep. The second time was when I switched it to update mode. It hasn't happened again. To be honest, that was the first time I'd seen a USB tester freeze...

In any case, the update completed successfully. I haven't seen any more freezes (perhaps the update helped), and it certainly doesn't seem to have made things worse.


I'll start by checking how well it identifies the protocols supported by various power supplies, since that's one of its main functions.
First, a pair of Choetech chargers rated at 61 and 100 W. The protocols were identified correctly.


But an attempt to check an 18 W power supply from the same brand with a USB-A port failed. During protocol detection, the screen went dark and a notification appeared. After I confirmed it, the same thing happened again.


Out of curiosity, I repeated the same test with the FNB48 and got the same result. The tester did manage to identify more protocols, but then the screen went dark again and the notification popped up again.


Then I thought, “Okay, what the heck is going on here...” and dug out my old ZY1276 from YZXstudio. I tested the 100 W charger first, then the weird 18 W one that had stumped both Fnirsi products. And guess what? It tested it without any problems.
People often ask me why I use different testers. It's simple: situations like this are exactly why. Newer models can offer a lot more features, but sometimes you have to bring out the old-timer.


While I was at it, I tested a couple of power banks with the new tester. First, one of the models I sell, with a real capacity of 78 Ah/285 Wh, using its USB-A and USB-C ports. Then I tested a small 20 Ah model that I reviewed earlier, using the same ports.



The function for reading information from cables with chips can come in handy sometimes.
1, 2. First, some no-name cable and a 60 W Baseus cable. Neither has a chip, so there's no information to read...
3–6. A 2 m Choetech cable and some 240 W cable I bought on AliExpress. The latter is a bit odd, though: its chip reports a length of 2–3 m whether the cable is 50 cm or 1 m long.
In any case, the function works. Just make sure the tester isn't powered through that cable, or it won't be able to read it.


Now for the measurements. I'll start by checking the current draw.
1. The tester starts up at around 3.9 V; at 3.8 V, the screen only flickers. It immediately displays the correct voltage, and the current draw is 80 mA.
2–4. As the input voltage rises, the current draw drops. The manufacturer used a DC-DC converter rather than a linear regulator, which is definitely a good thing.


And now for the objective tests.
Voltage measurement accuracy: I first checked the 1–28 V range, then thought, why am I giving it such a generous allowance? I'll measure starting at 1 mV, especially since that's entirely feasible with an external power supply. There are a lot of photos, sorry, but I tried to group them in a convenient way. All tests were performed using the 1-2-5, 10-20-50, etc. sequence.

The stated accuracy is 0.2%+2 digits, but in this case we can ignore the last 2 digits because there are just too... many of them.

What can I say? In my opinion, the calibration is very good—excellent in places—and in the specified 4-28 V range, it's really hard to find fault with it.


I couldn't resist connecting it to a trusted voltage reference, and you can see that the error is clearly lower than the stated value. And since all the measurements are taken on the same range, the results are excellent.


Now for current. The stated accuracy here is 0.5%. I tested it from 1 mA to 6.1 A, using the same 1-2-5 sequence.
In places, the value in the "voltage" field is very similar to the value in the "current" field. That's just a coincidence, since the test was essentially performed with a short circuit at the output.

I think these results are very good too. Even at very low currents, the tester gives quite accurate and reasonable readings.


I came across a claim that the maximum power is 120 W. The description mentioned 140 W, but the tester measures at least 150 W without any trouble. I don't see why it couldn't display an even higher value.


Now for a rather controversial feature that was added quite a while ago, but for some reason doesn't get much attention: the oscilloscope.
I have no complaints about the current/voltage graphs or the voltage on the data lines; they don't need a high sampling rate. But you need a higher bandwidth to assess low-frequency ripple.

First, I applied square and sine waves at 1, 10, and 100 kHz, with an amplitude of 3 V peak-to-peak.
The 1 kHz square wave was a little rounded off, but that's generally normal. Otherwise, even at 100 kHz, it looked quite acceptable.
The sine wave was displayed too, but didn't look quite like a sine wave, even though this type of signal is usually displayed without any problems.

There are also some issues with frequency measurement, but I wouldn't say they're a major problem for this application.


What disappointed me most was that it measured the AC signal voltage incorrectly, reading much too low, so I decided to check that too.
At 1.5 V, the tester reads 1.075 V. At 1 V it's closer: 787 mV. But when I lowered it to 500 mV, I got 484 mV, which is much more plausible. With 200 mV at the input, the tester displayed 206 mV.
So when the input voltage is above 500 mV, the oscilloscope starts under-reading. You can see this on the waveform as well.


I lowered the voltage to 300 mV and tested again.
1-3. Sine waves at 1, 10, and 100 kHz
4-6. Square waves at 1, 10, and 100 kHz.
7, 8. Triangle and sawtooth waves at 100 kHz.
9. Generator output off; the tester picks up some interference at around 20 mV.

The results were encouraging: the oscilloscope handles 100 kHz more than adequately, given its limitations. I really don't see any problem using it to assess power supply ripple. The input bandwidth is roughly 500 kHz.



Next, I wanted to take it apart and see what was inside. I managed to remove the module and unscrew the circuit board, but then realized there was a problem: it's only partially designed to come apart. In other words, you can take it apart if you have no other option, but the odds of breaking something are close to 100%.
In my opinion, being so difficult to disassemble is a drawback. I once had to replace the voltage and current measurement chip in another tester, and I've lost count of how many times I've had to take another tester apart to solder on auxiliary wires. These days, I've set aside a separate tester just for that.


But I did manage to get a look inside, thanks to one of the YouTubers, and I'm very grateful to him. He also showed how to take the tester apart—and yes, there's no way to do it without swearing...


And finally, a few words about the software. Download everything here; the manuals, firmware, and so on are there too.

First, the computer software. It displays just about everything possible: the main window shows the current operating mode, a graph, and lets you view saved data.


The second window is for working with protocols.


And the third one is for firmware updates.


In my opinion, the smartphone app feels more cluttered, as if they tried to cram in everything that could possibly be crammed in. They could simply have split it across several screens, as they did with the computer software—just with more screens.

To connect, launch the software, find the tester, tap Connect, then tap Start in the app. That's it.


There are a few settings, including controls for displaying the current/voltage graph and the voltage on the data lines.


By the way, the graph clearly shows how the phone “negotiates” with the power supply: it first gives the phone 5 V, then 7, 9, and 12 V.


As I mentioned at the very beginning, in addition to the AliExpress discounts, the store offered an extra $5 off with the code ZKT555, bringing the price to $30.73 and $33.36 for the standard model and the Bluetooth version, respectively.
Based on my experience with similar testers, I can say that the version without Bluetooth is enough, since you can still connect it to a PC, and using it with a phone isn't very convenient anyway. That version costs a little over $30.



Conclusions.
All in all, I liked the tester, especially its decent measurement accuracy and support for a large number of protocols. As far as I remember, though, it doesn't support the very latest QC version. There's also a caveat with PD support for the latest version, but I can't remember that ever causing me a problem.
My main complaint is the lack of USB 3.0 support, at least on the USB-A ports. But this is only critical if you use the tester to measure the current draw of USB 3.0 devices. If you plan to use it to check various power banks, power supplies, and so on, the lack of USB 3.0 doesn't matter at all—and I only needed it a few times myself. The thing is, the previous model did support it, and it bothers me that it's unlikely to come back :(

In short: a fairly capable model in a decent metal case, with support for a ton of protocols, an accurate meter, and a large screen. Whether to buy it depends on what you need it for, but in my opinion it's pretty good for $30.

If anyone has been using this tester for a long time, I'd be glad to hear your comments and feedback.

That's all for now. I hope this was useful; I'm off to prepare some more reviews :)