LM134 Series
APPLICATIO S I FOR ATIO
Basic Theory of Operation
The equivalent circuit of the LM134 is shown in Figure 1.
A reference voltage of 64mV is applied to the minus input
of A1 with respect to the V
–
pin. A1 serves the drive to Q2
to keep the R pin at 64mV, independent of the value of
R
SET
. Transistor Q1 is matched to Q2 at a 17:1 ratio so that
the current flowing out of the V
–
pin is always 1/18 of the
total current into the V
+
pin. This total current is called I
SET
and is equal to:
64mV
18
67.7mV
=
R
SET
17
R
SET
V
+
I
SET
Q1
Q2
+
A1
R
–
+
64mV
R
SET
–
V
–
134 F01
Figure 1.
The 67.7mV equivalent reference voltage is directly pro-
portional to absolute temperature in degrees Kelvin (see
curve, “Operating Current vs Temperature”). This means
that the reference voltage can be plotted as a straight line
going from 0mV at absolute zero temperature to 67.7mV
at 298°K (25°C). The slope of this line is 67.7mV/298 =
227µV/°C.
The accuracy of the device is specified as a percent error
at room temperature, or in the case of the -3 and -6
devices, as both a percent error and an equivalent tem-
perature error. The LM134 operating current changes at a
percent rate equal to (100)(227µV/°C)/(67.7mV) = 0.336%/
°C
at 25°C, so each 1% operating current error is equiva-
lent to
≈3°C
temperature error when the device is used as
a temperature sensor. The slope accuracy (temperature
coefficient) of the LM134 is expressed as a ratio com-
pared to unity. The LM134-3, for instance, is specified at
0.98 to 1.02, indicating that the maximum slope error of
U
the device is
±2%
when the room temperature current is
set to the exact desired value.
Supply Voltage Slew Rate
At slew rates above a given threshold (see curve), the
LM134 may exhibit nonlinear current shifts. The slewing
rate at which this occurs is directly proportional to I
SET
. At
I
SET
= 10µA, maximum dv/dt is 0.01V/µs; at I
SET
= 1mA,
the limits is 1V/µs. Slew rates above the limit do not harm
the LM134, or cause large currents to flow.
Thermal Effects
Internal heating can have a significant effect on current
regulation for I
SET
greater than 100µA. For example, each
1V increase across the LM134 at I
SET
= 1mA will increase
junction temperature by
≈0.4°C
in still air. Output current
(I
SET
) has a temperature coefficient of
≈0.33%/°C,
so the
change in current due to temperature rise will be (0.4)(0.33)
= 0.132%. This is a 10:1 degradation in regulation com-
pared to true electrical effects. Thermal effects, therefore,
must be taken into account when DC regulation is critical
and I
SET
exceeds 100µA. Heat sinking of the TO-46 pack-
age or the TO-92 leads can reduce this effect by more than
3:1.
Shunt Capacitance
In certain applications, the 15pF shunt capacitance of the
LM134 may have to be reduced, either because of loading
problems or because it limits the AC output impedance of
the current source. This can be easily accomplished by
buffering the LM134 with a FET, as shown in the applica-
tions. This can reduce capacitance to less than 3pF and
improve regulation by at least an order of magnitude. DC
characteristics (with the exception of minimum input
voltage) are not affected.
Noise
Current noise generated by the LM134 is approximately 4
times the shot noise of a transistor. If the LM134 is used
as an active load for a transistor amplifier, input referred
noise will be increased by about 12dB. In many cases, this
is acceptable and a single stage amplifier can be built with
a voltage gain exceeding 2000.
W
U
U
5
LINER [ LINEAR TECHNOLOGY ]
