[time-nuts] 10 MHz clock distribution for the lab

johncroos at aol.com johncroos at aol.com
Sun Apr 21 02:21:34 UTC 2013


 Just a couple of comments from a RF guy and FMT-nut


10 MHz clock distribution for the lab

 

 In the small coax cables one might consider RG-223. This is a 1/4 inch more or less diameter cable that features two concentric braided shields. Each is a very dense weave and is silver plated. This a very high quality mil spec coax. the main drawback is that a normal BNC fitting will not pass the entire cable through the clamp nut or most of the shield through the hole in the internal ferrule. However, Pasternack, among others has the correct connectors for BNC, TNC, and type N. I have used this cable with good success to keep RF from radiating in test setup. It is way way better than even quality RG-58. I found part of a reel on eBay for next to nothing.

How many instruments? Some time ago I surveyed a number of instruments to determine how many had an actual 50 OHM input for the 10 MHz reference. Answer - some do, most don't, though many specify something like 1V rms from a 50 Ohm source.

I have been able to operate 1 spectrum analyzer, 2 Fluke synthesizers, A hp synthesizer, and a HP 5345 counter all at the same time from a LPRO RB source using Ts and only terminating at the end of the line. Actually the termination had no effect as I was not looking at phase. Most of these devices seem to present a relatively high Z load and as long as that is the case; the simple T scheme is OK as long as there are not too many and the distances between them are short in terms of wavelength. 

It is easy enough to crudely determine the input Z using a series 1 K resistor on the source side.and just measuring the voltage drop with a scope.

I did get a distribution amp as things got more complex around here, but I never had a problem with the simpler scheme either.

-john k6iql



 



More information about the Time-nuts_lists.febo.com mailing list