[time-nuts] Using commercial video amplifier for 10MHz clock distribution.

[ed breya]

I have a Hitachi VD-1000 video distribution unit that's been sitting 
unused for years, waiting until I finally get around to making it into a 
reference unit. I just dug it out and looked inside. It is full of RCA 
jack cables for jumping the signals around in different ways. I 
apparently had changed things around already, to make one section into a 
15-output distributor. I have forgotten all about it long ago, so now 
it's just a dizzying array of cabling. I must have made notes somewhere, 
that I'll have to find for when I eventually (probably fairly soon) do 
the proper conversion.

It uses CLC404 amplifiers, which appear to be older and a little noisier 
than the CLC409, but should be OK for this. Unfortunately, it uses a 
small switching supply, which I dislike for things like this, so 
ultimately will make a linear one for it.

This thing has 40 BNC connectors on the back, which would allow for lots 
of I/O and branching combinations. This is likely way more than I'll 
ever need, so the "new" plan is to incorporate some improvements, 
considering what I recently learned during my work on the frequency 
multiplier project. I'll be reducing the total fanout, and changing some 
of the outputs to have fully-floating transformer coupling, to reduce 
ground loop effects in high sensitivity applications, and also 
independent volume controls for some. I may do the same on the source 
inputs too. I don't know how many of each thing yet, but definitely 
some. In any of the BNC spots, I can easily put an isolating type BNC, 
or a volume control pot. The front panel is mostly blank, so there's 
lots of room for even more stuff there. I just have to be careful to not 
get carried away and make it too wonderful and complicated and never 
finished.

I also have an Efratom something-something distribution rack with five 
"MBF" modules that have four outputs each. They should pretty good too, 
but unfortunately it's all for 5 MHz only - there's low- or band-pass 
filtering built into each channel, so needs reverse engineering and mods 
for 10 MHz. I vaguely recall looking for info on this about a year ago, 
and stumbled upon an old time nuts thread. Someone (maybe Chuck Harris?) 
had reported successful mods for 10 MHz on these, but there were no 
links to any info about it. So Chuck, if you see this, and it was you, 
or anyone else, I'd sure appreciate this info. If it was just my 
imagination, then never mind.

Ed