Lab #1: Ohm’s Law (and not Ohm’s Law)

Download Report

Transcript Lab #1: Ohm’s Law (and not Ohm’s Law)

End of Semester Info
• Exam questions available on class web page: 2 of the 4 questions posted at
the “question bank” link will be chosen for the exam
• Practice for the exam next week during class (Nov 26,27). You are not
required to come, but are strongly encouraged to do so. Dan will run these
classes.
• The exam is Dec 3 / 4.
• Make up for missed labs is also Nov 26,27.
• Please check ELMS and make sure all your grades are correct. I have
finished grading up through Lab 5. I have not graded the lab 1 re-submits yet.
Let me know ASAP if any of the grades is wrong. Reports of incorrect grades
for the items that I have graded as of today will not be accepted after Dec 4.
Lab #7: Diode and Rectifier Circuits
• learn what diode is
• learn some simple circuits that use diodes
•Energy levels of electrons in
atoms are quantized (1s, 2s, 2p,
etc)
Diode
•When these atoms are brought
together in crystals, these energy
levels become energy bands with
gaps between them.
• Because of the “fermi exclusion
principal” no 2 electrons can have
the same quantum numbers.
• The distributions of electron
energies depends on the
temperature. The highest possible
energy when T=0 is called the
Fermi Energy (EF).
• some materials can be made into
semi conductors by adding small
amounts of impurities. p-type (one
fewer valence e than Si) and ntype (one more valence than Si).
• semi conductors do not conduct
at T=0.
1 eV
Diode
104 – 106 V/cm, 0.3-0.6 V
Electrons migrate from n to the holes in p, giving a net
charge to each side (neg on p side, pos on n side)
In steady state, there are conduction electrons who have
enough thermal energy to go over the potential barrier and
make a (positive) current from p to n. There is an equal and
opposite current of electrons that get enough thermal energy
to break their bonds in the p region, go up into the conduction
band, and then slide down the barrier back to the n region.
Diode
+
-
Here, the external bias aids the
electrons trying to go from n-p
and you get a net current in the
p to n direction.
-
+
Here, the external voltage stops the
electrons going from n to p. The
ones going from p-n can still go and
you get a (small) net current from n
to p (the prob to get enough thermal
energy to break the bond is
unchanged)
I  I 0 (e
qV / kT
 1)
Diode
symbol
Idealized I vs V
Realistic I vs V
I  I 0 (e
qV / kT
 1)
Half-wave rectifier
Full wave rectifier
RC differentiator
See pg 36
For t>>RC. Voltage
across resistor is
derivative of VIN
frequency to voltage converter
Hints
• lab begins pg 70 no error analysis this lab
• IV.A. Do 10 measurements.. RMS, not peak-to-peak, Try pulling/pushing the
amplitude knob on the scope. You should be able to go up to about 7.5V and
down to 0.1 V.
• IV.A fit the upper points (the part that looks linear)to a straight line,
extrapolate the line down to the lower points. Include a graph with all the
points and the line.
• IV.B. Replace the last sentence with: For 3 different generator voltages,
measure the peak-to-peak and RMS voltages for both the generator and load
voltage. For each, make a plot of generator versus load, fit to a straight line,
and comment on the slope and intercept
•IV.C. Be careful! Not the normal cap
• IV.C. Replace instructions with: for 8 different values of the frequency, plot
the mean and peak-to-peak voltage across the cap versus frequency. Make
sure at least 4 of the frequencies are between 20 and 100 Hz, and the highest
is around 60 kHz. When measuring the mean, make sure the scope input is
DC coupled. When measuring the peak-to-peak, make sure it is AC coupled
and that the scale is set to a small enough setting that you can see structure.
Use the cursors, not measure for the peak-to-peak. At very high frequencies,
the structure will go away. Report 0 for the peak-to-peak.
Hints
• skip V-A, V-B, and V-C