Non-Standard Coulombs Ranges; Logarithmic Currents - Keithley 6517A User Manual

Electrometer
Hide thumbs Also See for 6517A:
Table of Contents

Advertisement

Front Panel Operation

2.11.4 Non-standard coulombs ranges

In its standard form, the Model 6517A has four coulombs
ranges allowing it to measure charge between 10fC and
2.1µC. Different charge measurement ranges can be used by
placing an external feedback capacitor between the
PREAMP OUT and Input HI and then placing the instrument
in the external feedback mode.
Charge is related to capacitance and voltage by the formula:
Q=CV, where Q is the charge in coulombs, C is the capaci-
tance in farads, and V is the voltage in volts. The Model
6517A display will read charge directly in units determined
by the value of C. For example, a 10µF capacitor will result
in a displayed reading of 10µC/V.
In practice, the feedback capacitor should be greater than
100pF for feedback stability and of suitable dielectric mate-
rial to ensure low leakage and low dielectric absorption.
Polystyrene, polypropylene, and Teflon dielectric capacitors
are examples of capacitor types with these desirable charac-
teristics. The capacitor should be mounted in a shielded fix-
ture like the one in Figure 2-43.
To discharge the external feedback capacitor, enable zero
check. The discharge time constant will be given by: t =
(10M Ω) (C
). Allow five time constants for discharge to
FB
within 1% of final value.

2.11.5 Logarithmic currents

The use of a diode junction in the external feedback path per-
mits a logarithmic current-to-voltage conversion. This rela-
tionship for a junction diode is given by the equation:
V = mkT/q ln(I/I
Where:
q = unit of charge (1.6022 × 10
k = Boltzmann's constant (1.3806 × 10
T = temperature (K).
The limitations in this equation center on the factors I
and RB. I
is the extrapolated current for V
O
proportional constant, m, accounts for the different character
current conduction (recombination and diffusion mecha-
nisms) within the junction, typically varying between 1 and
2. Finally, RB constitutes the ohmic bulk resistance of the di-
ode junction material. I
and RB limit the usefulness of the
O
junction diode at low and high currents respectively. The fac-
tor m introduces non-linearities between those two extremes.
Because of these limitations, most diodes have a limited
range of logarithmic behavior.
2-56
) + I
O
RB
-19
)
-23
)
, m,
O
. An empirical
O
A solution to these constraints is to use a transistor config-
ured as a "transdiode" in the feedback path, as shown in Fig-
ure 2-44. Analyzing the transistor in this configuration leads
to the relationship:
V = kT/q[ln(I/I
where h
is the current gain of the transistor.
FE
From this equation, proper selection of Q1 would require a
device with high current gain (h
over a wide range of emitter currents. Suitable devices for
this application include Analog Devices AD812 and Preci-
sion Monolithics MAT-01. Use the enclosure in Figure 2-43
to shield the device.
Frequency compensation/stabilization is accomplished by
adding a feedback capacitor, C
depends on the particular transistor being used and the max-
imum current level expected. Compensation at maximum
current is required because the dynamic impedance will be
minimum at this point. It should be noted that the response
speed at lower currents will be compromised due to the in-
creasing dynamic impedance, which is given by the follow-
ing formula:
dV
Z
kT/qI = 0.026/I(@25°C
=
------- -
=
dI
Using the above transistors, a minimum RC time constant of
100µsec at maximum input current would be used. At I
(max) of 100µA, this value would correspond to 0.4µF. Note
that at 100nA, this value would increase the RC response
time constant to 100msec. A minimum capacitance of 100pF
is recommended.
Although the input signal to this particular circuit is assumed
to be a current, conversion to voltage input could be per-
formed by placing a shunt resistor across the input. However,
the nominal voltage burden of 1mV must be considered as an
error signal that must be taken into account.
Further processing of the current response can be achieved
by using the suppress feature. For example, REL could be
enabled with a reference input current applied. For all subse-
quent currents, the natural logarithm of the ratio of the mea-
sured current to the suppressed current would then be
displayed:
V
= V
kT/q (ln (I
DISP
REL
= kT/q (ln (I
READ
= 0.26/I (ln (I
READ
) - ln(h
/(1 + h
))]
O
FE
FE
), which is maintained
FE
. The value of this capacitor
FB
)
/I
) - ln (I
/I
READ
O
REL
O
/I
))
REL
/I
)) @ 25°C
REL
IN
))

Hide quick links:

Advertisement

Table of Contents
loading

Table of Contents