On 23/05/2011 21:38, Olin Lathrop wrote:
> Yigit Turgut wrote:
>> >  Works very fine, thank you. I guess sensitivity of LDR will be optimu=
m
>> >  at the exact middle point (a resistor at exact middle  impedance of
>> >  LDR connected). Is it any different than driving a NTC thermistor ?
> Not really.  This method is for measuring any variable resistance, includ=
ing
> positive temperature coefficient thermisters.
>
> The full scale output voltage corresponds to the resistor under test goin=
g
> from 0 to infinite resistance.  Resolution is greatest at half scale, whe=
n
> the resistor under test has the same value as the pullup or pulldown
> resistor.  Resolution approaches 0 at the ends of the range.

Look at resistance scale on an old Analogue Meter and you see what he means=
..
At middle the numbers are  clear. At one end is quickly bunched to zero=20
and at the probes open the high values to infinity.
Note how 0 =3D full scale of the current scale and Infinity is 0 of=20
current scale. The readings are not linear.
The constant current source gives a linear scale. But hard to ensure the=20
constant current source really is constant. So perhaps the maths is a=20
better approach.
Old analogue meters used 1.5V for the two lower ranges and 9V, 15V or=20
22V for measuring high resistors.

I have one 1954 "meter" that uses a "magic eye" valve and bridge to=20
measure 10 Ohms to 10M Ohms in 6 ranges and a 110V fed neon relaxation=20
oscillator that "measures" 100M Ohm to 1,000M Ohm approx by flashing the=20
magic eye at 1Hz for 100M Ohm. About 0.1Hz for 1000M Ohm and 10Hz for=20
10M Ohm. (Good for testing capacitor leakage. All electrolytic caps are=20
"poor", you need 4 x high voltage Tantalums in series to get anywhere=20
near as slow as 1Hz!)


--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.