linear output to logarithmic output?

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
perfectbite
Posts: 255
Joined: Tue Jan 07, 2003 1:01 am
Location: California
Contact:

linear output to logarithmic output?

Post by perfectbite »

Is there a relatively simple way (I'm a simple kind of guy) to make a linearly scaled output produce the range (and sensitivity) of a logarithmic scaled output even just over a limited, delineated, linear span. Or are the scales mutually exclusive? Thanks.<p>[ July 09, 2004: Message edited by: perfectbite ]</p>
Ron H
Posts: 360
Joined: Tue Dec 18, 2001 1:01 am
Location: Boise, ID
Contact:

Re: linear output to logarithmic output?

Post by Ron H »

Do a Google search for "logarithmic amplifier". Also "log amp".<p>[ July 09, 2004: Message edited by: RonH ]</p>
User avatar
haklesup
Posts: 3139
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: linear output to logarithmic output?

Post by haklesup »

For simplicity you cant beat a diode and a resistor in series.<p>Apply a linear voltage sweep across a series connected resistor and diode. You will get a non linear current response which happens to be logrythmic (a diode characteristic plotted on a log scale has a straight line)<p>Measure the voltage across the resistor while you sweep the input voltage and you have your linear-log voltage voltage converter.<p>From there you can apply that voltage to the inputs of an op amp (differential or comparator mode) to get a single ended voltage relative to ground or whatever reference you want.<p>This also happens to be the crux of curve tracer operation.
Ron H
Posts: 360
Joined: Tue Dec 18, 2001 1:01 am
Location: Boise, ID
Contact:

Re: linear output to logarithmic output?

Post by Ron H »

Actually, the voltage across the diode is proportional to the log of the current through the diode.
The diode equation:<p>I=Is(e^(qV/kT)-1),
where Is is the diode saturation (reverse leakage) current, q is the charge on an electron, k is Boltzmann's constant, and T is temperature in degrees Kelvin.<p>Ignoring the -1 (insignificant when I>>Is)<p>qV/kT=ln(I/Is)<p>The resistor/diode scheme sorta works if you measure the voltage across the diode instead of the resistor, but it is inaccurate until the voltage across the resistor begins to look like a voltage-to-current converter, i.e., the diode impedance is much less than the value of the resistor.<p>[ July 09, 2004: Message edited by: RonH ]</p>
dyarker
Posts: 1920
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Re: linear output to logarithmic output?

Post by dyarker »

Is the project an analog circuit, or a microprocessor? You've got analog suggestions (I favor op amp approach). In a uP use a look-up table and interpulate for the in-between values.<p>Cheers,
Dale Y
Post Reply

Who is online

Users browsing this forum: No registered users and 157 guests