Physics 160Lecture 5R. JohnsonDiode RectifiersHalf WaveFull WaveApril 15, 2014Output SmoothingNote that there is no ground connection on The load (shown here as a 1 kohmresistance) draws current from thecapacitor when a diode is not supplyinggthis side of the rectifier!the current.=R1C1 decay time constant• If R1 is small (high load; i.e. high current), then C1 must be large inorder for the time constant to be long enough to minimize the ripple.• Later in the quarter we will see how to improve on this using activeltlttli i tildhldthttltApril 15, 2014voltage regulatorstoeliminaterippleandholdtheoutputvoltageconstant even with a changing load.Spice Simulation of the RectifierOutput peak is tdi d d500 ohm load resistancetwo diode drops less than the input peak.This region of zero output is where nooutput is where no diode is conducting, because of the 2-diode drops(aboutdiode drops (about 1.4 V) needed to flow current through the two diodes in April 15, 2014series.Rectifier With FilteringApril 15, 2014With Too Much LoadI decrease the load resistance by a factor of 10 to draw 10 times more current. Then the ripple becomes unacceptably large.April 15, 2014Voltage ClampCommon Example: protecting circuitry in an ICfrom external voltage spikes (e.g. electrostaticdischarge)ontheinputpadsorpinsdischarge)ontheinputpadsorpins.April 15, 2014Rectified “Differentiator”One of the lab projects for this week.• We see only the positive approximate “derivative” at the outputoutput.• This won’t work if the 2.2k resistor is removed. Why?• What are the charge and discharge paths for the capacitor charge?April 15, 2014charge?Zener Voltage ReferenceOutput currentZenercurrent•Why is the resistor R necessary?VinVoutcurrentyThe Zener is always operated in the reverse bkd ibreakdown region.As long as the Zenerreverse current is not close to zero thenVoutA large change close to zero, then Voutis close to the Zenerbreakdown voltage over a wide range of outputin Zener current corresponds to a very small voltage change.April 15, 2014a wide range of output current.Semiconductors• Pure silicon conductivity (“intrinsic”):– Cu conductivity: ~6×105S/cm (S=siemens=1/ohm)– Si conductivity: ~3×106S/cm (but impossible to get this pure)– C (diamond) conductivity: ~1016S/cm (pure, undoped)• Conductor: No energy gap between the “valence” energy levels (“bands”) and the “conduction” energy levels (“bands”).– Electrons near the Fermi level are easily moved into slightly higher unoccupied energy levels where they are free to move.• Insulator: Large energy gap between filled “valence bands” and unoccupied “conduction bands.”– Normal temperatures are extremely unlikely to excite an electron into the conduction band where it would be free to move.• Semiconductor: band structure essentially the same as that of an insulator, but with a small “band gap” of energy not too much greater than ~kT at room temperature.April 15, 2014– A small but significant number of electrons are excited into the conduction band at room temperature.Energy BandsApril 15, 2014April 15, 2014Non-Zero TemperatureIntrinsic semiconductor (pure silicon)Band gap ~1.1 eVkT~ 0.03 eV at room temperaturepFermi Energy (chemical potential)Fermi-Dirac distributionpotential)April 15, 2014Extrinsic (doped) Semiconductors• Add a tiny bit of phosphorus to the silicon (n-type doped silicon):• New states are produced just below the conduction band• Electrons on those states easily get excited into the conduction band•Add a tiny bit of boron to the silicon (ptype doped silicon):April 15, 2014•Add a tiny bit of boron to the silicon (p-type doped silicon):• Holes at the top of the valence band can conductPN JunctionWhat happens if we bring an N-type semiconductor into very close contact with a P-type semiconductor?–Some electrons in the N-type material move into the P-typeSome electrons in the Ntype material move into the Ptype material to fill in some of the holes.– This movement of charge builds up an electric field.–Eventually the electric field prevents any more net movement of ypycharge.• At that point the system is in equilibrium.• Then the chemical potentials match between the P and N type materials.• There is a thin charge-free “depletion region” at the junction where the electric field is established.April 15, 2014PN Junction in EquilibriumApril 15, 2014PN Junction in Reverse BiasSmall net reverse electron flow, li it d b b f i it ilimited by number of minority carriers.April 15, 2014PN Junction in Forward BiasILarge net forward electron flow, growing exponentially with voltage.Large net forward hole flow, growing exponentially with voltage.April 15, 2014Diode Review• A diode conducts in the forward direction because1. the external potential V lowers the potential barrier between the N and P doped silicon by an amount eV,2. and as a result, there are exponentially more electrons on the N side with thermal energy fluctuating high enough to get over the barrier.It cond cts onl a tin c rrent in the re erse direction beca se•It conducts only a tiny current in the reverse direction because, independent of V, there are few electrons in the p-type material to conduct current (and few holes in the n-type material).Therefore the dependence of current on voltage will be•Therefore, the dependence of current on voltage will be1kTeVIIFollows from the Boltzmann factor •where I0is the tiny reverse current (which itself depends 10kTeVeIIrelated to the population of the conduction band.April 15, 2014ee0s t e t y e e se cu e t ( c tse depe dsexponentially on temperature but not on V).Transistor Prelude• But, if we could inject electrons into the base somehow, they would easily fall “downhill” into the collector, making a flow of current far greater than the normal tiny reverse current.April 15, 2014Transistor Action (NPN)• Attach a second PN junction and forward bias it.• The voltage VBEcontrols the gBEbarrier height between base and emitter.•Raising VBElowers the barrier, gBE,allowing electrons to flood into the base from the emitter.• The base is very thin, so most forwardreversey,of those electrons quickly diffuse to the collector junction, where they fall The base should be very thin and li htl d d“downhill” into the collector.• Only a few percent of the electrons flow to the base lightly doped:• Most injected