Simple circuit lets you characterize JFETs
John Fattaruso, Dallas, TX - April 6, 2012
Figure 1 is a plot of this characteristic
for N-channel JFETs showing the variation
possible in a collection of devices.
For example, the 2N4416A’s data sheet
lists a pinch-off voltage of −2.5 to −6V,
and the zero-bias drain current can range
from 5 to 15 mA. You can observe the
correlation between these two parameters
across a sample of devices. The
outer curves in the plot represent these
extreme cases, and the center curve represents
perhaps a typical case of a pinchoff
voltage of −4V and a zero-bias drain
current of 8 mA.Although you can design around a certain amount of device variation for a mass-produced circuit, you sometimes need a tool to quickly characterize an assortment of discrete devices. This tool allows you to select a device that will optimize one circuit or perhaps to find a pair of devices with parameters that match reasonably well.
Figure 2 shows a simple test circuit for this purpose. Although the figure shows the JFET as an N-channel device, the JFET DUT (device under test) may be of either polarity, as selected by switch S1. An external voltmeter connects to the terminals on the right. Switch S2 selects two distinct measurement modes—one for the pinch-off voltage and another for the zero-bias drain current. In the pinch-off-voltage mode, the external voltmeter directly reads the pinch-off voltage; in the zero-bias-drain-current mode, the measured voltage is the zero-bias drain current across an apparent resistance of 100Ω.

In the zero-bias-drain-current mode, however, the resistance from JFET source to ground is only 10Ω, so the drain current is a close approximation of the zero-bias drain current. The op amp’s feedback also switches to a gain-of-10 configuration, with the inclusion of R4 and R5 in the feedback-voltage divider. This gain allows the voltmeter to easily read the small voltage across R2, with the resulting reading being the zero-bias drain current times 100Ω. For example, if the voltmeter reads 1V, this voltage corresponds to a zero-bias drain current of 10 mA.
For an N-channel device, both
voltage readings are positive; for a
P-channel device, the circuit functions
in the same manner except that the
voltage readings are negative. If you
wire the test JFET to this circuit with
test leads and clips, each with some parasitic
series inductance, you may need
to add C1 to suppress any tendency for
high-frequency oscillation. R6 isolates
the op-amp feedback loop from any
parasitic capacitance in the voltmeter
and its leads, preserving the loop stability.
R7 protects against accidental shorts,
and you can replace R4 and R5 with one
1.1-kΩ resistor. You are more likely to
have on hand resistors with the values
in the figure, however.By clipping in samples from a collection of JFETs and throwing a switch, you can very quickly find the two parameters that determine where each JFET’s characteristic falls in the range that Figure 1 illustrates and select devices to optimize circuit performance.
Signal distortion from high-K ceramic capacitors
Simple reverse-polarity-protection circuit has no voltage drop
Comparator-based buffer using a VVCCS
Why bypass caps make a difference - Part 1: How a regulator and its output capacitor can interact
Two-IC circuit combines digital and analog signals to make multiplier circuit
Jim Williams: The light side and classic electronics art sculptures
Understanding the basics of setup and hold time
Simple reverse-polarity-protection circuit has no voltage drop
Signal distortion from high-K ceramic capacitors
Wideband fully differential amplifier noise improved using active match
Filter quashes 60-Hz interference
NE555 timer sparks low-cost voltage-to-frequency converter
ACE Awards Jim Williams Contributor of the Year: Steve Hageman
