Agilent Technologies 6622A Service Manual page 34

* An output channel cannot be programmed to 0 amps. If the output channel receives a command to go to 0 amps (or any positive
current below the minimum
3-15 CV Noise (PARD). Periodic and random deviations
(PARD) in the output (ripple and noise) combine to produce
a residual ac voltage superimposed on the dc output
voltage. CV PARD is specified as the rms or peak-to-peak
output voltage in a frequency range from 20 Hz to 20 MHz.
This test measures the rms and peak-to-peak noise on the
output.
a. Turn off the supply and connect the output to be tested
as shown in Figure 3-3 to an oscilloscope (ac coupled)
between the +S and -S terminals, the Load switch
closed, and the Short switch opened. Be sure to keep
the leads from the 50 ohm impedance matching resistor
and the 50 ohm coaxial cable shield that run to the + S
and - S terminals as short as possible to avoid external
noise pickup.
b. Turn on the supply and select the output to be tested
(OUTPUT SELECT key on the front panel).
c. Program the current to the Low Range Maximum
Programmable Current Value and the output voltage to
the Low Range Full Scale Voltage value (See Table 3-2)
by sending the following strings:
ISET < ch >, < 5.15, 2.06, 10.3, or 4.12 >
VSET <ch>, <7 or 20>
d. Adjust the load for the Low Range Full Scale Current
value (see Table 3-2) as indicated on the front panel
display. The CV annunciator on the front panel must be
on. If it is not, adjust the load down slightly.
e. Note that the waveform on the oscilloscope should not
exceed 3 mV peak to peak.
f. Disconnect the oscilloscope and connect an rms
voltmeter in its place. The rms voltage reading should
not exceed 500 µV.
g. Repeat steps a through f for each output in your
supply.
3-16 CV Down Programming Speed. This test measures
the time required for the output voltage to fall to 37% of the
High Range Full Scale Voltage (time constant). Also
measured is the time an output takes to change from full
scale to zero volts and settle within the specified voltage
settling band (response time).
a. Turn off the supply and connect the output to be tested
as shown in Figure 3-4.
b. Turn on the supply and select the output to be tested
(OUTPUT SELECT key on the front panel).
programmable current), it will set itself to
c. First, program the selected output to zero volts by
sending the string:
VSET < ch >, 0
d. Using Channel A on the oscilloscope, set the
volts/division switch to 5 V/div (40WLV/80WLV
outputs) or to 10 V/div (40WHV/80WHV outputs) dc
coupled and position the trace on the bottom horizontal
line. Using Channel B on the oscilloscope, set the
volts/division switch to 50 mV/div dc coupled and
position the trace on the bottom horizontal line.
e. Program the output voltage in a loop which alternately
programs the output voltage between 0 and the High
Range Full Scale Voltage value by running the program
listed below. Refer to Table 3-3 for the High Range Full
Scale Voltage value for the particular output being
tested.
10 OUT 705; "VSET < ch >,0"
20 WAIT 0.05
30 OUT 705; "VSET < ch >, < High
Range F. S. Voltage >"
40 WAIT 0.05
50 GOTO 10
60 END
The tested output's CV annunciator should remain on at all
times while the test is in progress.
f. Observe Channel A on the oscilloscope and adjust for a
stationary waveform by using Channel A as the trigger
source set to trigger on a negative edge. Be sure to
trigger as close as possible to the time when the output
voltage just begins to fall.
g. On Channel A, observe the output voltage transition
from the High Range Full Scale Voltage to the scope's
bottom horizontal line. Look for a smooth exponential
waveform with no "kinks" or aberrations. Perform a
time contstant check by insuring that the output voltage
falls to about 37% of the High Range Full Scale in less
than 250 µsec. (40WLV/80WLV outputs) or 750 µsec.
(40WHV/80WHV outputs). Refer to the Channel A
waveform shown in Figure 3-5.
3-6
the mum programmable
NOTE
current.