[time-nuts] RAPCO 1882M-2 autochance over distribution amplifier

[GandalfG8 at aol.com]

In a message dated 22/10/2010 14:09:13 GMT Daylight Time,  
luciano.paramithiotti at hp.com writes:

I bought  a distribution amplifier Rapco model 1882M-2 auto changeover on 
ebay. This  distribution amplifier was intended to be used in conjunction 
with a GPSDO to  distribute 5MHz and 1PPS.

It has two inputs for 5MHz and and two for  1PPS, and have nr14 5Mhz 
outputs and two 1PPS. Switching between the two  couple of input is automatic. 
Normally each master input works, when the the  master drop,the input source is 
switched to the second input.

This  equipment uses wideband amplifiers  with ft100 MHz and I verified 
that it  works perfectly with no attenuation at 10 Mhz. The gain of the Rapco 
is 1 and  is not adjustable.

I noticed, however, that without the input signal,  at the exit, there is 
noise due to the two internal switching regulators that  generate -6 and +11 
volts from +24 vdc input. The noise at the 5 Mhz output is  60mV pp(sum of 
the two switching noise).

I reduced the noise to 10mVpp,  cuttingin two tracks on pcb, from power 
output from the switchings and  inserting an inductance 300uH (.8 A)than 
putting an electrolytic 470UF 16V.  from inductance to ground (respecting their  
polarity).




---------------
Hi Luciano
 
Thanks for the information, I'd been planning on replacing those two  
switching regulators with linear but that's not so straightforward for the  
negative supply so might experiment with your suggestion.
 
>From the description in the manual I guessed in advance these might be  
wideband and took a gamble on that, once confirmed on my first one I  bought a 
couple more:-)
 
Aside from frequency distribution I was also curious to try them as  LF/MF 
antenna distribution units.
I knew they were likely to be quite noisy at microvolt levels but  there's 
quite a bit of noise at LF anyway and the op amp noise actually  wasn't too 
bad.
However, when feeding into a radio with a 30MHz wide spectrum  display the 
effects of those switching regulators are immediately obvious.  They're 
running at 100KHz and the harmonics generate a nice picket fence that's  clearly 
visible up to at least 20MHz.
 
I'd still like to get rid of those switching regulators altogether but  
could really do with a schematic before trying that.
 
I assume you got a copy of the basic user manual from the seller? If not  
let me know and I'll send it to you.
 
regards
 
Nigel