Robin van Spaandonk wrote:
In reply to Stephen A. Lawrence's message of Thu, 19 Jan 2006
08:35:01 -0500:
Hi,
[snip]
All op-amp's have a leakage current. How much resistance do you
measure with one of the alligator clips disconnected?
Infinite. (Exactly 0, sort of ... but it wanders around a bit....)
I tweaked the voltage on the inverting input on the opamp which reads
the signal in order to zero it (see diagram below). The offset voltage
for the opamp appears to be on the order of a millivolt, which times
1000x is on the order of a volt. Without the tweak the "null reading"
was somewhere between about -1/2 and +1 volt; the exact value depended
on which particular opamp I used.
The voltage response _ought_ to be pretty much linear, so with the tweak
in place it _ought_ to read correctly for small inputs. Or at least, I
think so. Note that the voltage across the paper will be almost exactly
9 volts (<< 1 volt is dropped across R1), so a fraction of a volt offset
on the opamp side of it won't make a significant difference to the
current through the paper. And that current should show up as an
additional voltage on resistor R1 (in diagram below).
Leakage current in high-impedence circuits always confuses me but in
this case I think it's just showing up as an additional "null" voltage
across the 10M resistor, in which case it just looks like part of the
voltage offset to me.
The circuit's here (more or less):
http://physicsinsights.org/images/resistance-circuit.jpg
R1 = 10M
R3 = 100 ohms
R4 = 100K
The unnumbered capacitor is 100 pf.
The opamps are TL082's
The power supply's actually +9/0/-9 and the opamp which splits the 18
volts isn't really there.
The volt-ohm meter's an analog model from Lafayette Radio Electronics
which kind of dates me, I guess :-)
After reading Bill Beaty's comments, I took another stab at it, and
actually got a result!
With a #6B artist's pencil I drew (free-hand) two dark solid lines on a
sheet of printer paper, about 200 mm long and about 2.5 mm apart. I
checked them, and each line was electrically "intact" (#6 pencil marks
have a tendency to leave electrical gaps if one isn't careful). I
connected to the marks with alligator clips -- which were verified to
make decent contact in these conditions -- and put the paper in series
with a 10M resistor (paralleled with a 100pf capacitor), with 9 volts
across the lot.
I read out the voltage on the resistor through an op-amp, amplified
1000x, and got a reading of about 3.5 volts, which wandered around a
lot. Breathing on the paper sent the reading off-scale; it recovered
slowly. Laying a thumb across the marks shot it off-scale as well.
3.5 volts works out to a resistance of about 27 gigohms.
Dividing that by the width of the strip and multiplying by the length,
that comes out to a (surface) resistivity of about 2000 gigohms for 20
pound 100% recycled ink-jet paper.
At some point I'll try it again with somewhat better-controlled
conditions, including trying to get connections to _both_ sides of the
paper rather than just one side -- though that shouldn't actually make
much difference.
Regards,
Robin van Spaandonk
http://users.bigpond.net.au/rvanspaa/
Competition provides the motivation,
Cooperation provides the means.
