As launched final October, the summer time of 2024 was as soon as once more brutal from a lightning-induced electronics-culling standpoint on the Dipert family. I’ve already coated the recent tub management board that obtained zapped, in addition to documenting the laundry checklist of different now-DOA gadgets:
As soon as once more, a number of multi-port Ethernet switches (non-coincidentally on the ends of these exterior-attached community cable spans) obtained fried, together with a CableCard receiver and a MoCA transceiver (each related to exterior-routing coax). My three-bay QNAP NAS additionally expired, presumably the results of its connection to one of many lifeless multi-port Ethernet switches. All these things shall be (morbidly) showcased in teardowns to come back.
These teardown showcases begin right now with the final merchandise on the checklist, the QNAP TS-328, which I’d bought on sale for $169.99 again in January 2019 and fired up for service the next September in 2020. Right here once more to begin are the inventory photographs I shared again in December 2020 once I editorially coated the TS-328 intimately for the primary time (dimensions are 5.59” (H) × 5.91” (W) × 10.24” (D), and its internet weight absent HDDs is 3.62 lbs.):
The NAS (network-attached storage system) is tipped over on its left aspect for “present” in that second picture, if not already apparent. In regular operation, it’s the HDDs which are on their sides.
Now for some real-life snapshots of right now’s affected person, as common, accompanied by a 0.75″ (19.1 mm) diameter U.S. penny for dimension comparability functions. Entrance (numerous standing and exercise LEDs alongside the left aspect, with the facility and USB copy buttons, the latter with an built-in USB 3.0 connector for a tethered exterior storage system, under them):
(Bland) prime:
(Much more bland, a minimum of on the surface…maintain that thought) left aspect:
Proper aspect:
Again, with the fan dominating the panorama at higher left, a Kensington safety lock web site at decrease left, the “full vary” system speaker (vs the PCB buzzer, which you’ll see later) to its proper, and (top-to-bottom alongside the correct aspect) the recessed reset change, an undocumented-details (TS-328 consumer handbook is right here) “upkeep port”, an audio line-out jack, two RJ45 Gigabit Ethernet ports, side-by-side USB 2.0 (left) and three.0 (proper) connectors, and the facility enter:
And final, however not least, backside:
Earlier than I neglect, right here’s the exterior energy provide (which nonetheless appears to be totally purposeful):
These three thumbscrews you will have seen on the NAS’s again panel are our path inside:
Voila:
For orientation functions, what you’re right here is the within of the right-side shell, with the entrance of the NAS to the correct within the picture:
And listed here are the now-exposed guts of the NAS, tipped over on its left aspect and searching down on the first PCB mounted to the left aspect’s inside:
Right here’s one other perspective on the inner meeting, with the NAS nonetheless sideways:
And three extra views, with the NAS again in its “regular” orientation. Prime:
Backside:
and entrance:
Because it seems, there are two PCBs on this design, the principle one we’ve already caught a glimpse of, and one other related to the three SATA connectors behind the HDDs’ “cage” (I’ve briefly “jumped to the longer term” for the next shot, displaying the “cage” already indifferent):
Again to the current second. Detaching the “cage” entails eradicating seven screws, two every on the prime and backside:
And three extra that usually mate with the again panel:
With them eliminated, all that’s left is to detach the 2 halves of the connection between the 2 PCBs:
and the “cage” (with PCB nonetheless hooked up) is now free:
4 extra screws to go, to detach the PCB from the cage:
Mission achieved:
Discover what seems to be like corrosion on this rectangular steel area?
Sorry for ruining the shock, however this received’t be the final time you see it. I used to be unable to take away any of it with my fingernail, and belief me, the NAS was by no means uncovered to moisture, so it’s not rust. I don’t know whether or not it’s an artifact of being lightning-zapped, or (my suspicion) simply the end result of long-term publicity to a few heat-generating fast-spinning HDDs in a tiny enclosure.
The dominant IC on the backside is unsurprising given the PCB’s operate, an ASMedia ASM1062 PCI-to-SATA bridge and SATA controller. That mentioned, I’m nonetheless considerably stunned, as a result of the ASM1062 supposedly solely helps “two ports of Serial ATA 6Gbps”, however there are clearly three SATA connectors (due to this fact three SATA ports) on this design. Concepts, readers?
Within the ASM1062’s decrease left nook is a Macronix 25L4006E 4 Mbit serial interface (SPI, to be exact) NOR flash reminiscence. Given the 25L4006E’s low capability, to not point out its location on the opposite aspect of a PCI interface from the host CPU (whose heatsink you will have already glimpsed in a earlier picture), I’m assuming that it solely homes the firmware for the ASM1062, not the whole system. And no, it isn’t a NVM cache for the HDDs’ contents, both…
The opposite aspect of this PCB is relatively unmemorable other than an entire lot extra of what seems to be like corrosion. Provided that this aspect is extra instantly uncovered to the warmth coming off the HDDs, coupled with the truth that the HDDs remained totally purposeful after the NAS’s demise, my working principle as to the discoloration’s root trigger (excessive temperatures) is seemingly bolstered.
Now for the principle system PCB (look, extra corrosion!):
Earlier than diving in, listed here are some close-ups of the entrance whereas the PCB remains to be put in, displaying the gentle pipes from the LEDs to the entrance panel, together with the USB3 port and change assemblies:
In that earlier overview picture, you may need glimpsed 5 crimson marker-augmented (presumably to tip off the corporate to warranty-voiding proprietor removing) screw heads. Sadly, eradicating them didn’t allow the PCB to budge. However then I seen a bulge within the guarantee sticker in a single nook:
Sneaky, QNAP. Very sneaky!
That additional loosened the PCB, however I nonetheless couldn’t get it to completely detach from the steel bracket surrounding it, so I eliminated these 4 screws too:
Getting nearer:
And eventually, after disconnecting it from the zip cable-clustered multi-harness morass above it (which on reflection might have been what saved it caught in place after the preliminary six-screw removing course of):
The PCB was lastly free:
Earlier than diving in, a quick diversion; let’s look extra intently on the inside the again panel, rotated 90° from its “regular” place within the photographs that comply with. That’s the system fan to the left (duh) and the mounting bracket for the speaker to the correct:
Take away the 2 screws that usually maintain the mounting bracket in place:
And there’s your transducer!
Again to the PCB. There was, you will have already seen from the sooner overview picture, nothing of specific be aware on the bottom…except you’re into solder factors (or corrosion patches), that’s. The entrance aspect, nevertheless, was extra fascinating. Right here’s the far-right finish:
Prime-to-bottom alongside the correct edge, once more, are the recessed reset change, the mysterious “upkeep port”, an audio line-out jack, two RJ45 Gigabit Ethernet ports, side-by-side USB 2.0 and three.0 connectors, and the facility enter. On the backside left is the black-color connector, which initially mated with the SATA PCB. And the right-most two off-white coloured ones above it go to the speaker (two-pin) and fan (four-pin). Curiously, as you will have already seen, the opposite four-pin connector, instantly above the higher proper nook of the Faraday Cage, appears to be unused on this specific system design.
Talking of the Faraday cage:
There’s one other seemingly unused connector above it, eight-pin and black in coloration. And the IC in its higher left nook is the place, I consider, the first system firmware resides; it’s a Toshiba (now Kioxia) THGBMNG5D1LBAIL 4 GByte eMMC NAND flash reminiscence module.
Shifting over as soon as extra to the left…
At far left is the sooner alluded-to PCB “buzzer”. Above it’s a Weltrend Semiconductor WT61P803, which seems to be a microcontroller optimized for energy administration. Above that’s one other unpopulated four-pin connector. To the correct of the buzzer is the RTC (aka, CMOS) battery (which, by the way in which, I confirmed post-NAS failure was nonetheless purposeful however swapped anyway, not that it helped…generally a lifeless battery can thwart a profitable system begin).
Let’s get that heatsink off, we could? Needle-nose pliers did the trick:
The system CPU, a Realtek RTD1296 based mostly on a 64-bit Arm Cortex-A53 four-core 1.4 GHz cluster, is now uncovered to view.
And underneath the rest of the Faraday cage:
are 4 Samsung K4A4g165We BCRC 4 Gbit DDR4-2400 SDRAMs, collectively comprising the NAS’s 2 GBytes of (non-expandable, clearly) system unstable reminiscence:
I’ll shut with a few PCB finish pictures:
and a premise, making an attempt to reply the elemental query, “What precipitated the NAS to fail?” As I’ve talked about in previous protection of the QNAP TS-328, this NAS doesn’t have a very stellar long-term reliability report; see, for instance, this Reddit thread or this one, each of that are harking back to what I skilled. So, it could have simply coincidentally chosen that time limit to run out, pushed by long-term heat-induced failure of some element on the PCB, for instance. However the chronological proximity to final summer time’s lightning debacle is tough to disregard, provided that it’d been reliably operating for just a few weeks shy of 4 years at that time.
I don’t assume the DC energy enter was the failure level, because the PSU nonetheless appears totally purposeful (as I discussed earlier). The one different factor bodily linked to the NAS was its higher Gigabit Ethernet port; I’d wager that was the Achilles’ Heel, as a substitute. Subsequent non-operation traits (a quick twitch of the system fan on every powerup try, for instance) are past-history reminiscent to me of a PC whose CPU has gone awry. Basically, in spite of everything, what’s a NAS however a tailored-function, Arm- and Linux-based (on this case) pc? Though I’m unable to discover a detailed datasheet on the Realtek RTD1296 on-line, the overview info I’ve dug up makes repeated point out of dual-port Gigabit Ethernet help, suggesting that, at minimal, the RTD1296 integrates the MAC, thereby offering the requisite failure path.
Agree or disagree with my premise? Anything that jumps out at you from my dissection? Pontificate together with your ideas within the feedback!
Associated Content material
- Lightning strikes…thrice???!!!
- NAS successors add notable options
- Exposing a NAS safety subject
- NAS failure launches an information restoration mission
- Beefing up backup
- Lightning strike turns into EMP weapon
- Floor strikes and lightning safety of buried cables
The submit Analyzing a lightning-zapped NAS appeared first on EDN.
