10 MHz over unshielded twisted pair works very well. That's what Ethernet 10BaseT is after all. Either scrounge some pulse com transformers out of ancient Ethernet cards or use a pair of 'video baluns' which are sold into the closed circuit television industry for transporting video over cat5 cable. Dale NV8U
Sent from my iPhone On Apr 18, 2013, at 4:35, Iain Young <[email protected]> wrote: > Hi Everyone, > > Recently, I have acquired a HP Frequency Counter and Signal Genny, and > have set up a small "lab" in the house. This is great, but I'd like to > hook it into my 3816A, which is 70 ft away in an outhouse, along with > all my radio gear, to at least compare it to the 10811 in the Frequency > Counter. > > > I'd rather not drill a hole and run a cable (There are other issues > with that as well as the hole, the outhouse is the other side of the > garden path from the lab!) > > I do have fibre to the house for N/W connectivity, and (unshielded) > CAT6 from the patch panel to the "lab". > > Two problems here. One the patch panel is the other side of the house > from the lab (so running a dedicated piece of coax is out without > taking up the floors..), and Two, 10MHz over unshielded CAT6 is not > good practice, to say the least, and simply not going to happen. > > So I started looking at other possibilities. It seems a lot of GPSDOs > steer the Oscillator by using the PPS. Is that right ? 1 Hz over UTP is > a bit more reasonable than 10 MHz, but I did not find many 10MHz > oscillators with a PPS input. > > > I thought of using a Z3801 instead of the Z3816, but patching out from > the EFC SBM connector, then (optionally) converting that to fibre, > sending it up the garden to the house, converting back to copper, then > the CAT 6 to the Lab where a second Z3801 would sit > > I would rather fibre between the house and outhouse for EMC and > grounding reasons. My hope is that thee 10MHz Osc would then be steered > from the remote Z3801, although the lab Z3801 itself would complain > bitterly about no lock no doubt. > > > Does anyone have any comments on this madhat scheme ? Or have other > suggestions of how I might go about getting that 10MHz signal > converted to fibre, and back again to send into the "lab" equipment ? > What are other people's experience with similar issues ? > > What do the "big boys" like NIST and NPL do to manage this ? I know > they transfer time over large distances, and I know NPL transfer > frequency as well to certain customers, so I guess other similar > institutions do as well > > > [Note, for me, plug and play is better. Soldering irons do not like me, > and I wouldn't trust myself with one anywhere near anything like a > precision instrument :), although putting pre-built modules in a metal > box I'm okay with, but plug-and play preferred.] > > (Googling for fibre converters or similar these days brings up such > a noise floor of Ethernet, Any suggestions on the best terms > or part numbers to use to find raw (assembled) fibre transmitter / > receiver modules that might be suitable would be gratefully received) > > > Best Regards > > Iain > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
