[sci.electronics] Log amps, sqare root amps

gmg@hcx.uucp (Greg M. Garner) (10/16/88)

I am currently working on a video project, and I need a logarythmic amplifier
or a square root amplifier. The catch is that I need it to run at 5-10 MHZ. 
Analog devices has one they specify to work at 10 MHZ to 250 MHZ, but I am 
not sure it is perfect for the job. Any body out there know of any others
that I could use?

  Greg Garner
  gmg@hcx.uucp
  501-442-4847

/* Come on guys, lets get busy and build a FTL drive! */

commgrp@silver.bacs.indiana.edu (10/18/88)

>I am currently working on a video project, and I need a 
>logarythmic amplifier or a square root amplifier. The catch is 
>that I need it to run at 5-10 MHZ.  Analog devices has one they 
>specify to work at 10 MHZ to 250 MHZ, but I am  not sure it is 
>perfect for the job. Any body out there know of any others that I 
>could use?
>
>  Greg Garner
>  gmg@hcx.uucp
>  501-442-4847

I don't know where to get the device you need, but here's 
a precaution when using log and other nonlinear amps:

The amplifier's gain-bandwidth product is constant, the gain 
varies depending upon input signal level, therefore the bandwidth 
varies in inverse proportion.  Inspect manufacturers' data sheets 
cary carefully; their bandwith specification is probably at unity 
gain.

--

Frank
reid@gold.bacs.indiana.edu

ornitz@kodak.UUCP (barry ornitz) (10/18/88)

In article <913@cseg.uucp> gmg@hcx.uucp (Greg M. Garner) writes that he is
looking for a high speed log amplifier.

As Frank Reid has pointed out, the simple diode in the feedback path of an
op-amp design is limited due to gain-bandwidth considerations.  This circuit
is often known as the transdiode configuration.  The main problem with this
circuit is one of closed-loop stability; since gain is a function of signal
level, there is a choice in the design between stability at high signal levels
and bandwidth at low signal levels.

For a good discussion of this problem consult:

   "Nonlinear Circuits Handbook", ed. Daniel Sheingold, Analog Devices, Inc.
   Norwood, MA, 1974.

This book is recommended highly for assistance in your application.  It also
discusses analog multiplication and powers and roots.  You may find exactly
what you need in this book.

The wideband logarithmic amplifier made by AD (AD9521) is likely not what you
are looking for.  This circuit should more properly be known as a wide band
log detector.  It is used in many radar applications where its DETECTED output
provides a DC current logarithmically proportional to the AC signal level at
the input.  Plessey Semiconductors make a number of circuits similar to these.
I do not know if it is still made but the Plessey SL1531  *might*  do what you
need.  This is a five stage log IF strip with a 50 db dynamic range with +/-
1 db error.  Its typical frequency range is 10 to 160 MHz, however.  Some other
Plessey devices cover the 3 to 500 MHz ranges but have much smaller dynamic
range.  You should also be able to find a number of discrete and hybrid log
amplifiers intended for use in radar applications where compression of the IF
frequency is needed.  Check any of the numerous microwave journals for ads.

BTW, Plessey makes a number of interesting IC's that I would love to play with.
Unfortunately, I have never found a source for small quantities of these.  Good
luck with the project.
                                            Barry  WA4VZQ

 -----------------
|  ___  ________  |
| |  / /        | |  Dr. Barry L. Ornitz  UUCP:..rutgers!rochester!kodak!ornitz
| | / /         | |  Eastman Kodak Company
| |< < K O D A K| |  Eastman Chemicals Division Research Laboratories
| | \ \         | |  P. O. Box 1972
| |__\ \________| |  Kingsport, TN  37662       615/229-4904
|                 |
 -----------------