Lab 6: Power Supplies

Virtually all electronic devices need electrical power. Most of the time, our devices need DC power (anything with smarts has transistors, and needs DC and cares about polarity). Batteries are one option, but these are not rock-solid (needing regulation at least) and run out of juice. Our other readily available option is 120 V AC out of the wall—but this needs to be converted to DC before it's useful.

Though a properly equipped lab will have packaged power supplies, it is very useful to know how to "roll your own." Not only might you want to make a power supply for home electronics exploration, but you may:

• need a quick extra power supply to augment your limited lab resources
• want to package a power supply in a custom electronics box so it's AC-ready;
• want to transform DC power already present on a board to some other value;
• be able to make do with a "dirty" but simple regulator, in which case you need to know how to construct it.

So in this lab, you will be exposed to:

1. Center-tapped transformers;
2. Full-wave diode rectification;
3. Ripple elimination/characterization using capacitors;
4. A crude zener-diode regulator for small loads;
5. A transistor-buffered zener diode regulator for larger loads;
6. A fully-regulated ±15 V power supply.

The Actual Lab

Some of the instructions below are less than fully descriptive. You are left to figure some things out on your own.

WARNINGS: Important!

Warning: Unplug the AC cord from the wall when changing the setup. 36 VAC is still dangerous to be poking with fingers!

Warning: Stick transformer leads into breadboard before attempting to measure anything: holding the active leads with your hands while could be dangerous.

Warning: Do not connect a diode directly across the transformer leads: this is not the circuit you want, and could be dangerous.

Warning: never connect the scope ground to anything that disagrees with ground. You'll have to think through this: the transformer outputs are all floating, so that any one can connect to scope ground, but if you choose to ground one of the leads, the others become off limits to the scope ground.

Safety Advice: Once you have established a transformer ground (center tap, in step 2 below), connect this to the wall ground via the ground (green) lead coming out of the AC power cord.

Lab Procedure

1. Characterize the transformer outputs as fully as you can. Use the scope and the DVM. Use the measurement capabilities of the scope (in addition to reading on the scale, use the measurement menu to get at peak-to-peak, frequency, RMS amplitude, etc). Make sure the scope is in DC coupled measurement mode, and that you are using a 10× probe and the scope is also set to 10×. Compare these to the DVM readings (AC RMS and DC measurements), and look at all lead combinations (1 to 2, 2 to 3, 1 to 3). You will need to use proper scope leads, as the scope's range is too small otherwise. See related tip below. You will not be tying the center tap to ground in this step.
2. Use four diodes to rig a full-wave rectifier for + and - outputs, relative to the center tap (which you will now tie to ground). Ignore the single-sided example from class: we're building the dual power supply. Verify the full-wave behavior for both + and - outputs on the scope and sketch the result (both + and - outputs) for the write-up. Measure the heck out of the result (e.g., peak-to-peak, RMS, frequency, etc.) using the scope and the DVM: AC and DC. Sometimes the DVM AC measurement is odd without some load attached, which could simply be the scope itself. As such, you might want to make DVM settings while the scope is probing the same thing.
3. Put capacitors on the outputs (be careful about the polarity!) and measure the effect on DC level and ripple. Then add a load resistor and see what happens. Watch on the scope, and characterize with a battery of numbers. Try two values of capacitors (we have 22 μF and 100 μF available for this lab) and describe/characterize the differences. For the load resistor, try the lowest-valued 0.25-W that will work with the supply. Hint: not the 150Ω resistors. Measure the heck.
4. Now with the weaker (22 μF) capacitors in place, we'll start to regulate. Ordinarily, we might use the beefier capacitors, but this way we exaggerate the ripple and appreciate the regulator's ability. First up is the simple zener. For now, we'll restrict attention to the + side. We'll use a 15 V (±5%) zener.
   1. Pick a resistor that will present approximately a 10 mA load to the 15 V source (be sure to specify choices in write-up). Whenever you do something like this, confirm that the resistor is up to the task in terms of power rating. Most normal-size resistors are rated at 0.25 Watts. Is your resistor safe?
   2. Let's vow to maintain at least 2 mA in the zener. Based on the approximate DC level of the supply (coming off the capacitor), pick a resistor to go between the zener and the supply so that the sum of the zener and load currents will not drag the node below the zener voltage. Be sure to evaluate if the chosen resistor can handle the power. Don't over-compensate in this choice: we'll want the regulator to fail just beyond a 10 mA load.
   3. Hook up the zener and current-limiting resistor, leaving off the load at first. Make sure the zener is in reverse conduction and measure the AC and DC voltages—both at the zener node and at the top-end supply. Use the DVM to measure the AC performance, rather than straining to see it on the scope.
   4. Add the load and verify continued operation. Re-evaluate the AC and DC characteristics of the zener node and the top-end. Is your load high enough that you see increased ripple on the top-end? Does the zener squelch this properly?
   5. You may want to try some intermediate loads (less than 10 mA) and characterize the operation.
   6. Now exceed the design specification: pick a resistor that will demand something like 15 mA. What happens? Explain why. Characterize well. Make sure you don't exceed power limitations in the resistor.
5. Install an npn transistor as an emitter-follower, connecting the zener node to the transistor base (via a roughly 100 ohm resistor so that you can monitor base current). If you're looking at the flat side of the transistor, the collector is on the right and the emitter is on the left (base is in middle). The collector is connected to the top-end (ripply) supply, and the load resistor sits between the emitter and ground. Use the 100 μF capacitor for this section. Characterize the scene using the original 10 mA-yielding resistor. Measure the DC voltages at the top-end, at the zener, and at the emitter (delivered to the load). Measure the AC voltage at the top-end and at the load. Is the regulator regulating? Measure the voltage across the base resistor (mA scale) to get at the base current, and compare this to the load current. Now go to the 15 mA-yielding resistor, performing the same measurements. If it still works, try the 150 Ω 3-W resistor, again performing the same measurements. Does this work? Why or why not?
6. Now pull out all the zener stuff, and replace with an integrated voltage regulator. This time do both the positive and negative supplies. Be aware that the pinouts of the LM7815 and LM7915 differ. Characterize the performance as a function of load. You should also place a small capacitor (0.1 μF or so) at the regulator output (to ground) for the regulator to function properly. Choose loads that do not exceed the power ratings of the resistors. Can you see any "droop" of the regulated voltage? Is the top-end healthy? Is the top-end maintaining a voltage at least 3 V over the regulated voltage (scope's minimum measurement useful here)? If not, is this responsible for any ripple present in the regulated voltage under load? You might want to try the 22 μF capacitors together with the 150 Ω resistor to see if you can "break" the regulator performance.
7. (Optional if you have time and/or are curious): Unplug the transformer and detach the voltage regulator from the rectifier circuit. Set up the lab DC power supply to feed the regulator. Use only the 7815 +15 V regulator. You will probably need to put a capacitor across the input voltage on the breadboard. How high must you keep the input voltage for the output to remain rock steady at its nominal value? Explore this question as a function of load.

Tips

• Trigger the scope on LINE (input AC line) to freeze all your waveforms and ripples, since these will always be at line frequency.
• For evaluating the power dissipated in a resistor, R with V volts across it and I current through it, P = VI = I²R = V²/R. Always perform this calculation before turning on the circuit.
• The DVM allows you to read AC volts and DC volts. In the case of a ripply voltage, the DC will tell you the average, and the AC will tell you the RMS variation about this average (very useful!). So use the DVM to make most of your voltage/ripple measurements, rather than the scope
• If you use a direct connection to the scope, it's voltage scale is limited to 5 V/division. This is not enough to see/measure the sine waves we deal with in this lab. You will need to use 10× scope probes, and set the Probe value to 10× in the channel menu.
• The scopes we use in the lab have a variety of measurement options on the waveforms. Hit the Measure button near top-center. Then select the measurement field you want to modify using the soft-menu keys on the right side of the scope. Now you can cycle through Type. Note that the measurement appears in the field below "Type" so that it is easy to cycle through the various available measurements. Don't trust the number it spits out unless you can confirm that it's reasonable based on the waveform's appearance on the screen. It may not be measuring what you think it is.
• Be aware that the electrolytic capacitors you are using are polarity sensitive. The negative leg must be at a lower voltage than the unmarked positive leg. This means that the negative leg will be at ground for the positive supply, but at the negative output on the negative supply.
• Make sure the heat sink is attached to the voltage regulator: it can get hot.
• Beware that your load resistors can also get hot to the touch.
• If you're pushing (or worried about) power limitations, make an effort to keep the ON duty cycle low: unplug during times when you are not making a measurement. Keep the ON-time short to prevent overheating.

Lab 6 Parts List

Note: some parts—especially many of the resistors you'll use—will come from the general lab stock (drawers).

Lab 6 Write-up

In the write-up:

This is a pretty standard write-up. Explain your component choices (with related calculations). Present your measurements. Summarize performance for each section. Calculate time constants when appropriate, relevant currents, voltages, relationships. In short, let us know that you understood the lab, your results, and what you learned from it.