WEBVTT NOTE Measurement Technique and S Parameters 1 00:00:00.000 --> 00:00:14.864 S parameters. Scattering parameters describe a network by what it does to waves entering and leaving its ports, which is the natural description at RF, where you cannot easily measure voltage and current directly. 2 00:00:14.864 --> 00:00:20.064 Question. What do the subscripts of S parameters represent? 3 00:00:20.964 --> 00:00:26.804 Answer. The port or ports at which measurements are made. 4 00:00:27.404 --> 00:00:45.378 The convention is S(out, in), the first subscript is where the signal comes out, the second is where it went in. Parameter, S11; Meaning, input port return loss or reflection coefficient. 5 00:00:45.378 --> 00:01:06.139 Parameter, S21; Meaning, forward gain (or insertion loss). Parameter, S12; Meaning, reverse gain, isolation. Parameter, S22; Meaning, output port return loss. S21 is in at port 1, out at port 2, forward transmission. 6 00:01:06.139 --> 00:01:16.039 S11 is in at port 1, back out at port 1, reflection. Once you have the ordering, the whole set follows. 7 00:01:16.039 --> 00:01:29.784 For an amateur, S11 is what an antenna analyser reports as SWR, and S21 is what you measure to characterise a filter. Vector network analysers. 8 00:01:29.784 --> 00:01:37.204 Question. What three test loads are used to calibrate an RF vector network analyzer? 9 00:01:38.104 --> 00:01:44.198 Answer. Short circuit, open circuit, and 50 ohms. 10 00:01:44.798 --> 00:01:54.962 Three known loads, the SOL calibration, let the instrument solve for its own errors: cable loss, connector mismatch, directivity. 11 00:01:54.962 --> 00:02:06.229 Calibration is done at the end of the cables you will actually use, which is why a NanoVNA must be recalibrated whenever you change test leads. 12 00:02:06.229 --> 00:02:13.648 Question. Which of the following can be measured by a two-port vector network analyzer? 13 00:02:14.548 --> 00:02:18.428 Answer. Filter frequency response. 14 00:02:19.028 --> 00:02:36.008 Question. Which of the following can be measured with a vector network analyzer? A. Input impedance. B. Output impedance. C. Reflection coefficient. D. All these choices are correct. 15 00:02:36.908 --> 00:02:40.428 Answer. All these choices are correct. 16 00:02:41.028 --> 00:02:58.668 A VNA is the instrument that made serious RF measurement affordable for amateurs. A NanoVNA costs less than a decent multimeter and does what a laboratory instrument did twenty years ago. Accuracy limits. 17 00:02:58.668 --> 00:03:06.150 Question. Which of the following factors most affects the accuracy of a frequency counter? 18 00:03:07.050 --> 00:03:10.970 Answer. Time base accuracy. 19 00:03:11.570 --> 00:03:19.432 A counter counts cycles in a known interval, and the whole measurement is only as good as that interval. 20 00:03:19.432 --> 00:03:32.736 A counter with eight digits and a poor crystal displays eight digits of which three are meaningful, which is why the GPS-disciplined oscillator from the circuits lesson exists. 21 00:03:32.736 --> 00:03:40.163 Question. What is the significance of voltmeter sensitivity expressed in ohms per volt? 22 00:03:41.063 --> 00:03:51.289 Answer. The full scale reading of the voltmeter multiplied by its ohms per volt rating is the input impedance of the voltmeter. 23 00:03:51.889 --> 00:04:10.055 An old analogue-meter specification with a live consequence. A 20,000 ohms-per-volt meter on a 10 V range presents 200 kΩ, which visibly loads a high-impedance circuit and gives a low reading. 24 00:04:10.055 --> 00:04:19.990 The General track's "voltmeters have high input impedance" is this, quantified. Computing absorbed power. 25 00:04:19.990 --> 00:04:36.590 Question. How much power is being absorbed by the load when a directional power meter connected between a transmitter and a terminating load reads 100 watts forward power and 25 watts reflected power? 26 00:04:37.490 --> 00:04:41.010 Answer. 75 watts. 27 00:04:41.610 --> 00:04:59.550 P absorbed equals P forward minus P reflected 100 − 25 = 75 watts. Simple subtraction, and worth noting what it implies: 25 W is heading back to the transmitter. 28 00:04:59.550 --> 00:05:19.272 (For interest, 25/100 reflected corresponds to an SWR of 3:1, the reflected power fraction is the square of the voltage reflection coefficient, so Γ = 0.5.) Measuring Q and IMD. 29 00:05:19.272 --> 00:05:26.632 Question. Which of the following can be used to determine the Q of a series-tuned circuit? 30 00:05:27.532 --> 00:05:32.452 Answer. The bandwidth of the circuit's frequency response. 31 00:05:33.052 --> 00:05:56.212 Rearranged from BW = f 0/Q: measure the −3 dB bandwidth and the resonant frequency, and Q = f 0/BW. That is a measurement you can make with a signal generator and a detector, or, in one sweep, with a VNA. 32 00:05:56.212 --> 00:06:04.372 Question. Which of the following methods measures intermodulation distortion in an SSB transmitter? 33 00:06:05.272 --> 00:06:16.912 Answer. Modulate the transmitter using two AF signals having non-harmonically related frequencies and observe the RF output with a spectrum analyzer. 34 00:06:17.512 --> 00:06:34.249 The two-tone test from the General track, stated in full. Non-harmonically related so the distortion products do not hide under the tones, and a spectrum analyser to see them. Check yourself. 35 00:06:34.249 --> 00:06:52.827 Your VNA shows S21 falling steeply above 30 MHz on a filter. What have you measured? A directional meter reads 200 W forward, 8 W reflected. How much reaches the load? Your frequency counter reads 146.520037 MHz. 36 00:06:52.827 --> 00:06:55.631 How much of that do you believe? 37 00:07:01.631 --> 00:07:22.553 Forward transmission, in at port 1, out at port 2. You have measured the filter's response, and it is rolling off above 30 MHz. 200 − 8 = 192 watts. Only as many digits as the time base supports. 38 00:07:22.553 --> 00:07:37.788 A counter with an ordinary crystal is good to a few parts per million, so about 146.5200 MHz is honest and the last two digits are decoration.