Devices which operate in A24 address space have an offset register located in their A16 configuration register space that gets set by the command module. My multimeters however are purely register based. So, everything about them is contained within their A16 memory area and there is nothing to configure.

I'm thinking, the issue could be
- related to the physical bus: For example some pull-up resistor missing somewhere on a control line.
- power management: the PSU in the mainframe has a bit of smarts to it and does a power up procedure. Maybe this absolutely needs to be observed for the individual modules to behave properly

And I'm wondering if I can figure out more by using the system itself. Find out why exactly it thinks there is a configuration error and what exactly is misconfigured.

The command module provides me with an extensive set of commands I can use to poke around. And since I need to familiarize myself with it anyway, let's explore that now

The manual for the command module.

After start up the system seems to be in some form of bootloader context. The number of commands that it would respond to there is very limited. One command is the terminal configuration command ST.
You need to enter the context of an instrument first, for example the context of SYSTEM, the command module. You do so by using the SI command (select instrument):
SI SYSTEM

The bar on the bottom shows you what actions the function keys currently perform. They are just shortcuts for written commands. On boot up it will provide you with a list of available instruments. Notice how the multimeter is missing there. It should already appear there.

Let's enter the SYSTEM.

Calls to
VXI:CONF:DLAD?
VXI:CONF:DLIS? 0
VXI:CONF:DLIS? 16
Using function keys F1 and F2.

It recognizes there being an instrument at LADDR 16

So, let's understand the output of DLIS?

The Command module:
- has device ID 0
- has no commander
- is made by manufacturer 4095 (guess that's HP)
- has the model code 1306 (HP E1306A)
- is in an unknown slot. For B sized mainframes this is always the case.
- has "Slot 0 Logical Address" of 0. Whatever that is
- is a HYBRID device
- takes up no memory space on the VXI bus
- has a memory offset of 0
- has a memory size of 0
- reports it's READY to receive commands
- not used
- not used
- not used
- the last field can contain strings. It lets us know there's been error 13

The multimeter:
- has device ID 16
- has device 0 as its commander
- is made by manufacturer 4095 (guess that's HP)
- has the model code 65344. It's an HP E1326B, so something seems odd there.
- is in an unknown slot. For B sized mainframes this is always the case.
- has "Slot 0 Logical Address" of 0. Odd that this is also 0
- is a REGISTER based device
- lives within the A16 address space of the VXI bus
- has a memory offset of 0
- has a memory size of 0
- reports it's READY to receive commands
- not used
- not used
- not used
- the last field can contain strings. It lets us know there's been error 13

Now, let's access those configuration registers

User manual for the HP E1326B and HP E1411B multimeters

Annoyingly enough, it returns the register contents as signed decimals. Ugh

And there is no way to change that.

Poking around in software I can't find a reason why Error 13 occurs. There's got to be a hardware difference between the VXI mainframe and this plain VME backplane, despite the standard saying that VXI is a superset of VME.

So let's look at the VXI frame again.

There's a little board in the VXI frame that supplies the ACFAIL* signal. Just two LM339 quad comparators. Disconnecting it does not result in the same behavior as the plain VME backplane.