Where in the manual does it say that?
It doesn't say that, it's obvious from the (limited) wiring diagram.
I emailed Redarc and got this:
So, opposite of what you said.
Not opposite of what I said...
He's assuming a single shunt measuring all input/output to/from the battery. Since the bcdc ground is for powering the electronics of the bcdc unit - with a single shunt setup - it's draw should run through the shunt to be measured.
But you're not running a single shunt setup, and one of your shunts is intended to only measure what charge goes into the battery.
He didn't say where to earth the solar/alternator though.
On the charge shunt, at the end opposite the battery connection, so the charge going into the battery gets measured.
This is due to the way the bcdc is setup...only the pos runs through it. The bcdc's ground is only for the bcdc electronics.
It would be different with say an inverter/charger, where the battery neg would be connected to the unit with the same size wire as the pos.
The bcdc isn't doing that, it only handles the pos and needs no neg connection except to power itself. Same with an ACR - it only needs a neg connection to power the on-board computer and the electromagnetic solenoid.
My current theory (here we go again

) is the solar/alt should be grounded at the same contact point of the BCDC earth, on the charge shunt.
It's pretty much irrelevant where you ground it...it only carries the few watts needed to power the bcdc electronics.
If it's grounded on the common side of the battery monitor shunt, current will flow from solar/alt to the BCDC, skipping the battery monitor shunt
(Not sure what you mean by "common side". Assuming you mean "battery side"...)
Not if the alt/solar are grounded to the charge shunt on the side opposite the soc shunt.
Power would flow from alt/solar - through both charge and soc shunts - to the bcdc neg, but only a few watts and only when alt or solar are active.
The bcdc neg is not really important.
, flow across the "charge shunt" and cancel itself out with the charge current from the BCDC earth. The "charge shunt" would show somewhere near zero amps, possible negative. The battery monitor shunt would be correct, because it sees the solar/alt earth and the BCDC earth on the same side. So the BCDC "consumes" the current from the solar/alt earths, and then produces current on its own earth.
Uh...I gotta run and will have to wait till later to try and parse that.