[RedDevil]

The control box may be temporary after all, as it is really crowded inside.
Not that it matters much as everything is isolated and fixed.

I also put a 12V fuse in there, but I expect it will never blow so I made it inaccessible. If it ever blows I would take the whole thing apart anyway.

I expected to do all this in two hours, but always testing every step and one or two minor mistakes like soldering a connector on the wrong wire (it happens just when, and only when, you start to believe it is impossible to mix it up) made me spend all evening.

But here goes.
The 3 battery boxes have been hooked up to the control unit so it feeds its internals with the combined voltage.
I hooked up my Fluke to the two terminals - which are a good 2 inches apart, even if they seem to touch in the photo - that's just line of sight.

So, here is the system hooked up to 12V and switched OFF:

The internal volt meter gets power and shows the battery voltage.
The relays are not powered so the terminals are dead.

Flip that switch to ON:

There you have it. Power on the terminals.
My volt meter does have a slight deviation after all. Oh well.

Now I switch off the 12V, like when I key off in the car:

Control unit is dead, relays are off (of course). The terminals are disconnected, the reading shows static noise of a few millivolt.

Once the harness is installed in my car I can flip the switch with the batteries disconnected to check the IMA system voltage.
Then I switch it off and hook up the batteries.
That way I know what the voltage difference is between both battery systems, so I can anticipate on the current flow (the amperage would be 1.2 times the voltage differential, max) and decide whether it is safe to flip the switch to connect both batteries or not.

A 25 Volt difference between both packs for 30 Ampere would blow the fuse. I feel no need to test that in a hurry.