[time-nuts] Firmware and antenna for Stanford Research FS700

[Charles Steinmetz]

Dave wrote:

>Do you really mean the ZTX849? It is a power transistor (5A continuous
>current collector current, 20 A peak collector current), with no
>specification for noise on the data sheet. It would on the face of it
>hardly seem a wise choice for a low noise amplifier, but perhaps you know
>something I don't.

Yep.  They are not marketed (yet) for low noise amplification, but 
they have about the lowest voltage noise you can get while at the 
same time not having huge junction capacitances and exhibiting good 
current gain down to 1mA and below.  Innovative process features that 
were developed to provide very low saturation voltage also produced 
benefits that were not targeted by the designers -- in this case, 
very low voltage noise.

Attila wrote:

>I always wonder how you figure out whether a transistor is low noise
>or not. What part of the datasheet hints at which transistors have low
>noise and which have not? Even if it's just try and measure, how
>do you find good candidates to measure?

Noise in BJTs is well correlated with the "base spreading 
resistance," Rbb.  But Rbb is almost never given on datasheets, 
except in the case of a very few transistors marketed specifically as 
low-noise amplifiers (e.g., the obsolete Rohm 2SB737 and 2SD786).  It 
is well known that Rbb goes down as the die size increases, so large 
power transistors have sometimes been used as low-noise, small-signal 
amplifiers.  However, the current gain of 20 amp power transistors 
(TO-3, TO-220) generally falls off severely at the low currents used 
for small-signal amplifiers, and the large die have very large 
junction capacitances, which severely limits bandwidth.  Zetex makes 
a series of high-current, low Vsat transistors on much smaller die, 
using a new process that also maintains high current gain below 10 
mA.  I've been using the 849s as low-noise amplifiers since last 
century, and have recommended them a number of times both here and on 
the volt-nuts list.  As those who have read the new 3rd edition of 
The Art of Electronics know, the authors recommend the 849's cousin, 
ZTX/FZT851, in this same role.

One finds good candidates to measure by applying an understanding of 
basic solid state physics to what manufacturers *do* say about their 
products.  Or now, by reading the 3rd edition of The Art of 
Electronics.  (Like its predecessors, I consider this an absolutely 
indispensable book for anyone who dabbles with circuit design, from 
the newbie nimrod to the crusty old designer who has forgotten more 
than any newly-minted PhD/EE knows.)

I posted a lab note on measuring base spreading resistance to 
Didier's site long ago:

<http://www.ko4bb.com/manuals/download.php?file=06_Misc_Test_Equipment/3_Manuals_to_be_sorted/Measuring_base_spreading_resistance.pdf>


Best regards,

Charles