1. Parallel LNBs I only know from Unicable/JESS technology.
They are not so common yet in America? I got my knowlegde about them from some german forums; they are the experts IMHO.
With multiple (parallel) Unicable/JESS LNBs, a splitter/combiner is needed, and usually also a power inserter (as 2 LNBs sometimes need more current than 1 receiver can supply).
Oh I once saw a post from someone that wanted it tested: 2 LNBs parallel, without a diseqc switch. I never read if that test was put to practice.
I expect it might work a bit (for part of the frequencies), but not with any reliability.
2. Parallel diseqc client devices (switches, motors): for that you need a
power pass in both directions splitter/combiner.
I've got it working with a 5-1000 MHz splitter that I had lying around (instead of a x-2150 MHz or x-2300 MHz splitter)!
A diy combining of 3 cables, to form your own splitter/combiner, also works (I've had it); though it will have a resulting impedance mismatch, so it isn't a 'neat' solution.
3. Parallel receivers need indeed protection in the splitter: the positive current of the one receiver must not reach the F-connection of the other receiver.
The diodes in your
power pass in one direction splitter do exactly that.
The receiver with the highest voltage 'wins' (so 18 volt wins over 13 volt), and 22kHz ON wins over 22kHz OFF.
Theoretically if both supply 22kHz, there could be interference of the two 22kHz signals, causing the 22kHz to extinct. But in practice, I don't believe that happens (or very very very rarely), I've never seen it reported in a forum.
For running two receivers parallel simultaneously, you could also use a splitter/tap with
power pass on one port, and power block (capacitor) on the other port. They are also available (see internet).
In that case you have one fixed 'master' receiver, and one fixed 'slave'.
A
sat priority switch (master/slave switch) is also available, for running two receivers
NOT simultaneously.
Usually they will work also when the two receivers are both ON simultaneously, but then the slave port will have relevant dB-loss (as there really is diseqc-like signal switching going on inside, with PIN-diodes, it seems). You can also find those on the internet.
Alas on many simple splitters, there's no clear indication or decription if they are power pass or power blocked.
Usually splitters with indication '5-1000 MHz' are power blocked here in Europe, but not all! With other splitters meant for several purposes, it isn't clearly indicated either.
So I need to measure them with a ohmmeter, to know what I've got...
Greetz,
A33