-2
Page D3, para. 3, line 2 - change 2000 ohms to 1000 ohms; line 4 - change "3k to 4k" to "1.5k to
2k".
Modifications to the A1 and A3 amplifier stages to simplify negative and bipolar output operation: Changes
should be made to section 2.6 (pages 10 and 10A) of the application notes (Rev. 5), as follows:
1.
The current sink provision for A1 to swing -2.3 V now only needs to be 8 to 10 mA.
2.
The 5-ohm resistor shown in Fig 2.6 is no longer present, and the consequent re-balancing
procedure which is described is no longer necessary.
3.
The A3 linear stage now has adequate negative swing, and thus no external current sink is
required, nor is the re- balancing procedure for A3 described on page 10A required.
Also, in Appendix B, Fig. B1 (lower right-hand corner) should be changed (for the modified design) to show
the internal load to ground as 370 ohms (rather than 250 ohms) and Fig. B3 should be changed (also in the
lower right-hand area) to show the output-stage current as 1.5 mA (instead of 1 mA).
The 200-ohm internal resistors leading to pins 29 and 26 (the A2 and A3 external outputs) are no longer
made of nichrome: The purpose of this is to make these outputs less sensitive to electrostatic damage.
However, good anti-ESD handling and fab practices should be observed.
A design modification has been made to the output-stage amplifier to decrease its gain temperature-
dependence: It is now likely that the use of the output slope temp. comp. sensistor S1 (refer, e.g., to Fig.
2(c), the first half of page 27, and the first paragraph of page 30B) will no longer be necessary for most
applications.
B: Slope-Matched Pairs for the L-17C Pinout-40
Some L-17C pinout-40 users with two-arm DLVA applications have noted that when using the S+ and S-
summing junctions (pins 8 and 15) to sum the outputs of the two DLVA arms, there is sometimes a
noticeable change in the transfer function slope which occurs at the transition point between the low- and
high-power arms. Consequently, ANADYNE now offers pinout-40 L-17Cs in slope-matched pairs.
There is no additional cost to the purchaser for this screen, which is performed by ANADYNE as part of its
pre-shipment test procedures. The slope-matching is to within 0.5%. This ensures that the transfer function
will not deviate from a fitted line over the full two-arm dynamic range (90+ dB of RF power with tunnel diode
detectors) by more than about 0.1 dB (solely as a result of slope mismatch). Of course, once the transfer
slopes of the two arms are matched, the overall transfer slope can easily be adjusted by trimming the
output- stage feedback resistor of whichever L-17C output is being used (this would normally be the low-
power arm, with the output-stage gain on the high-power arm being greatly reduced).
For the benefit of L-17C users who currently have pinout-40 L-17Cs in stock, and who wish to perform a
slope-match screen using their own L-17C test fixture, the procedure is as follows:
1.
Using a voltmeter with a precision over the required range of 0.5% or better, measure the
voltage between pins 8 and 15 at two widely different input power levels (use CW), for
example 0 dBm and -30 dBm. Subtract the low-power voltage from the high-power voltage.
This gives the slope over the measured power range for the S+ S- outputs of a single IC.
2.
Since the summed-signal contribution of each member of an L-17C pair also depends on
the exact values of the resistors which internally connect pins 8 and 15 to 6 volts, the value
of the resistance between pins 8 and 15 is also accurately measured (with the IC
unpowered). The voltage difference measured in step 1, above, is divided by this
 |
 |
 |
 |
 |