This is my first teardown in a while. My focus this time is NANOTE, a mini-laptop available at less than 20,000 yen and no doubt highly sought-after by fans of ultra-cheap gadgets. That much-talked-about “Jonetsu Kakaku” model.
Weighing in at around 520 grams, NANOTE’s specs are as follows: an Intel Atom x5-Z8350 as the CPU, 4GB of RAM (LPDDR3), 64 GB of eMMC storage, and a 7 inch LCD display (1920×1200 pixels).
Respectable reviews have already been released in the form of articles, so I recommend reading those if you want to know what kind of product NANOTE is and get a proper overview. This report, however, is purely for gadget geeks such as myself.
Related articles (Japanese):
The NANOTE used in the below teardown was courtesy of Gian Suzuki, a writer who cannot resist purchasing whatever gadget catches his attention. Naturally, he bought NANOTE in a heartbeat and seemed to be having a fantastic time playing around with it. After a while, I was asked to give a review on the product’s hardware.
The fact that I would be given an actual NANOTE to work with was a major deciding factor. Not to mention,
“I don’t mind if it gets broken so dissect it to your heart’s content.
Mr. Suzuki, you are too kind! Greatly appreciated!
After he said that I thought to myself that even I couldn’t do such a thing...but...
I’m so sorry!
Just to be clear at the outset, this article does not recommend the tearing down or modification of electronics. Engadget’s Editing Department bears no responsibility under any circumstances, so please do not try to imitate the actions described below. Also, this article contains a lot of my own personal thoughts so there may be some inaccuracies in the information. Thank you for understanding.
Checked out the point that puzzled me after removing the baseplate
I had heard in advance that the Wi-Fi antenna cable cut across the CPU and if you updated Windows the bottom of the NANOTE would swell up. This only raised my feeling of anticipation all the more.
Before I did the teardown, one aspect that I’d noticed about NANOTE’s appearance was that the keyboard side was a little warped. Still, the warping was so slight you could only notice it by putting a ruler up against the side. The thing I found the most suspicious about NANOTE was that the technical standards conformity certification mark was elongated. This could be because of space restrictions, but it just seems a little shady to me.
Incidentally, when NANOTE was first released I couldn’t find it in TELEC’s registry, but when I did a search a few moments ago, it was definitely there. It seems NANOTE was registered on April 20th, but it took some time to be reflected.
The disassembly of the main body portion is extremely easy and only requires removal of the six screws visible on the base. You can take these out with a regular Phillips head screwdriver.
Once the baseplate is removed, you will see that, if the hinges are at the top, the left side comprises the main PCB and copper plate, while the right side holds the battery. Despite being a compact PC, there’s not much inside. The speakers’ connector is disconnected but that’s because I had removed it myself and forgot to reconnect it before taking this photo. Pay no mind.
The battery is 5000 mAh and 3.8 V. There’s even a PSE mark. Unlike the technical standards conformity certification mark, the PSE mark isn’t elongated.
The blue object near the center is a heat conductive sheet which transmits the heat from the copper plate to the baseplate to release it. NANOTE is not equipped with a fan, so the only cooling component is this heat conductive sheet.
In fact, this is the reason why the baseplate lifts up and a gap exists. As you can see, the sheet is made of slightly thick material and the base of the device is closed as if to press the sheet down. If this portion heats up, warping increases due to the different expansion rates of the baseplate (aluminum alloy) and the frame (resin), and a gap is formed.
Unless tightly pressed together, it is difficult for heat transfer to occur and the CPU, etc. cannot be cooled, therefore to some degree this cannot be avoided in a fanless design. Even still, there really is quite a noticeable gap if you lay the baseplate down on top of the main unit without screwing it on. I think it might have been perfectly fine to use a slightly thinner heat conductive sheet.
Also, the baseplate thickness is only 0.8 mm, which is relatively thin. If it was just a bit thicker, it would curve more gently and I believe the gap would not be as significant, so it really is unfortunate this was not reflected in the design.
Right, let’s return the focus to NANOTE’s internal components.
When I removed the black tape attached to the battery cable, I found a connector. I don’t like how electronic devices with non-removable batteries continue to be live even when they are torn down, so I am happy that, in NANOTE’s case, I can disconnect this easily.
I have no idea what the connector directly under the battery’s connector is used for. It only has four pins, so maybe it is for fan power or a temperature sensor.
The dark yellow sticker in the upper middle is the Wi-Fi antenna. The back of the keyboard is covered in a metal plate, making it hard for radio waves to pass through. As such, the Wi-Fi antenna is attached to the resin frame. It sort of resembles a chip, but this is common in the antennas of many Wi-Fi devices.
Removed the copper plate and unveiled the main PCB
The copper plate is only secured by three screws, so it too is easily removed using a Phillips. Once removed, it looks like how it’s shown in the below photo.
I had been excited as I’d heard the Wi-Fi antenna cable cut across the CPU, however it seemed like the particular NANOTE I was tearing down had the perfect layout, and the cable passed through the middle without interfering with anything, as per normal. This discovery was disappointing for me, however this was in fact how the layout was supposed to be from a product perspective.
The two pink portions are also heat conductive sheets. The CPU is shown in the top left, and the power IC is shown in the lower right. These two heat-generating chips are cooled by the heat conductive sheets.
What I can’t work out though, is the reason for the empty spot in the middle of the PCB. What could this be? If it was intended to be the location of some kind of IC then the pin arrangement is kind of peculiar. Not to mention, the numerous pads on the inside are a complete mystery. Perhaps they are terminals for testing the PCB? But if that was the case, I feel it’s strange that they have soldering. Above all, this portion just seems to stand out far too much.
I turned over the copper plate I’d removed and saw a sheet I believe is used for insulation purposes attached to the back face. This sheet was torn due to swelling of the CPU and power IC portions. As such, the copper plate only touched these two locations and nowhere else.
The main PCB can be removed by disconnecting each connector and undoing the four screws. This is also possible with a Phillips head alone, so the disassembly itself is easy. The connectors are small so it may take some effort to disconnect if you aren’t used to it, but it really isn’t that difficult a task.
As such, to make the most of this teardown opportunity, I thought I’d include a photo of both sides of the main PCB.
The rear face had a sheet attached to it probably for insulation and protection (I took this photo without removing the sheet so there is some reflection). I’ll introduce some of the large components equipped on the PCB.
Firstly, the CPU. It’s an Atom x5-Z8350 stamped with SR2KT (the spec code) and the die is unexpectedly large - that is my honest opinion. The NANOTE is equipped with a 4-core, 4-thread CPU, GPU, and related I/O, therefore of course the die is going to be somewhat large.
Right beside this CPU is the memory. The stamp reads SK hynix, but there are rumors it is an imitation product. I don’t have the skills to distinguish between authentic and fake so I can’t say either way. All I know is this RAM enables the PC to operate smoothly enough so there doesn’t seem to be any harm in using it from a functional perspective. Of course, I don’t know how the quality, error ratio, etc. would be after prolonged use.
Incidentally, there are two 16Gbit chips, so a total of 32Gbit. In other words, 4GB.
The green module set slightly apart is the Wi-Fi. The familiar crab mark indicates the manufacturer of this RTL8723BS chip is Realtek.
Looking closely, I see that the SDIO connect supports IEEE802.11n. This chip also supports Bluetooth 4.0 technology. I was curious as to how much this module cost so I did a search and found an identical one, Aliexpress, for around two or three dollars. That’s all they cost.
In the same way as the CPU, the power IC had a heat conductive sheet attached to it for cooling. The power IC is an AXP288C manufactured by Shenzhen X-Powers.
Function-wise, this power IC supports a single-cell lithium-ion battery and is customized for use with Intel Cherry Trail. In other words, it is an IC suited to use in compact UMPCs such as NANOTE, and tablets.
Another large component on the back of the PCB is the eMMC. This is a FORESEE NCEMASLD-64G made by Longsys.
I like how you can sort of tell that it’s 64GB from the stamp. Upon close investigation, I learnt that the specs are a 150MB/s or higher light and 200MB/s or higher lead, meaning that it is much slower than most SSDs, but still fairly fast for an eMMC.
One thing on the PCB other than the main components that caught my attention was the empty spot next to the microSD card slot. It is about the same size as the microSD card slot but only has seven connection pins.
Perhaps it’s safe to assume this might be reserved for an SIM slot? A quick search led me to a slot such as this one that would fit neatly in the vacant area.
Maybe it is the case that the PCB manufacturer customizes the product by changing the mounted components to suit customer requirements. What’s surprising is that such a universal circuit board is possible at this tiny size.
While we’re here, please take a look at the connector. It isn’t a common type of connector mounted on the PCB’s face, but rather a type that is mounted by cutting out a notch in the PCB.
Connector thickness is often a troublesome issue in flat devices, so this mounting method has the benefit of keeping the thickness relatively minimal. Incidentally, the tact switch at the back of the USB Type-C is the power button.
Introducing other peripheral components
To make the most of this teardown opportunity, I not only removed the main PCB, but also peripheral components from the body. Allow me to introduce them.
Oh, by the way, I included photos to show what is connected to where in the PCB. After all, it’s difficult to make out the connector positions and so on.
Considering it is instantly obvious what this is considering that it’s the largest component after the PCB, I feel maybe there isn’t a need to introduce the battery, but I will anyway. So here is the lowdown. The battery is fairly easy to remove as it’s only held down by double-sided tape attached to the two rows on the rear face. Of the two rows, only about half of one row was stuck on. This configuration is very considerate of people who like to tear devices down.
I was surprised to find that the left and right speakers were shaped differently. These are also only fixed to a plastic frame with double-sided tape, so they can easily be removed by wedging a finger in-between and pulling. Again, great for dismantlers like me.
The microphones are small, but there are two of them. One of the mics was stuck on with something that looked like adhesive but the other wasn’t so I was able to easily remove it. I thought this was just the way it was but when I looked closely, I could see a mark indicating a failed attempt at sticking the battery in place.
The circular space immediately under the cream-colored adhesive mark is the location of the mic. It’s a little out of place, isn’t it? Incidentally, this is what the mic should look like if properly fixed down.
This is the way it is supposed to be arranged. It wouldn’t be very easy to tear down.
A metal plate broadly covers the back of the keyboard to improve the rigidity of the resin body and prevent the main unit from warping when the user is typing. This metal plate is fused into the resin so unfortunately, I can’t disassemble it. If I did, I could never return it to its original state.
By rotating the display to about 270 degrees, then taking out the four screws of the hinges, I was able to easily remove it. The movement of these hinges is quite interesting in that, up until 180 degrees, only the portion connected with the LCD panel rotates then, beyond that, the portion connected with the main unit rotates. This method was adopted to achieve both a notebook computer and tablet style. It’s quite impressive, really.
Up to this point, the teardown is extremely easy to do and reassembly is also straightforward, making it hard to go wrong. I suppose if one had to be careful of anything, it would be the four different types of screws that are somewhat plain and hard to tell apart.
The two pictured on the left are countersunk head screws for fixing down the baseplate. The long one is used close to the hinges and locks in firmly. The two screws on the right have thinner heads, with the shorter one used for securing the PCB and the long one for holding the hinges in place. The diameter and pitch are common to all of these screw types so care is needed when reassembling as its hard to notice if you use the wrong one in the wrong place.
Got carried away and even tried dismantling the LCD display
I should have just stopped at this point while things were going well, but I realized that the information I’d acquired through the teardown process could easily be found in an online search, so I decided to challenge myself by disassembling the display panel.
After removing the cover next to the hinges, I got the impression that the LCD was attached using double-sided tape. Removing it would require persistent effort, namely involving heating the tape to reduce its adhesiveness then inserting a thin object like a spatula and pushing it out.
If the tape’s adhesiveness wasn’t that strong, one could simply heat it up a little, then use suction caps to pull it off, but in NANOTE’s case, the tape is stuck on stubbornly. Even pulling it a little didn’t create the slightest gap.
As such, I used a dryer to heat up the area where I believed the tape to be located and slid in a thin metal plate. It budged a little, but when I forced it a bit too much, I heard a horrid cracking sound.
I’d done it now.
It’s easy for force to transfer to the corners so, despite my carefulness to begin with, I let my guard down once I’d removed the panel to a certain point and tried using my finger to pull at the corner. This move led to ultimate failure. There’s definitely a need to use suction caps to remove the panel. Humblest apologies, Gian.
Anyway, because I grew fairly bold from this point, I persevered, and a few cracks later, managed to get the display fully off. Incidentally, with the panel portion removed from the main unit it was difficult to apply enough force to the LCD, so I put it back on the main unit before attempting this task.
As a result, I was able to separate the two pieces, not without incident though.
Let’s take a look at the back of the LCD display. The first thing I noticed was the LCD circuit board. The circuit board is often attached to the rear face of the display, but this one was just hanging there loosely.
Broadly speaking, there are two cable routes coming from the main unit; the one from the right hing is for the LCD display, and the one for the left hing is for the touchpanel and camera. Please take a look at the adhesive tape portion that gave me a hard time during my removal attempt.
This is on the left of the display, where the camera is located. Tape fills the entire area and I felt a bit impressed with myself for being able to remove the display at all (albeit with a few cracks). The flexible cable connecting to the touchpanel is extremely close to the outer periphery so if you tried to slide a cutter in between, you’d cut it for sure.
The opposite side of the LCD looks like this. While there are no cables nearby to be concerned about, it seems as though the tape here is stuck on even more firmly, and it really was a struggle to peel it off. The smartphone and tablet repair people of this world are truly amazing. If I ever attempt this again, I won’t try to make do with a hair dryer, but actually use a hot plate or heat gun to do the job right.
Other than the LCD display, there was an independent PCB for the camera. Let’s take a closer look.
This camera is a module-type typically used for built-in cameras, and the controller is inscribed with SONIX 9C268CJ. A quick online search led me to SONIX’s site where I only acquired a snippet of information that the SN9C268 image processor was released in May 2017. SN9C291B, a controller for HD cameras believed to be from the same series adopts USB connection and so does this one. There are four connection wires.
Took a peek at the power source too
Finally, moving away from the notebook computer itself, let’s take a look at one of its accessories, the AC adaptor.
Naturally the AC adaptor accompanying NANOTE is Type-C because the product is USB Type-C. However, output-wise, it is fixed at 5V, therefore even though it’s 3A, it doesn’t really seem like a PD.
I checked the current of the adaptor when charging the main unit and as far as I can see, this ranges from 5V 1.5A to 1.9A. The moment the adaptor is connected, the current is 0.5A, which increases after around 0.5 seconds. It’s a fairly decent item. With this much current, 3A is enough, in fact, more than enough. While feeling a little nervous, I hooked the adaptor up to my Nintendo Switch, and the current was only around 5V 1.4A.
So one thing that played on my mind here was “Can this adaptor really output 3A?” I decided to investigate to satisfy my curiosity. When I tried using an electronic load with USB Type-C input, I saw that I could raise it to around 3.3 A.
The current amount is certainly not stable, but I was able to output between 3.1 and 3.3A without dropping below 5V. Certain to say, the adaptor is capable of securing 3A.
Wanting to see what would happen if I increased the load further, I tried staying over 3A, but the voltage suddenly dropped to 2.5V.
Most ultra-cheap USB AC adaptors with no particular control gradually lose voltage due to being unable to withstand the load, however NANOTE’s AC adaptor seems to be designed so that the voltage drops majorly when subjected to an excessive load in order to protect the device. It seems the controller inside the AC adaptor is doing its job properly, and I was quite surprised at how well it performs.
Setting aside its performance as a PC, NANOTE’s hardware specs are pretty solid
NANOTE is quite tiny, even for a notebook, so I assumed it would be crammed full of components but in reality, there were few components and few screws; making for an extremely simple design overall. In addition, all peripheral parts such as the microphones, speakers, and batteries use connectors, meaning there is ample space and making teardown/reassembly work a breeze.
Setting aside NANOTE’s performance and usefulness as a computer, I got the impression that its internal hardware is made really well. This is extremely surprising and appealing considering the product’s price of less than 20,000 yen.
However, it’s obvious that NANOTE is assembled by hand, so the finished quality varies between individual products. Also, it would be nice if the manufacturer could somehow improve the main body which is susceptible to warping and gaps but I suppose being too meticulous here would significantly raise the price.
Personally, I think it’d be interesting to only sell the main PCB under the demise of a microcontroller for insertion. If that was the case, users would seek options like an LCD display and recommended batteries, and start thinking they’d like to buy a compact case to put it all in...so I guess, after all, the verdict would be “We may as well sell it as a complete notebook PC!”
I may have broken the display, but when I temporarily assembled it, at least it functioned alight - for the moment. Still, the bottom half of the taskbar could not be seen and the area around the start menu was also beyond saving. With another crack in the top right corner of the screen, it would be difficult to continue using this NANOTE.
Incidentally, the touchpanel does respond to contact, but proper operation is not possible. Maybe I’d damaged the cable, causing the touchpanel to break along with the LCD. So sorry...
This article was originally written in Japanese. All images and content are directly from the Japanese version at the time of publication.