[time-nuts] Spectracom 8140 distribution amplifier and the Taps- some application info

Bob Camp lists at rtty.us
Wed Jul 27 21:19:43 UTC 2011


Hi

If you:
1) Run a reasonable length of coax (say > 5 M and < 100M)
2) Neglect the 50 ohm termination
You can get quite a bit of 10 MHz RF running around on the outside of the
coax shield. Depending on grounding it may or may not be a problem.

The taps appear to deliver an ADEV of ~ 1x10^-11 at a tau of 1 second. That
of course assumes the source you are driving into the system is better than
that. 

They are designed to drive the standard input of a test instrument. There is
processing (clean up) between the standard input and the instrument's clock.
Most instruments you plug in will be quite happy with a signal that has that
level of stability.

Bob

-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of walter shawlee 2
Sent: Wednesday, July 27, 2011 3:00 PM
To: time-nuts at febo.com
Subject: [time-nuts] Spectracom 8140 distribution amplifier and the Taps-
some application info

There have been many questions posted here about the units, here are a few 
answers I have collected:

1. the 8140 can drive up to 25 total downstream 8410T taps on the combined 
outputs. The mopdels are frequency specific, and are not easily changed,
10Mhz 
is the most common. The front selector provides TTL divided waveforms from
the 
input rear clock.

2. yes, the taps are needed, as the output of the 8140 itself is a sinewave
and 
+12VDC.  using the tap allows long cable distances, and provides the
extracted, 
buffered sinewave. long cable runs of RG58 up to 500 feet are possible, with

T-taps all along the way, and all 4 outputs can have long runs. the manual
says 
a capactively coupled 50 ohm load is meant to terminate the line, but I have
not 
found it to be required so far, as the sine wave drive produces very little
ringing.

3. yes, it works WITHOUT the taps, if you add a 0.47uF good quality cap in 
series with each BNC output, run distance is then short into 50 ohms.

4. YES, taps are different, if you have a 10Mhz source, you must have 10Mhz
taps 
(8410T10). beware of the 8410TTL10, it has a TTL 10 Mhz output, not a
sinewave.

5. the operating quality of the system is excellent, do not be deterred by
it's 
modest appearance, it works just as well as the racal, fluke and hp
distribution 
amps I have and use. Late ones are much better made (they have black, not
blue 
vinyl tops), if you can score one.

6. Phasing: my measurements show approximately a 190 degree phase difference

between the input sine, and the output waveform at 10Mhz...not an issue for
me, 
but maybe for others?

7. The system checks for DC output shorts (excessive loading) and loss of
input 
clock, these drive the red front indicator (output fault). An option can
also 
drive a sonalert.

8. the taps are not generally repairable, as they are potted with foam.
bummer.

9. System details, and downloads are here at Spectracom:
http://www.spectracomcorp.com/ProductsServices/TimingSynchronization/Timeand
FrequencyDistribution/8140FrequencyDistributionSystem/tabid/133/Default.aspx

10. Note, there are no schematics for Taps, if anybody has some to send me,
I'd 
really appreciate it, just to satisfy my curiosity.

so, if you find one in good condition, it is a great addition to your lab,
for 
distributing your master reference all around the shop area to counters, 
spectrum analyzers, generators, etc.  it works with literally all common
clock 
sources, from 0.3Vrms to 2Vrms, should be a sine wave.

regards,
walter

Walter Shawlee 2, President  -  Sphere Research Corporation
3394 Sunnyside Road, Kelowna, B.C.,  CANADA    V1Z 2V4
URL: http://www.sphere.bc.ca   E-Mail: walter2 at sphere.bc.ca
Phone: +1 (250) 769-1834             Fax: +1 (250) 769-4106
             --Home of the Slide Rule Universe--
WS2:    If wishes were fishes, we'd all be covered in cod.
WS2:  We're all in one boat, no matter how it looks to you.

_______________________________________________
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.




More information about the time-nuts mailing list