The BT3101 Bluetooth smart dimmer supports both on/off and variable-intensity dimming capabilities, including flexible programming facilities.

Inside the box you'll find a product pamphlet and owner's manual.

Here is the dimmer out of the box.

The voltage input comes in the form of a three-prong AC plug.

The switched-and-dimmed output takes the form of a two-plug socket.

The button on top handles simple on/off switching, but when pressed for a more prolonged period, dimming as well.

Here's a closeup of the backside spec markings.

The screws holding the front panel to the rest of the chassis were hidden behind easily removed rubber plugs.

They used atypical 'spanner' dual-notch screw heads.

The screws can be removed easily with the iFixit 64-bit driver kit.

With the front panel removed, you can see inside.

The metal plate will be notable. At first I thought 'Faraday shield' and indeed it may indirectly serve this purpose. I got the assemblage out of the chassis by removing the two screws holding it in place, in addition to three more holding down the primary PCB.

Its primary purpose, as a passive heat sink, became immediately obvious. That TO-220-packaged device that it's thermal-pasted and riveted to is, I believe, a TRIAC, which takes the more efficient place of the variable resistor used in early dimmer switch designs (but apparently still radiates a notable amount of heat).

The BT3101 PCB markings can be seen here.

This image gives you a bit more insight into what the PCB contains. Note, for example, the toroidal inductor that takes the place of the relay in previous models.

Now for the primary PCB backside, shown with its 'earth ground' prong already removed, which is fully exposed to view.

And here's the secondary PCB, handling Bluetooth connectivity and other digital functions, including the CSR (now Qualcomm) model 1010 single-chip Bluetooth radio with integrated microprocessor and memory at its nexus.

The 'earth ground' input prong is screw-connected to the PCB.

Although the output voltage plug was three-prong, at least on the inside, as you can see, the earth ground portion is unpopulated, not to mention un-exposed to the outside via a chassis hole. Power engineers, any ideas as to why earth ground was ignored?