[time-nuts] NIST isolation amplifiers

Claudio Girardi claudio.girardi at virgilio.it
Wed Nov 26 13:01:43 EST 2014

I have built a similar circuit some time ago (from http://www.ke5fx.com/norton.htm ) and did some experiments with different devices to see how the isolation changed. Note that the isolation is limited by the output feed back to the input via two different paths : the device collector-base impedance (Ccb) and the device base-emitter impedance (r_be//Cbe), thru the transformer. It turns out that, since the transformer is inverting (and scaling) the output signal one could add an RC network between collector and base to have the two feedback contributions cancel, achieving a perfect isolation. Without resistors in series with the emitter and/or collector, the overall base-collector impedance should be n times the overall base-emitter impedance to have the two feedback contribution cancel each other (n is the transformer turns ratio, 3 for the circuit in the link). In practice everything is frequency dependent and other terms are contributing to the isolation, but nevertheless 10/15 dB additional isolation are possible. Note that this will lower the input impedance, though.

Enclosed are the measured forward and reverse gain of that circuit with a 2N5109 and a 2N3904 (with lower current) with and without the additional parallel RC network between base and collector; with the 2N3904 you can tweak the network to achieve a null, even if that's not recommended of course; you can see the isolation change a little when the device is warming up...


----Messaggio originale----
Da: csteinmetz at yandex.com
Data: 26-nov-2014 14.52
A: "Discussion of precise time and frequency measurement"<time-nuts at febo.com>
Ogg: Re: [time-nuts] NIST isolation amplifiers

Bruce wrote:

>A single 2N2222 or equivalent transistor in a suitable circuit 
>dissipating about 200mW or so can achieve a reverse isolation of 
>35dB with distortion of around -40dBc (output +13dBm) with a gain of 
>unity, and an output impedance of 50 ohms with a PN floor of around 
>-180dBc/Hz or so.

For those wondering, I suspect Bruce had in mind something like the 
attached  (he posted the basic design a few years ago).  I built 8 
channels using toroids on FT37-61 cores.  I think the Mini-Circuits 
T622 should work, but I have not tried it.  The analyses are from my 
simulation, and the constructed unit performed similarly.  The Miller 
effect limits fan-out to about 10 for a 10MHz distribution 
amp.  [Note: the 50 ohm resistors on the outputs represent the 
external loads, and are not part of the amplifier.]

Best regards,


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.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 2N5109_buffer.png
Type: image/png
Size: 81064 bytes
Desc: not available
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20141126/4a0dcacd/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 2N3904_buffer.png
Type: image/png
Size: 92252 bytes
Desc: not available
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20141126/4a0dcacd/attachment-0003.png>

More information about the time-nuts mailing list