Lenovo’s Smart Clock 2: A “charged” device that met a premature demise

Back in April 2022, EDN published my teardown of Lenovo’s first-generation Google Assistant-cognizant diminutive smart speaker, the Smart Clock Essential:

As I mentioned at the time, the second-generation Smart Clock Essential had already been introduced, and switched to supporting Amazon’s Alexa ecosystem. Lenovo had at that point also already introduced two generations’ worth of its larger touchscreen-inclusive Smart Clock, both Google Assistant-based, most recently (nearly a year earlier, to be exact) the redesigned Smart Clock 2, which initially cost $69.99 standalone:

or $89.99 bundled with an optional charging dock for both it, a wireless-charging (with MagSafe support, to boot) smartphone or other device, and a USB-tethered device (the USB charging port moved from the back of the speaker itself to the dock in this second-generation design):

That said, I bought mine brand new direct from Lenovo at the end of 2022 for only $29.99, complete with the docking station (which I’ll save for another teardown to come). Why the substantial price drop, along with the broader Amazon-to-Google voice interface service switch? I don’t have direct insight, but I suspect it had something to do with coverage that appeared a few months later, like “Google cuts off third-party smart displays as Assistant support dwindles”:

Google has stopped pushing updates for some third-party smart displays, reflecting a broader shift away from Assistant products. On a support page spotted by 9to5Google, the company says it will no longer provide software updates for the Lenovo Smart Display, JBL Link View, and the LG Xboom AI ThinQ WK9 Smart Display.

All three displays made their debut in 2018, just months after Google first announced the Smart Display platform and its own Home Hub (now called Nest Hub) as it sought to compete with Amazon’s Alexa. While Google provided some new features to these devices in the years that followed, they never received the same kind of attention it gives to its Nest Hub displays.

Google’s decision to end support for its third-party devices doesn’t mean they’ll stop working, but it is indicative of what’s been happening over the past few years: they just won’t receive any new features or updates.

That’s unfortunate, because the Smart Clock 2 (along with its Smart Clock 1 and Smart Clock 1 Essential predecessors) were generally quite well-reviewed. And, as you’ll soon see, it’s also interesting internally.

Let’s dive in, beginning with some external packaging shots:

Pop open the (outer) box lid:

and inside, you’ll find a box at the bottom, which I assume is the dock:

a cardboard space-filler:

and the box containing our victim (perhaps obviously, the packaging modularity affords optional smart speaker standalone sale absent the dock, as per the earlier noted pricing differential):

(As shown here, the smart speaker comes in three color options, Shadow Black—which is oddly not listed on the product spec sheet, Heather Gray and Abyss Blue. Mine’s Heather Gray. The charging dock comes only in white.)

Open sesame, once again:

To the right, obviously, is the power adapter, with a barrel connector to the smart speaker:

Below are two pieces of documentation:

And last but not least, here’s the star of the show:

Now freed from its clear-plastic confines, and as-usual accompanied by a 0.75″ (19.1 mm) diameter U.S. penny for size comparison purposes. The Smart Clock 2 has dimensions of 3.67″ x 4.47″ x 2.81″ (93.30mm x 113.48mm x 71.33mm) with a weight “starting at” (whatever that means) 0.66 lbs. (298 grams):

Compare it to the first-generation Smart Clock, released in January 2019 (and available only in gray and black, by the way):

and the particulars of the dock-friendly redesign will likely be immediately obvious. Here are some more perspectives on our patient:

(note the “pogo plug” contacts for the charging dock interface)

I’m betting that the rubber “foot” around the bottom-side edges is our pathway inside. Agree or disagree, readers?

Ayuh (or, if you prefer, ayup):

Putting aside for the moment the now-revealed bulk of the device’s “guts”, let’s check out the inside of the bottom panel first:

Perhaps obviously, considering its proximity, the circuitry on this mini-PCB handles two main tasks: managing the power coming in either via the barrel connector (if used standalone) or the “pogo pins” (when used with the dock), as well as as-appropriate muting the microphone array based on the next-door switch setting. The flex cable you likely already saw earlier transfers them both (analog DC power and digital on/off) to the main PCB.

Let’s get that mini-PCB out of the bottom chassis piece for a closer look:

Now-revealed underside of the mini-PCB first:

And now the (standalone) other side you saw already:

Not much to write home about, eh? (post your thoughts in the comments if you disagree!)

At this point, returning to the main assembly, I admittedly was a bit stuck:

Let’s first orient ourselves; at bottom of this photo, resting on my desk, is the front of the device. See those two screws toward the bottom?

Unfortunately, removing them didn’t get me anywhere. The insides wouldn’t budge. But diving into the photos included with the FCC certification documentation (ID O57CD24502F) clued me in that the touchscreen display was enclosed within an assembly that, with a bit of dexterity, could be popped out the front of the device:

Voila:

Removing those two screws, by the way, had absolutely been necessary; they’re what held the display assembly in “permanent” place.

Here’s a closeup of the currently-reinforcement-taped flex PCB cable connection to the remainder of the system:

Along with another closeup, this our first peek at the two-microphone-array PCB:

(note that one of the two rubber seals between the mic ports and the display assembly ended up still attached to the former, with the other instead clinging to the latter)

A quick aside, while we’re staring at this particular circuit board. If you look closely at the tech specs on the product page, you’ll see two sensors listed:

L-Sensor
G-Sensor

Neither is described in any detail whatsoever. I’m guessing (limited online chatter concurs) that:

“L” stands for “luminance”, i.e. “ambient light”, and it’s probably the sensor in-between the two microphones, one on either end. It adjusts the LCD backlight based on the ambient illumination (or lack thereof) in whatever environment the speaker’s located.
“G” does not stand for “gyro”, because whatever would be the reason to discern the orientation of a device designed to sit on a flat surface? Instead, I think it stands for “g-force”, i.e., an accelerometer. It senses when the Smart Clock 2 has been tapped (or slapped? Punched?) by the (cranky, sleepy) owner and shuts off the alarm. Dunno if we’re going to be able to definitively locate (nearby the top-side volume switches would be a reasonable guess, methinks) and ID this one…

Onward, again. Let’s disconnect that touchscreen LCD assembly:

That’s all the further I’m going here. The display on the first-generation Smart Clock is specified as “4” HD+ (800 x 480) IPS, touch compatible”, but the one on this successor is only spec’d as being a “4.0″ LCD IPS Touchscreen” (I’m betting the resolution’s the same, and reviews concur).

Now let’s get that mic array PCB outta there:

The microphones are MEMS-based (and PCB backside-mounted), as I suspected:

And now once again to the front for a standalone (and closer) inspection:

Look back at the earlier shot of the LCD assembly separated from the main chassis and within the latter you’ll see four more screws (in addition to the four holding the speaker in place, which we’ll get to later), one in each newly revealed corner. Let’s get those out next:

And now the “guts” slide easily out of the remainder of the device chassis:

Let’s again orient ourselves at this point:

You’re looking at the front of the device; the speaker points toward and slightly downward below the backside of the display, and the (majority of the) sound exits through the mesh below the display. The flex cables for the display (to the left) and mic array PCB (right) come out the top, and the one going to the power-and-mute-switch PCB comes out the bottom.

Side views of the assembly:

Note the antenna on this side:

The bottom:

And finally, on the back, there’s the main system PCB!

Let’s free it from its captivity, unhooking the antenna connection while we’re at it:

Here’s the antenna peeled off and standalone:

The Smart Clock 2 supports 2.4 GHz-only 802.11 Wi-Fi, along with Bluetooth 4.2. I can’t find any other discrete antennae anywhere, and (foreshadowing) I don’t see any embedded in the main PCB, either, so I assume this antenna does RF double-duty for both 2.4 GHz technologies. The Smart Clock 1 precursor had handled 5 GHz Wi-Fi, too; was this a Gen2 cost-reduction decision?

Once again, onward. I’m betting that thick two-wire cable harness goes to the speaker:

I’d also wager a bet that the square hole the speaker harness goes through also acts as a rear bass port for the speaker:

And here we are, first showing the now-revealed heat sink attached to the PCB backside:

which pops right off:

Oh goodie, a Faraday Cage!

To orient (and educate) myself, I did an after-the-fact partial reassembly to remind myself how these pieces fit together in the first place. Here are the photos I snapped, which may also be useful to at least some of you:

And here’s the PCB initially-exposed side again, standalone this time:

Pressing “pause” on the inevitable Faraday cage top lid peel-off to come, I returned my attention to the transducer at this point:

The spec sheet says that it’s a “1.5″ 3 W front-firing speaker”, although to be precise, it says “speakers” (unless the other one’s invisible, that’s a typo). As comparison, the specs for the first-generation Smart Clock state “1.5″ 3 W speaker, with 2x passive radiators”:

And now, back to that Faraday cage:

Time for a closer view:

Along the top, left to right, are first two ICs with Micron logos and the following package marks:

IFP77
D9SHD

Google tells me that they’re MT41K256M16LY-093 4 Gbit DDR3-1066 SDRAMs, which jives with the Smart Clock 2 spec sheet claim that the device contains 1 GByte of total system DRAM. Next to them at far right is a Toshiba-now-Kioxia THGBMJg6C1LBAIL 8 GByte eMMC flash memory. And in the bottom row are two Mediatek SoCs. The larger one at left is the MT8167A application processor, in contrast to Lenovo’s product spec sheet, which claims that it’s the “MT8167S Processor (1.50 GHz)”. And at right is the smaller MT6392A, information for which is quite difficult to ascertain but which appears to be a power management IC.

One other region of this side of the PCB, in the upper right corner, begs for closeup attention:

The notable IC here is Texas Instruments’ TAS5805M 23 W (stereo) and 45 W (mono, in this configuration) Class D amplifier. Its function is unsurprising given that it’s in close proximity to one of the volume switches (above it; the other one is in the upper left corner of the PCB) and the speaker wiring harness cable below it.

Flipping the PCB back over, the only thing of note is a closeup of the connector for the flex cable that mates this PCB to the power-and-mute one you saw earlier:

I’ll close with some side views:

and an interesting link that I found during my research to the documented progress of some enterprising hackers that have attempted to tap into the Smart Clock 2’s hardware and Android 10 firmware foundations. And with that, it’s over to you for your thoughts in the comments!

—Brian Dipert is the Editor-in-Chief of the Edge AI and Vision Alliance, and a Senior Analyst at BDTI and Editor-in-Chief of InsideDSP, the company’s online newsletter.

<!–
googletag.cmd.push(function() { googletag.display(‘div-gpt-ad-native’); });
–>

The post Lenovo’s Smart Clock 2: A “charged” device that met a premature demise appeared first on EDN.

Lenovo’s Smart Clock 2: A “charged” device that met a premature demise

Become a member

Become a member

Become a subscriber

Become a sponsor