MAX4206
Precision Transimpedance Logarithmic
Amplifier with Over 5 Decades of Dynamic Range
10 ______________________________________________________________________________________
Detailed Description
Theory
Figure 2 shows a simplified model of a logarithmic
amplifier. Two transistors convert the currents applied
at LOGIIN and REFIIN to logarithmic voltages accord-
ing to the following equation:
where:
V
BE
= base-emitter voltage of a bipolar transistor
k = 1.381 x 10
-23
J/K
T = absolute temperature (K)
q = 1.602 x 10
–19
C
I
C
= collector current
I
S
= reverse saturation current
The logarithmic amplifier compares V
BE1
to the refer-
ence voltage V
BE2
, which is a logarithmic voltage for a
known reference current, I
REF
. The temperature depen-
dencies of a logarithmic amplifier relate to the thermal
voltage, (kT/q), and I
S
. Matched transistors eliminate
the I
S
temperature dependence of the amplifier in the
following manner:
VVV
kT
q
I
I
kT
q
I
I
kT
q
I
I
I
I
kT
q
I
I
OUT BE BE
LOG
S
REF
S
LOG
S
REF
S
LOG
REF
=
=
=
=
12
ln ln
ln ln
ln
=
()
kT
q
I
I
K
I
I
see Figure
LOG
REF
LOG
REF
ln( ) log
log ( )
10
3
10
10
V
kT
q
I
I
BE
C
S
=
ln
MAX4206
LOGIIN
CMVIN
V
CC
REFIIN
V
CC
CURRENT
CORRECTION
V
CC
LOGV1
SCALE
OSADJ
LOGV2
V
CC
V
CC
REFISET
CURRENT MIRROR
CMVOUTREFVOUT
REFIOUT
GND
V
EE
0.5V
1.238V
V
EE
V
EE
SUMMING
AMPLIFIER
AND
TEMPERATURE
COMPENSATION
Figure 1. Functional Diagram
MAX4206
Precision Transimpedance Logarithmic
Amplifier with Over 5 Decades of Dynamic Range
______________________________________________________________________________________ 11
where:
K = scale factor (V/decade)
I
LOG
= the input current at LOGIIN
I
REF
= the reference current at REFIIN
The MAX4206 uses internal temperature compensation
to virtually eliminate the effects of the thermal voltage,
(kT/q), on the amplifier’s scale factor, maintaining a
constant slope over temperature.
Definitions
Transfer Function
The ideal logarithmic amplifier transfer function is:
Adjust K (see the Scale Factor section) to increase the
transfer-function slope as illustrated in Figure 3. Adjust
I
REF
using REFISET (see the Adjusting the Logarithmic
Intercept section) to shift the logarithmic intercept to the
left or right as illustrated in Figure 4.
Log Conformity
Log conformity is the maximum deviation of the
MAX4206’s output from the best-fit straight line of the
V
LOGV1
versus log (I
LOG
/I
REF
) curve. It is expressed as
a percent of the full-scale output or an output voltage.
Referred-to-Input and Referred-to-Output Errors
The log nature of the MAX4206 insures that any addi-
tive error at LOGV1 corresponds to multiplicative error
at the input, regardless of input level.
Total Error
Total error (TE) is defined as the deviation of the output
voltage, V
LOGV1
, from the ideal transfer function (see
the Transfer Function section):
Total error is a combination of the associated gain,
input offset current, input bias current, output offset
voltage, and transfer characteristic nonlinearity (log
conformity) errors:
where V
LC
and V
OSOUT
are the log conformity and out-
put offset voltages, respectively. Output offset is
defined as the offset occurring at the output of the
MAX4206 when equal currents are presented to I
LOG
and I
REF
. Because the MAX4206 is configured with
a gain of K = 0.25V/decade, a 4 should multiply the
(±V
LC
±V
OSOUT
) term, if V
LC
and V
OSOUT
were derived
from this default configuration.
VKK
II
II
VV
LOGV
LOG BIAS
REF BIAS
LC OSOUT210
1
2
14
±± ±
()
()log
VVTE
LOGV IDEAL1
VK
I
I
IDEAL
LOG
REF
log
10
LOGIIN
CMVIN
V
CC
REFIIN
V
CC
V
BE1
V
BE2
V
EE
V
EE
I
LOG
I
REF
Figure 2. Simplified Model of a Logarithmic Amplifier
IDEAL TRANSFER FUNCTION
WITH VARYING K
MAX4206 fig03
CURRENT RATIO (I
LOG
/I
REF
)
NORMALIZED OUTPUT VOLTAGE (V)
100.1
-2
-3
-1
0
1
2
3
4
-4
0.001 1000
V
OUT
= K LOG (I
LOG
/I
REF
)
K = 1
K = 0.5
K = 0.25
Figure 3. Ideal Transfer Function with Varying K
MAX4206
Precision Transimpedance Logarithmic
Amplifier with Over 5 Decades of Dynamic Range
12 ______________________________________________________________________________________
I
BIAS1
and I
BIAS2
are currents in the order of 20pA, sig-
nificantly smaller than I
LOG
and I
REF
, and can therefore
be eliminated:
Expanding this expression:
The first term of this expression is the ideal component
of V
LOGV1
. The remainder of the expression is the TE:
In the second term, one can generally remove the
products relating to K, because K is generally much
less than 1. Hence, a good approximation for TE is
given by:
As an example, consider the following situation:
Full-scale input = 5V
I
LOG
= 100µA
I
REF
= 100nA
K = 1 ±5% V/decade (note that the uncommitted ampli-
fier is configured for a gain of 4)
V
LC
= ±5mV (obtained from the Electrical Character-
istics table)
V
OSOUT
= ±2mV (typ)
T
A
= +25°C
Substituting into the total error approximation,
TE ± (1V/decade)(0.05log
10
(100µA/100nA)
±4 (±5mV ±2mV) = ±[0.15V ±4(±7mV)]
As a worst case, one finds TE ±178mV or ±3.6% of
full scale.
When expressed as a voltage, TE increases in proportion
with an increase in gain as the contributing errors are
defined at a specific gain. Calibration using a look-up
table eliminates the effects of gain and output offset
errors, leaving conformity error as the only factor con-
tributing to total error. For further accuracy, consider tem-
perature monitoring as part of the calibration process.
Applications Information
Input Current Range
Five decades of input current across a 10nA to 1mA
range are acceptable for I
LOG
and I
REF
. The effects of
leakage currents increase as I
LOG
and I
REF
fall below
10nA. Bandwidth decreases at low I
LOG
values (see
the Frequency Response and Noise Considerations
section). As I
LOG
and I
REF
increase to 1mA or higher,
transistors become less logarithmic in nature. The
MAX4206 incorporates leakage current compensation
and high-current correction circuits to compensate for
these errors.
Frequency Compensation
The MAX4206’s frequency response is a function of the
input current magnitude and the selected compensation
network at LOGIIN and REFIIN. The compensation net-
work comprised of C
COMP
and R
COMP
ensures stability
over the specified range of input currents by introducing
an additional pole/zero to the system. For the typical
application, select C
COMP
= 100pF and R
COMP
= 100.
Where high bandwidth at low current is required, C
COMP
= 32pF and R
COMP
= 330 are suitable compen-
sation values.
TE K K
I
I
VV
LOG
REF
LC OSOUT
±
±± ±
()
log
10
4
TE K K
I
I
KKVV
LOG
REF
LC OSOUT
±
±±±±
()
∆∆log ( )
10
41
VK
I
I
KK
I
I
KKVV
LOGV
LOG
REF
LOG
REF
LC OSOUT
210 10
41
±
±±±±
()
log log
( )
VKK
I
I
VV
LOGV
LOG
REF
LC OSOUT210
14 ±
±± ±
()
()log
IDEAL TRANSFER FUNCTION
WITH VARYING I
REF
MAX4206 fig04
I
LOG
(A)
OUTPUT VOLTAGE (V)
100µ
1µ 10µ100n10n
-1.0
-0.5
0
0.5
1.0
1.5
-1.5
1n 1m
I
REF
= 100µA
I
REF
= 1µA
I
REF
= 10nA
Figure 4. Ideal Transfer Function with Varying I
REF

MAX4206ETE+T

Mfr. #:
Manufacturer:
Maxim Integrated
Description:
Logarithmic Amplifiers Transimpedance w/ 100Db Dynamic Range
Lifecycle:
New from this manufacturer.
Delivery:
DHL FedEx Ups TNT EMS
Payment:
T/T Paypal Visa MoneyGram Western Union

Products related to this Datasheet