where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
true-log amp
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

1705 views
09:33 Aug-22-2001

Guido Crapels

NDT Inspector
Nuson Inspection Services bv,
Netherlands,
Joined Oct 2000
1
true-log amp


Please can anyone help me with info regarding the electronics of true-logarithmic amplification of ultrasonic RF-signals, preferably beyond 100 MHz bandwidth ? Any useful info is welcome, also concerning commercial applications. I already have been looking into matters, but most info is coming from radio and video applications and imply demodulating electronics, which is something I do not want to do ! Thanks in advance.



 
00:25 Aug-23-2001
Pim van Andel
Re: true-log amp :
: Please can anyone help me with info regarding the electronics of true-logarithmic amplification of ultrasonic RF-signals, preferably beyond 100 MHz bandwidth ? Any useful info is welcome, also concerning commercial applications. I already have been looking into matters, but most info is coming from radio and video applications and imply demodulating electronics, which is something I do not want to do ! Thanks in advance.

E.g. Analog devices AD640, yet accuracy may be insufficient.



 
03:37 Aug-24-2001

Bruce Maxfield

Engineering, electronics and instrumentation
Analogic Engineering, Inc.,
USA,
Joined Jan 2000
10
Re: true-log amp : :
: : Please can anyone help me with info regarding the electronics of true-logarithmic amplification of ultrasonic RF-signals, preferably beyond 100 MHz bandwidth ? Any useful info is welcome, also concerning commercial applications. I already have been looking into matters, but most info is coming from radio and video applications and imply demodulating electronics, which is something I do not want to do ! Thanks in advance.
.
: E.g. Analog devices AD640, yet accuracy may be insufficient.
.

Although the AD 640 produces a very useful log video output, I believe that the actual RF signal output is not a log function of the input. I have not checked the AD 640 data sheet but I suggest anyone thinking of using the AD640 as an RF log amp check this carefully before proceeding.



 
02:32 Sep-01-2001
Joerg Schulze-Clewing
Re: true-log amp : : :
: : : Please can anyone help me with info regarding the electronics of true-logarithmic amplification of ultrasonic RF-signals, preferably beyond 100 MHz bandwidth ? Any useful info is welcome, also concerning commercial applications. I already have been looking into matters, but most info is coming from radio and video applications and imply demodulating electronics, which is something I do not want to do ! Thanks in advance.
: .
: : E.g. Analog devices AD640, yet accuracy may be insufficient.
: .
.
: Although the AD 640 produces a very useful log video output, I believe that the actual RF signal output is not a log function of the input. I have not checked the AD 640 data sheet but I suggest anyone thinking of using the AD640 as an RF log amp check this carefully before proceeding.
.
The AD640 is a multi-stage log amp, pretty much like
the ones we used to design discrete in the 80's. I have
used it in a medical ultrasound system years ago and it
performed very well.The log amp is generally part of the
detector (RF-to-baseband) section, a placement within
the RF section wouldn't make much sense.

Two AD 640 can be cascaded to give a range somewhat above
90 dB, and freq response is ok to 100 MHz (maybe higher
but I've never used it above that).

Only downside I found with it is that it is single
sources (but so was the Texas 440), and in the late 90's
it was quite pricey.

One word of caution: As with any log-amp it is very, very
crucial to maintain the utmost in feedback avoidance.
Solid, common ground and clever layout is important or
the whole circuit will oscillate. The main reason for
the log amp being so sensitive is that a single device
alrady has 50 dB of gain, all the way into the VHF band.
I usually bury the RF and output traces between circuit
board V+ and ground layers, and use SMT parts exclusively.

All the best, Joerg.


 
05:20 Sep-04-2001

Thomas Murphy

NDT Inspector
USA,
Joined Jul 2000
16
Log amp instrument Hello,

My company currently uses a field friendly instrument (log amp) with a dynamic range of 80 dB. These instruments were made by ourselves and we are interested in replacing them at some point with off the shelf instruments. If no alternative is available we would consider building another set of these custom instruments. Ideally the replacement instrument would have the characteristics listed below. If anyone knows of, or sells such an instrument please contact me via email.

Regards,
Tom

Dynamic range: 80dB
Pulse voltage >300v fast spike or tunable square wave
Broad banded and narrow banded operation
Variable Rep rate as low as 60 hz
Field portable, not laboratory instrument:
calibrated sweep/delay, or ability to zoom in on part of the sweep

: Please can anyone help me with info regarding the electronics of true-logarithmic amplification of ultrasonic RF-signals, preferably beyond 100 MHz bandwidth ? Any useful info is welcome, also concerning commercial applications. I already have been looking into matters, but most info is coming from radio and video applications and imply demodulating electronics, which is something I do not want to do ! Thanks in advance.
.



 
09:03 Sep-06-2001
chris gartside
Re: true-log amp Our company has manufactured for several years a PC expansion board
Pulser Receiver with a log detector. This is used in systems principally
on composites and highly attenuating thick metal structures (eg nuclear
fabrications) where a large dynamic range is needed. I use the term
"log detector" because this is different from a true "log amplifier".
The AD640 is a log detector. I doubt whether you can get a usable 90dB
dyanmic range by coupling two together, unless your input signals reach
4V and the input noise is extremely low (ie lower than we have ever
been able to achieve, even using the absolute best practice). I think 80dB
is a more likely maximum.

I also find it hard to understand the use of a log amp at high frequencies.
usually at high frequencies near surface and axial resolution is paramount
and we find it difficult to use log amps at frequencies over 15MHz. This
is from practical considerations in industrial systems - it doesn't necessarily
apply to an R&D application.



 
03:22 Nov-07-2001
Marek Stepan
Re: true-log amp Sorry, but Analog devices doesn#t manufacture
true log-amplifiers.
One chip, whch I have found is for example ICL8048
from www.intersil.com
Analog devices makes only log-anps,
AND NOT true-log-amps.



 
08:37 Nov-28-2002
Shashank Dwivedi
Re: true-log amp Theory of true log amp is anyway different from that of demodulating log amps, what most log amps from Analog Devices offer. A normal log amp will actually perform a logarithmic function on the input signal, which means that the output will contain only positive signal since logarithm of any negative number is not defined; just like the definition of a mod of any real number.
On contrary, a true log amp have a symmetric transfer characteristic i.e., the log function also applied for the negative going input signal. Unlike demodulating log amps, it will not detect the signal but will provide with an ideal log output of input RF/pulsed signal.


 
07:42 Nov-30-2002

rodney fordham

Consultant
United Kingdom,
Joined Feb 2002
12
Re: true-log amp the logarithm of zero is minus infinity !!!

the logarithm of a negative number is 'not defined'

arithmetic can be performed with the logarithm
of the modulus of a negative number.

it all depends upon what you are trying to do ???



 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|