The specified timebase aging rate may or may not be achieved before the specified stabilization period. Because the specified period is in days, this performance test is optional.
|
|
The specified timebase aging rate may or may not be achieved before the specified stabilization period. Because the specified period is in days, this performance test is optional. |
This manual test verifies the aging rate of the signal generator’s internal timebase. The time required for a 360 degree phase change (relative to the cesium beam reference) is measured both before and after a specified waiting period. The aging rate is inversely proportional to the absolute value of the difference in the measured times. The results of this test can be recorded using the Agilent ESG Signal Generator Timebase Aging Rate Performance Test Record.
The overall aging rate of the internal timebase is a function of:
TBC ± TE ± LE where:
TBC = timebase calibration
TE = temperature effects
LE = line effects
After the settling period is adjusted, the timebase should stay within the aging rate for the following situations:
The timebase oven does not cool down.
The instrument keeps the same orientation, with respect to the earth’s magnetic field.
The instrument stays at the same altitude.
The instrument does not receive any mechanical shock.
If the timebase oven cools (the instrument power switch is set to off), you may have to readjust the timebase frequency after a new warm-up cycle. Typically, however, the timebase frequency returns within ±1 Hz of the original frequency.
Frequency changes due to either a change in orientation, with respect
to the earth's magnetic field, or to a change in altitude usually go away
when the instrument is returned to its original position. A frequency
change due to mechanical shock usually appears as a fixed frequency error.
|
Instrument |
Critical Specifications for This Test |
Recommended Model |
|
Dual-Channel Oscilloscope |
Inputs: BNC (f) |
Agilent 54600B oscilloscope |
|
Frequency Standard |
Output: BNC (f) |
Use a 10 MHz house standard. |
|
Cable |
BNC (m) to (m) |
Agilent 8120-1840 BNC (m) to (m) cable |
Connections and Setup Procedures
|
|
Allow the unit (signal generator) to warm up for 24 hours. All other equipment requires a 1 hour warm up period to ensure accurate performance. In the following test setup, direct cable and adapter connections are designated as solid lines. Changeable cable and adapter connections are designated as dashed lines. |
Connect all test equipment as shown.
Preset all other equipment and let them warm up for at least one hour to ensure accurate performance.
If the oscilloscope does not have a 50 ohm input impedance, connect channel 1 through a 50 ohm feedthrough.
On the oscilloscope, adjust the external triggering for a display of the 10 MHz REF OUTPUT signal from the signal generator.
Set Channel 1 to the following:
Display: On
Volts/Division: 500 mV
Input Coupling: DC
Input Impedance: 50 ohms (or use a 50 ohm feedthrough)
Set Channel 2 to the following:
Display: Off
Input Coupling: DC
Input Impedance: 50 ohms (or use a 50 ohm feedthrough)
Set the Timebase to the following:
Time/Division: 5 ns
Set the Trigger to the following:
Trigger Source: CH 2
Trigger Mode: Normal
Trigger Level: 0 V
Determine and record the number of cycles (N) and time (T1) to complete N cycles.
Monitor the oscilloscope display for a minimum of two minutes and count the number of full cycles (N) that pass the center reference point (the center is being used as a convenient reference point, but this can be anywhere on the oscilloscope display).
Record
the time (T_) that is required to complete all cycles.
The number of cycles (N) observed during the monitoring time is
caused by the small amount of frequency difference between the 10 MHz
reference (of the signal generator being tested) and the frequency standard
(that it is being compared against); as little as 0.001 Hz difference
causes multiple cycles to occur.
Wait 6 to 24 hours. Record the time between measurements as T2.
Repeat steps 4 through 7 and record the time as T3 to complete the number of cycles (N) recorded in step 7.
Calculate the aging rate as follows:
Aging Rate = (N cycle/10 MHz) (1/T1 - 1/T3) (24
hours/T2)
Example: with T1 = 351 seconds, T2 = 3 hours, T3 = 349 seconds,
N = 1
Aging Rate = (1 cycle/10 MHz) (1/351s - 1/349s) (24h/3h) = (1.306
x 10-11 per day)
Compare the aging rate to the specification and if the signal generator:
meets the aging rate, write the results on the performance test record located at the end of this chapter.
does not meet
the aging rate and the signal generator has not met the specific settling
time, use the last value recorded for T3 as T1 and repeat steps 9 and 10.
In Case of Difficulty
Verify that the equipment is set up as shown in the equipment setup for this performance test.
If after the specified settling period, the aging rate
is not met, replace the reference assembly.
To find these procedures, turn to ”Assembly Replacement"
in the troubleshooting section of the service guide. If you do not have
a hard copy of the service guide (Option OBW), refer to the .pdf file
available on the CD-ROM that came with your shipment.
If the absolute frequency of the standard and the timebase oscillator are extremely close, you can reduce the measurement time (T1and T3) by measuring the time required for a phase change of less than 360 degrees. In step 9, change 1 cycle to 0.5 cycle for 180 degrees, or 0.25 cycle for 90 degrees.
If necessary, obtain service from Agilent Technologies. Refer to Contacting Agilent Technologies.