Having a second hicap on a 282 might break the principles of star-earthing, but connecting a Supercap to a 282 using two Snaics creates a ground loop because each Snaic provides it’s own path to ground.
When connecting to a 282, here are the connections used on the Supercap
Here are the sockets the Supercap connects to on the 282
Socket 1 on the Supercap connects to “Upgrade 1” on the 282. Socket 2 on the Supercap connects to “Upgrade 2” on the 282. Note both Supercap sockets include a 0v (or ground) connection. We know that the “Upgrade 1” socket must include a ground connection because that’s where a Hicap would connect to in the case of a single Hicap. It must also be the case that the “Uograde 2” socket must also include a ground connection because that’s the socket a 2nd Hicap would connect to and the 2nd Hicap wouldn’t work if it didn’t include a ground connection back to the 282. Just for confirmation though, here’s the wiring of the “Upgrade 2” socket.
(Green wire is the ground connection.
Is this dual connection for the ground path a serious issue? Well, here’s what the Forum FAQ on the the use of DIN plugs says about ground loops:
https://community.naimaudio.com/t/why-naim-use-din-connectors/63
Many manufacturers point to the great trouble they take to “star ground” everything. Sadly, this is all wasted when you connect your system together with RCA-plugged cables. Why?
When you connect, for instance, a CD player to a preamp with RCA-plugged cable, you automatically have two separate ground wires - the left and right shields going between them. This creates a ground loop, which degrades the musical performance dramatically, and negates any efforts that were taken to ground the internal circuits properly.
If you were to connect these same two components together with DIN-plugged Naim interconnects, you would have only ONE cable with only ONE ground shield surrounding both the left and right signal wires. Hence, only one ground path for each connection and no ground loop.
So there does appear to be a performance compromise here when using two Snaic connections from Supercap to 282. From my own findings comparing a HCDR to a freshly recapped Supercap-2, it was like the Supercap was two steps forward, sonically, but 1 step back musically. As impressive as it sounded, I just wasn’t getting the raw emotional involvement with the Supercap that I was with the Hicap. That’s why I did a bit of digging into the above. So after a bit of thought I arranged things so the ground loop didn’t exist, and bam, there it was, the involvement was back.