Electricity: measuring and testing – Measuring – testing – or sensing electricity – per se – Using radiant energy
Reexamination Certificate
2001-10-18
2004-05-18
Pert, Evan (Department: 2829)
Electricity: measuring and testing
Measuring, testing, or sensing electricity, per se
Using radiant energy
Reexamination Certificate
active
06737853
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to acquisition of voltage measurements with a photoconductive sampling probe.
2. The Prior Art
A prior-art sampling probe has a contact tip and a high-impedance photoconductive gate fabricated as an electrode structure on a substrate. To sample voltages on a conductor of a device under test (DUT), the contact tip is applied to the conductor and an optical probe laser beam is pulsed to close the gate. See J. KIM et al., Photoconductive sampling probe with 2.3-ps temporal resolution and 4 &mgr;V sensitivity, APPL. PHYS. LETT. 62(18), May 3, 1993, pp. 2268-2270. A system employing such a probe is described in U.S. Pat. No. 5,317,256 dated May 31, 1996 to Williamson. See also U.S. Pat. No. 5,331,275 dated Jul. 19, 1994 to Ozaki et al., and U.S. Pat. No. 5,442,300 dated Aug. 15, 1995 to Nees et al.
In an ideal photoconductive (PC) switch, the dark-resistance (when the laser pulse is off) is infinite, so that the rest of the sampling circuit is electrically connected to the DUT only when the laser pulse is on. For such an ideal PC switch, one operating method is to keep the output side of the PC switch at a fixed voltage (say 0 V) and to measure the net charge passed by the switch over an entire trigger period, where the trigger period is the product of the length of a repetitive stimulus-signal loop applied to the DUT and the clock period of the loop. A simple calibration allows this net charge to be interpreted as voltage.
FIG. 1
shows an equivalent-circuit view of an ideal prior-art PC sampling system. The DUT is represented as a voltage source
100
connected to the input terminal of a PC switch
105
which has a resistance
110
when closed of some value R
on
, such as 50 k&OHgr;. The current Ipc through PC switch
105
and resistance
110
R
on
is applied to the input of a current-to-voltage converter made up of a differential amplifier
115
and a feedback resistance
120
having a value R. The converter output voltage is thus V
out
=R·I
pc
.
FIG. 2
illustrates. A trigger pulse occurs once for each repetition of a stimulus-signal pattern applied to the DUT, as shown in line
200
. A laser sampling pulse is produced at some time after the trigger pulse as shown in line
205
. The ideal PC switch responds to the laser sampling pulse by changing resistance from R
off
=
∞ to R
on
=50 k&OHgr;, as shown in line
210
. Voltage on a conductor of the DUT to be sampled is shown at line
215
. The result of closing the PC switch in response to the laser sampling pulse is a signal V
out
, as shown in line
220
. The integral under V
out
for each optical sampling pulse is proportional to the sampled DUT voltage.
In practice, it has been observed that although the equilibrium dark-resistance of the PC switch R
off
is usually greater than several hundred megohms, the “dark-resistance” R
off
is much lower immediately following a laser pulse. In one experiment, the dark-resistance R
off
was found to be approximately 50 M&OHgr; for a few microseconds following a laser pulse. This “persistent photoconductivity” (PPC) effect will cause the net charge through the PC switch to be contaminated by the DUT voltage acting through the reduced dark-resistance R
off
for a few microseconds following the laser pulse, whereas accurate measurement demands that the net charge passed by the PC switch be sensitive only to the DUT voltage during the laser pulse interval.
An equivalent circuit is shown in FIG.
3
. At a laser pulse repetition rate of 500 KHz, rather than having an infinite resistance when open, the PC switch is found to have a dark-resistance
300
of R
off
=50 M&OHgr;.
FIG. 4
illustrates. A trigger pulse occurs once for each repetition of a stimulus-signal pattern applied to the DUT, as shown in line
400
. A laser sampling pulse is produced at some time after the trigger pulse as shown in line
405
. The non-ideal PC switch of
FIG. 3
responds to the laser sampling pulse by changing resistance from R
off
=50 M&OHgr; to R
on
=50 k&OHgr; as shown in line
410
.
A first example of voltage V
DUT1
on a conductor of the DUT to be sampled is shown at line
415
—in this example, the voltage is at a high level during much of the interval between trigger pulses, except for a negative-going pulse
435
just prior to the laser sampling pulse
440
. Closing the PC switch in response to the laser sampling pulse results in a signal V
out1
, as shown in line
420
. A second example of voltage V
DUT2
on a conductor of the DUT to be sampled is shown at line
425
—in this example, the voltage is at a low level during the interval between trigger pulses. Closing the PC switch in response to the laser sampling pulse results in a signal V
out2
, as shown in line
430
. The differences between lines
420
and
430
illustrate a problem with PC switch leakage. The integrated areas under V
out1
and V
out2
are not identical, even though the voltages V
DUT1
and V
DUT2
are the same at the sampling point.
PC switches also have other non-ideal characteristics which can lead to measurement errors. These include non-linear on-resistance (the conduction current saturates at high bias voltages) and temperature sensitivity. When the effect of the dark-resistance can be neglected, it is simple to avoid such errors: a hold-capacitor is charged up through the PC switch, so that when the voltage on the capacitor has reached equilibrium, that voltage is exactly equal to the DUT voltage at the time of the laser pulse, irrespective of non-linear on-resistances or temperature variations.
FIG. 5
shows an equivalent-circuit of a hold-capacitor sampling system for absolute voltage measurement with an ideal PC switch. The DUT is represented as a voltage source
500
connected to the input terminal of a PC switch
505
which has a resistance
510
when closed of R
on
. The current Ipc through PC switch
505
and resistance
510
R
on
is applied to a hold-capacitor C
hold
which is connected across the input terminals of a high-impedance amplifier
520
. Ideally, if leakage currents can be neglected, the voltage on C
hold
will charge exactly to the DUT voltage at the sampling point. The action of the hold-capacitor can be viewed as negative feedback. When the dark-resistance R
off
cannot be neglected, such a simple system will give erroneous voltage measurements due to current flow through the dark-resistance.
In addition to the above-noted limitations, the operating voltage range of high impedance input stages is usually limited. There is thus a need for improved methods and circuits of photoconductive voltage sampling.
SUMMARY OF THE INVENTION
In accordance with an embodiment of the invention, methods of probing voltage comprise: establishing electrical connectivity between a DUT conductor to be probed and a photoconductive switch; during a sampling interval n, applying a laser pulse to the photoconductive switch while applying a voltage to the photoconductive switch terminal that is not connected to the DUT, corresponding to a voltage sample taken during a prior sampling interval n−1, such that current flow through the photoconductive switch is dependent on any difference between voltage of the DUT conductor and the applied voltage; converting the current flow to a voltage signal; passing the voltage signal during a gating interval T
elec;
and sampling the passed voltage signal to produce a voltage sample for the sampling interval n.
A repetitive test pattern is applied to the conductor, and the sampling interval is synchronized with the repetitive test pattern. Converting the current flow to a voltage signal can comprise applying the current flow to a current-to-voltage converter having a rise time which is less than the gating interval T
elec
. The voltage signal can be passed only during the gating interval so that the voltage sample is insensitive to any leakage through the photoconductive switch outside of the gating interval. Passing the voltage signal during a gating interva
Ho Francis
Wilsher Kenneth R.
Chan Emily Y
NPTest LLC
Pert Evan
Riter Bruce D.
LandOfFree
Photoconductive-sampling voltage measurement does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Photoconductive-sampling voltage measurement, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Photoconductive-sampling voltage measurement will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3234244