WEBVTT NOTE RF Effects in Real Components 1 00:00:00.000 --> 00:00:01.447 Skin effect. 2 00:00:01.447 --> 00:00:06.927 Question. What is the result of conductor skin effect? 3 00:00:07.827 --> 00:00:16.644 Answer. Resistance increases as frequency increases because RF current flows closer to the surface. 4 00:00:17.244 --> 00:00:33.787 At DC, current uses the whole cross-section. At RF, induced eddy currents push it outward until it flows only in a thin surface layer, so the effective cross-section shrinks and resistance rises with frequency. 5 00:00:33.787 --> 00:00:51.118 Three consequences worth carrying: Flat strap beats round wire for RF bonding, because what matters is surface perimeter rather than cross-sectional area. That is the Technician station-setup answer, now with its reason. 6 00:00:51.118 --> 00:01:00.492 Silver plating helps at RF and not at DC, only the surface carries current, so only the surface's conductivity matters. 7 00:01:00.492 --> 00:01:10.339 Hollow tubing is as good as solid rod at RF and much lighter, which is why beam elements and amplifier tank coils are tubing. 8 00:01:10.339 --> 00:01:17.859 Question. What is the primary cause of loss in film capacitors at RF? 9 00:01:18.759 --> 00:01:21.599 Answer. Skin effect. 10 00:01:22.199 --> 00:01:25.119 Everything has parasitics. 11 00:01:25.119 --> 00:01:31.129 Question. What combines to create the self-resonance of a component? 12 00:01:32.029 --> 00:01:37.909 Answer. The component's nominal and parasitic reactance. 13 00:01:38.509 --> 00:01:46.729 Every real component is the thing it is supposed to be plus a small amount of the thing it is not. 14 00:01:46.729 --> 00:01:58.389 Those parasitics resonate against the nominal value, and above that self-resonant frequency the component behaves as its parasitic instead. 15 00:01:58.389 --> 00:02:05.229 Question. What parasitic characteristic creates an inductor's self-resonance? 16 00:02:06.129 --> 00:02:09.687 Answer. Inter-turn capacitance. 17 00:02:10.287 --> 00:02:23.567 Capacitance between adjacent turns. Above self-resonance the inductor is capacitive, the General components lesson stated that result; this is the mechanism. 18 00:02:23.567 --> 00:02:32.807 Question. What parasitic characteristic causes electrolytic capacitors to be unsuitable for use at RF? 19 00:02:33.707 --> 00:02:36.366 Answer. Inductance. 20 00:02:36.966 --> 00:02:49.414 An electrolytic is a long strip of foil rolled up, which is a coil. Its series inductance means it stops being a capacitor at a few hundred kilohertz. 21 00:02:49.414 --> 00:03:02.692 This is why a power supply needs both a large electrolytic for bulk smoothing and a small ceramic in parallel for RF bypass: neither works over the whole range. 22 00:03:02.692 --> 00:03:11.568 Question. Why is it important to keep lead lengths short for components used in circuits for VHF and above? 23 00:03:12.468 --> 00:03:16.828 Answer. To minimize inductive reactance. 24 00:03:17.428 --> 00:03:36.084 A wire's inductance is roughly 20 nH per inch. At 145 MHz that is about 18 ohms per inch of lead, comparable with the impedances in the circuit, so lead length is a circuit element whether you intended it or not. 25 00:03:36.084 --> 00:03:37.668 Electrical length. 26 00:03:37.668 --> 00:03:43.068 Question. Why are short connections used at microwave frequencies? 27 00:03:43.968 --> 00:03:48.535 Answer. To reduce phase shift along the connection. 28 00:03:49.135 --> 00:04:04.015 At microwave a connection is a transmission line, and its length is a meaningful fraction of a wavelength. A signal arrives phase-shifted, which changes what the circuit does. 29 00:04:04.015 --> 00:04:11.937 Question. As a conductor's diameter increases, what is the effect on its electrical length? 30 00:04:12.837 --> 00:04:15.626 Answer. It increases. 31 00:04:16.226 --> 00:04:29.076 A fatter conductor is electrically longer than a thin one of the same physical length, because increased capacitance to its surroundings slows the wave. 32 00:04:29.076 --> 00:04:44.546 This is why a fat dipole is cut shorter than a thin-wire one for the same frequency, the same end effect that gives 468 rather than 492 in the dipole formula. Real and reactive power. 33 00:04:44.546 --> 00:04:49.106 Question. What is reactive power? 34 00:04:50.006 --> 00:04:54.886 Answer. Wattless, nonproductive power. 35 00:04:55.486 --> 00:05:02.490 Question. What happens to reactive power in ideal inductors and capacitors? 36 00:05:03.390 --> 00:05:12.074 Answer. Energy is stored in magnetic or electric fields, but power is not dissipated. 37 00:05:12.674 --> 00:05:20.218 Question. What is the phase relationship between current and voltage for reactive power? 38 00:05:21.118 --> 00:05:25.216 Answer. They are 90 degrees out of phase. 39 00:05:25.816 --> 00:05:43.277 Reactive power flows into a reactance and back out again each cycle. Nothing is consumed, the current is real and does real work in the wire's resistance, but the reactance itself dissipates nothing. 40 00:05:43.277 --> 00:05:45.821 The quantitative consequence: 41 00:05:45.821 --> 00:06:00.461 Question. How much real power is consumed in a circuit consisting of a 100-ohm resistor in series with a 100-ohm inductive reactance drawing 1 ampere? 42 00:06:01.361 --> 00:06:04.402 Answer. 100 watts. 43 00:06:05.002 --> 00:06:24.567 P = I to the power of 2R = 1 to the power of 2 times 100 = 100 watts. The reactance contributes nothing to real power, even though it contributes to impedance and therefore to how much current flows for a given voltage. 44 00:06:24.567 --> 00:06:36.180 That is why power in a reactive circuit is P = E times I times cos(θ), and why the foundations power lesson flagged the qualifier. 45 00:06:36.180 --> 00:06:54.137 Here only the resistance dissipates, and I to the power of 2R with the resistive part alone is the whole answer. Check yourself. Why is silver plating useful on an RF coil and pointless on a DC busbar? 46 00:06:54.137 --> 00:07:12.272 Your bypass capacitor is a 10 µF electrolytic and the circuit still has RF on the supply rail. What is wrong? A circuit draws 2 A through 50 Ω of resistance in series with 200 Ω of reactance. Real power? 47 00:07:18.272 --> 00:07:31.790 Skin effect, at RF the current flows only in the surface layer, so surface conductivity is what matters. At DC the whole cross-section conducts. 48 00:07:31.790 --> 00:07:49.814 An electrolytic's series inductance makes it useless above a few hundred kilohertz. Add a small ceramic in parallel. P = I to the power of 2R using the resistance only: 4 times 50 = 200 watts. 49 00:07:49.814 --> 00:07:52.912 The reactance dissipates nothing.