AR 10008569, Operating And Service Manual, Page: 35 / 38

Models PH – Power Heads
3. At a frequency of 7 GHz a calibration factor will be applied to our measurement. The worst case
uncertainty of this calibration factor at 7 GHz is 3.1%; (See Table 3-1).
4. The next item to calculate is mismatch uncertainty. This uncertainty is calculated from the SWR’s of
both head and source. Once the SWR’s are known, we must calculate the reflection coefficient
(amplitude only) for both the head and source. To find reflection coefficient (p), use the formula:
p=SWR–1/SWR+1 . After the reflection coefficient is found, mismatch uncertainty is found by using the
formula.
[(1
±
p head x p source)
2
– 1] x 100.
Example: Source SWR = 1.50, Head SWR = 1.40
[(1
±
0.2 x 0.167)
2
– 1] x 100 = +6.79%, -6.57%
Average Uncertainty will be 6.68% (worst case)
5. The noise specification on the Model PH2005 is 30 pW, RMS, therefore, for a signal level of –55 dBm
(3.16 nW), the additional uncertainty due to noise will be 0.95%. This is calculated by using the formula,
(signal level/noise specification) x 100.
6. In this step, we will discuss the linearity of the PH2005. As shown in Table 2-1 (Notes), this uncertainty
will be 1% for levels below 10 dBm.
7. In our example, we mentioned that the temperature of our room has drifted 5
°
C from the time that we
established our 50 MHz, 0 dBm reference. Since all power heads are temperature sensitive, we must add
the heads drift, due to temperature, into our total uncertainty. If the drift in the PH2005 is 0.01 dB per
°
C,
the additional uncertainty in our measurement will be
±
0.05 dB or 1.16%. Please note that performing the
auto calibration at the new temperature will cancel out this additional uncertainty.
8. Now that all of the uncertainties have been established, we can combine them to determine the total
measurement uncertainty. To find the RSS uncertainty proceed as follows:
Source of Error Value (%) Value
2
(%)
Instrumentation 0.46 0.2116
Power Reference 1.20 1.4400
Calibration Factor 3.10 9.6100
Mismatch 6.68 44.6200
Noise 0.95 0.9025
Linearity (Shaping) 1.00 1.0000
Temperature 1.16 1.3456
Sum of Squares 59.1297
Total RSS Uncertainty 7.69%
To find the maximum worst case uncertainty, we would add each of the individual uncertainties (%) as
follows:
0.46 + 1.20 + 3.10 + 6.68 + 0.95+ 1.00 + 1.16 = 14.55%
p source = 1.50 - 1
1.50 +1
= 0.2
p head =
1.40 - 1
1.40 + 1
= 0.167
Rev C 21