← About PHYS 252

PHYS 252 · Electronics

Lecture 01 · Electronic tools & power distribution

1 Where Lab 1 Sits

Lab 1 has three parts. First you learn the bench and measure parts with the digital multimeter (DMM). Then you build a source out of resistors and watch what happens when something draws current from it. Last, you read a signal on the oscilloscope and save it.

Lab 1 in three rows. Know the bench: Task 1, the station (breadboard, supplies, grounding), then Task 2, measuring three resistors with the DMM. Build a source: Task 3, a divider from 12 V giving 5 V, 3.3 V and 1.8 V, then Task 4, loading the divider and watching it sag. Read a signal: Task 5, triggering, measuring, exporting and plotting on the scope.
Figure 1. Lab 1 in three rows: know the bench (Tasks 1 and 2), build a source (Tasks 3 and 4), read a signal (Task 5).

1.1 Task map

Lab 1 taskWhat you doConcept & section
1Learn the station: breadboard, supplies, groundingThe bench (§2)
2Measure 470 Ω, 22 kΩ and 2.2 MΩ resistors; compare with toleranceThe DMM and resistors (§3)
3Design a divider that gives 5, 3.3 and 1.8 V from 12 VSources (§4), the voltage divider (§5)
4Load a divider of two 22 kΩ resistors; find the loading effectLoading (§6)
5Trigger, measure, export and plot a 1 kHz sineThe oscilloscope (§7)

Every task needs an error analysis. Section 8 collects the main error sources for each kind of measurement.

2 The Bench

2.1 The breadboard

A solderless breadboard holds parts without solder. Under each strip of five holes sits one metal spring clip, so the five holes are one electrical node. A lead pushed into any of them is connected to the other four. The two halves of the board, rows a to e and f to j, are separated by a center channel. A chip in a dual in-line package (DIP) sits across that channel, so each of its pins lands in its own strip.

Breadboard connections. Columns of five holes, rows a to e and f to j, each form one node; an IC straddles the center channel so each pin has its own strip. Power rails run along the top and bottom edges; on some boards the top rails are split in the middle and need a wire to bridge the gap.
Figure 2. Breadboard connections. Each column of five holes is one node; an IC straddles the center channel. The power rails run along the edges. On some boards the top rails are split in the middle.

The long rows along the edges are the power rails, usually marked with a red (+) and a blue (−) line. Use one rail for the supply and one for ground, and keep that choice the same on every board you build. On some long boards, each rail is split into two halves. If half of your circuit seems dead, check for a split rail and bridge it with a short wire.

A breadboard with its backing removed, showing rows of metal spring clips, one under each strip of five holes, and the long clips of the power rails.
Figure 3. A breadboard with its backing removed: one spring clip under each strip of five holes, and long clips under the rails. Photo: Zeroping, CC BY 4.0.

2.2 Two kinds of power supply

Your station can power a circuit in two ways. An integrated supply is built into the prototyping station. It may have fixed outputs, adjustable ones, or both, and its ground is shared with the rest of the station. An external bench supply is a separate box with its own voltage and current knobs. Its outputs are often floating: the − terminal is not connected to anything until you connect it to your circuit's ground.

A bench DC power supply with red digital displays for current (0.00 A) and voltage (11.9 V), current and voltage knobs, and red and black output terminals.
Figure 4. A bench DC power supply with current and voltage displays and knobs. Photo: Derrick Parker, CC0.

A bench supply holds its output voltage until the load asks for more than the current limit. Then it switches to holding the current at the limit, and the voltage drops. Set the limit a little above what your circuit should draw. A wiring mistake then shows up as a sagging voltage instead of a hot part.

2.3 One ground for every instrument

A voltage is always measured between two points. "The output is 5 V" means 5 V above ground, so every instrument and every supply must agree on where ground is. Most strange readings in a first lab come from a missing or wrong ground connection.

MistakeWhat you seeFix
Supply − not tied to circuit groundReadings drift or make no senseWire the supply's − terminal to the ground rail
Two supplies, no common groundOne part of the circuit floatsJoin their − terminals
Scope ground clip on a signal nodeThat node shorts to earthClip the ground lead to circuit ground only
Split rail left openHalf the board is deadBridge the gap with a wire

On most bench oscilloscopes and function generators, the outer shell of each BNC connector is tied to mains earth. Every scope ground clip is therefore the same node, and so is the generator's ground. Clip a ground lead to a node that is not ground and you short that node to earth through the instruments.

Practice 2

(a) You power a circuit from an external bench supply and measure its output with the scope. The trace jumps around and the DMM shows a different value every time you touch the board. What is the first connection you check?

(b) A circuit needs +5 V and +12 V from two separate supplies. Draw how the four supply terminals connect to the board.

3 The DMM and Resistors

3.1 Connecting the meter

A DMM measures voltage, current and resistance. To measure a voltage, connect the meter across the part, in parallel with it. To measure a current, break the circuit and put the meter in series, so the current flows through it. The current jacks are separate (A or mA), and the meter's resistance there is small. Leave the red lead in the current jack and switch to volts, and you short whatever you touch.

Two circuits. Left: a voltmeter connected across a resistor R powered by a source V_S, labeled voltage: across the part. Right: an ammeter in series between the source V_S and the resistor R, labeled current: in series.
Figure 5. Left: a voltmeter across the resistor. Right: an ammeter in series with it.

To measure resistance, take the part out of the circuit and turn the power off. In circuit, the meter also sees every other path between the two leads, and a powered circuit can damage the meter's resistance range.

A handheld digital multimeter with a large display, a rotary range switch and four input jacks at the bottom.
Figure 6. A handheld DMM: display, range switch, and the COM, V/Ω and current jacks. Photo: Anselm Schüler, CC BY-SA 4.0.

3.2 The color code

Through-hole resistors carry their value in colored bands. On a four-band resistor, the first two bands are digits, the third is the power of ten, and the fourth is the tolerance.

ColorDigitColorDigitTolerance band
black0green5brown ±1 %
brown1blue6red ±2 %
red2violet7gold ±5 %
orange3gray8silver ±10 %
yellow4white9
A through-hole resistor with brown, black, orange and gold bands: 10 kilohms, plus or minus 5 percent.
Figure 7. Brown, black, orange, gold: 1, 0, then three zeros, so 10 000 Ω = 10 kΩ, ±5 %. Photo: oomlout, CC BY-SA 2.0.

Precision resistors often have five bands: three digits, the multiplier, then the tolerance. When you are unsure which end to start from, measure it.

3.3 Standard values

Resistors are not made in every value. They come in E-series: the E12 series has 12 values per decade for ±10 % parts, and the E24 series has 24 values per decade for ±5 % parts. The values are spaced so that each one's tolerance band reaches its neighbors. Any resistance you might need is then within tolerance of some standard part. The same values repeat in every decade: 2.2 Ω, 22 Ω, 220 Ω, 2.2 kΩ and so on.

The E12 and E24 standard values in one decade on a log scale, each drawn with its tolerance band. E12 values 1, 1.2, 1.5, 1.8, 2.2, 2.7, 3.3, 3.9, 4.7, 5.6, 6.8, 8.2 with plus or minus 10 percent bands; E24 values from 1 to 9.1 with plus or minus 5 percent bands. Neighboring bands touch or overlap, so the decade is covered.
Figure 8. E12 (±10 %) and E24 (±5 %) values in one decade, each with its tolerance band, on a log scale.

3.4 What the datasheet says

A resistor's datasheet gives more than its value. Figure 9 is the specification table of a common family of metal film resistors. Three lines matter in this lab. The tolerance tells you how far a new part may be from its nominal value. The rated dissipation tells you how much power it can turn into heat: 0.4 W for the smallest size, at 70 °C. The operating voltage is the most you may put across it.

Technical specifications table from the Vishay MBA/SMA 0204, MBB/SMA 0207 and MBE/SMA 0414 datasheet: resistance range 0.22 ohm to 10 or 22 megohm; tolerance plus or minus 5, 1 or 0.5 percent; temperature coefficient 50 or 25 ppm per kelvin; rated dissipation at 70 degrees C of 0.4, 0.6 and 1.0 W; operating voltage 200, 350 and 500 V.
Figure 9. Technical specifications of the Vishay MBA/SMA 0204, MBB/SMA 0207 and MBE/SMA 0414 metal film resistors. Source: Vishay datasheet, document 28766, p. 1.

Your kit may hold carbon film parts from another maker. Read the band colors for the tolerance, and look up the series for the power rating.

3.5 Deviation from nominal

The percentage deviation compares what you measured with what the bands say:

deviation=Rmeas−RnomRnom×100 %\text{deviation} = \dfrac{R_{\text{meas}} - R_{\text{nom}}}{R_{\text{nom}}} \times 100\,\%(1)
Worked example: a 4.7 kΩ, ±5 % resistor

The DMM reads 4.62 kΩ. The deviation is (4.62 − 4.70)/4.70 × 100 % = −1.7 %. A ±5 % part may lie anywhere from 0.95 × 4.70 = 4.465 kΩ to 1.05 × 4.70 = 4.935 kΩ, so this one is in tolerance.

3.6 The meter's own uncertainty

The DMM is not perfect either. Its manual gives the accuracy for each range, usually as a percentage of the reading plus a number of counts (steps of the last digit). A deviation smaller than the meter's uncertainty cannot be told apart from zero.

Worked example: meter uncertainty

Suppose the manual gives ±(0.8 % + 2 counts) on the 6.000 kΩ range, where one count is 0.001 kΩ. For a reading of 4.620 kΩ:

δR=0.008×4.620 kΩ+2×0.001 kΩ=0.039 kΩ\delta R = 0.008 \times 4.620\ \text{k}\Omega + 2 \times 0.001\ \text{k}\Omega = 0.039\ \text{k}\Omega

So R=4.620±0.039R = 4.620 \pm 0.039 kΩ, or ±0.84 %. The −1.7 % deviation above is about twice that, so it is a real difference.

A resistance axis from 4.4 to 5.0 kilohms. A shaded band from 4.465 to 4.935 kilohms is the plus or minus 5 percent tolerance of a 4.70 kilohm resistor, marked at its nominal value. A measured point at 4.620 kilohms with an error bar of plus or minus 0.039 kilohms (minus 1.7 percent) lies inside the band.
Figure 10. The worked example on one axis: the ±5 % band of a 4.70 kΩ resistor, and the reading of 4.620 ± 0.039 kΩ inside it.

These numbers are an example. Find the specification for your own meter and the range you used. At the extremes, other errors join in: the resistance of the test leads matters for a few ohms, and your fingers on both leads put your body's resistance in parallel with a megohm resistor.

Practice 3

(a) A resistor has bands red, red, yellow, gold. What is its nominal value, and what range of values is in tolerance?

(b) A 1.5 kΩ, ±1 % resistor reads 1.48 kΩ. Is it in tolerance? With the meter specification of §3.6 on the same range, can you be sure?

(c) Why might a 2.2 MΩ resistor read low if you hold both leads between your fingers?

4 Sources

4.1 The ideal source

An ideal voltage source keeps its voltage whatever you connect to it. Connect a resistor RR across an ideal 9 V source and the voltage across it, VRV_R, is 9 V for every RR. With 1 kΩ the current is 9 mA, and with 1 MΩ it is 9 µA. With 1 Ω, the ideal source would push 9 A. No 9 V battery can do that.

A 9 V source V_in connected across a resistor R, with a voltmeter V_R across R. The bottom wire is grounded.
Figure 11. An ideal 9 V source across a resistor RR, with a voltmeter reading VRV_R.

4.2 A real source has a resistance inside

A real source behaves like an ideal one in series with a small resistance RSR_S, the source resistance. The load RLR_L is in series with RSR_S, so the same current II flows through both (IS=ILI_S = I_L). Ohm's law around the loop gives the load voltage:

VL=RLRS+RL VinV_L = \dfrac{R_L}{R_S + R_L}\, V_{\text{in}}(2)
A real source: inside a dashed box, the source V_in in series with a source resistance R_S, carrying current I to two output terminals. The output voltage V_L appears across the terminals, which connect to a load resistor R_L outside the box.
Figure 12. A real source: VinV_{\text{in}} in series with RSR_S, connected to the load RLR_L. The load sees VLV_L.

The voltage across RSR_S is lost inside the source. How much is lost depends on the load.

Worked example: a 9 V source with R_S = 50 Ω

With RLR_L = 1 kΩ: VL=9×1000/1050=8.57V_L = 9 \times 1000/1050 = 8.57 V, 95.2 % of 9 V.

With RLR_L = 1 Ω: VL=9×1/51=0.18V_L = 9 \times 1/51 = 0.18 V. The current is 0.18 A, not 9 A, and nearly all of the voltage is lost across RSR_S.

With RLR_L = 1 MΩ: VL=8.9996V_L = 8.9996 V, effectively 9 V.

4.3 Firm and stiff

Figure 13 shows VLV_L for every load from 1 Ω to 10 MΩ. When RL=RSR_L = R_S the load gets exactly half. Two rules of thumb name the region where the source is close to ideal:

Firm and stiff sources

A source is firm if RS<RL/10R_S < R_L/10: the load gets at least 10/1110/11 = 91 % of the open-circuit voltage.

A source is stiff if RS<RL/100R_S < R_L/100: the load gets at least 100/101100/101 = 99 %.

Load voltage against load resistance, from 1 ohm to 10 megohms on a log axis, for a 9 V source with R_S of 50 ohms. The curve rises from 0.18 V at 1 ohm through 4.50 V at 50 ohms (half) to 8.57 V at 1 kilohm and 9.00 V at 1 megohm, approaching the dashed 9 V line of an ideal source. Shaded regions mark firm (R_L above 10 R_S) and stiff (R_L above 100 R_S).
Figure 13. Load voltage against load resistance for a 9 V source with RSR_S = 50 Ω. The firm and stiff regions are shaded.

"Stiff" is not a property of the source alone. The same 50 Ω source is stiff for a 10 kΩ load and useless for a 10 Ω one.

4.4 The function generator is a 50 Ω source

The function generator on your bench is a source of this kind. Inside, an ideal AC source VocV_{\text{oc}} sits behind a 50 Ω output resistance. The scope input is about 1 MΩ, so the generator is stiff for it: the scope sees the full open-circuit voltage.

A function generator modeled as an AC source V_oc in series with 50 ohms, inside a dashed box, connected to the scope input, a 1 megohm resistor in a second dashed box.
Figure 14. The generator as VocV_{\text{oc}} behind 50 Ω, driving the scope's 1 MΩ input.
A small laboratory function generator with a frequency display, waveform and amplitude controls, and output jacks.
Figure 15. A function generator. Photo: Anselm Schüler, CC BY-SA 4.0.

Many generators have a load setting, 50 Ω or High-Z, that changes only what the display shows. In 50 Ω mode, the display assumes a 50 Ω load, which gets half of VocV_{\text{oc}}. Set 1.00 Vpp in that mode and the generator makes 2.00 Vpp open circuit: a 50 Ω load would get 1.00 Vpp, but the scope's 1 MΩ gets 2.00 Vpp. If your scope shows twice what you set, check this setting before anything else.

Practice 4

(a) A 12 V source has RSR_S = 2 Ω. What does a 100 Ω load get? Is the source firm, stiff, or neither, for that load?

(b) For the same source, what is the smallest load for which it is stiff?

(c) A generator in 50 Ω mode is set to 300 mVpp. What does the scope show at its 1 MΩ input? What would it show if you added a 50 Ω resistor across the scope input?

5 The Voltage Divider

5.1 Two resistors in series

Two resistors in series across a source carry the same current, so the source voltage splits between them in the ratio of their resistances. The output, taken across R2R_2, is

Vout=R2R1+R2 VinV_{\text{out}} = \dfrac{R_2}{R_1 + R_2}\, V_{\text{in}}(3)
A voltage divider: a 9 V source V_in drives R_1 of 1 kilohm in series with R_2 of 2 kilohms to ground; V_out is taken at the node between them.
Figure 16. A divider from 9 V: R1R_1 = 1 kΩ, R2R_2 = 2 kΩ, so VoutV_{\text{out}} = 9 V × 2/3 = 6 V.

Equation (3) holds only while nothing draws current from the output. Section 6 deals with the load.

5.2 Choose the total resistance first

Only the ratio of R1R_1 and R2R_2 sets the output. The total sets the current, and with it the power:

Worked example: the 1 kΩ and 2 kΩ divider

I=9 V/3 kΩ=3I = 9\ \text{V}/3\ \text{k}\Omega = 3 mA. The chain uses P=VinIP = V_{\text{in}} I = 27 mW: I2R1I^2R_1 = 9 mW in R1R_1 and 18 mW in R2R_2. Both are far below the 0.4 W of Figure 9.

Small resistors draw more current, heat up and drain a battery. Large resistors make a weak source that sags as soon as a load is connected (§6). A design starts by picking a total that balances the two, then splits it.

Two plots against the total resistance of a 6 V divider from 9 V, from 300 ohms to 3 megohms. Left, log scales: the current drawn falls from 3 mA at 3 kilohms to 0.3 mA at 30 kilohms and 0.03 mA at 300 kilohms. Right: the output with a 100 kilohm load stays near the unloaded 6.0 V for small totals (5.960 V at 3 kilohms, 5.625 V at 30 kilohms) and drops to 3.600 V at 300 kilohms.
Figure 17. A 6 V divider from 9 V (R2=2R1R_2 = 2R_1) for every total resistance. Left: the current it draws. Right: its output with a 100 kΩ load: 5.960 V at 3 kΩ total, 5.625 V at 30 kΩ and 3.600 V at 300 kΩ (§6 explains the sag).

5.3 Several outputs from one chain

A chain of three or more resistors gives several outputs at once. Each resistor gets the share of the total voltage that drops across it. Work from the bottom up: the bottom resistor sets the lowest tap, the next one the difference to the second tap, and so on.

Worked example: 6.0 V and 2.5 V from 9 V

Aim for about 30 kΩ in total. The bottom resistor carries 2.5 V, the middle one 6.0 − 2.5 = 3.5 V, and the top one 9 − 6.0 = 3.0 V:

RC=2.59×30 kΩ=8.33 kΩRB=3.59×30 kΩ=11.67 kΩRA=3.09×30 kΩ=10.0 kΩR_C = \tfrac{2.5}{9} \times 30\ \text{k}\Omega = 8.33\ \text{k}\Omega \qquad R_B = \tfrac{3.5}{9} \times 30\ \text{k}\Omega = 11.67\ \text{k}\Omega \qquad R_A = \tfrac{3.0}{9} \times 30\ \text{k}\Omega = 10.0\ \text{k}\Omega

The nearest E24 values are 8.2 kΩ, 12 kΩ and 10 kΩ, 30.2 kΩ in total. Recompute with the real values:

V6=9×12+8.230.2=6.02 V (+0.3 %)V2.5=9×8.230.2=2.44 V (−2.3 %)V_{6} = 9 \times \frac{12 + 8.2}{30.2} = 6.02\ \text{V}\ (+0.3\,\%) \qquad V_{2.5} = 9 \times \frac{8.2}{30.2} = 2.44\ \text{V}\ (-2.3\,\%)

The chain draws 9 V/30.2 kΩ = 0.30 mA and uses 2.7 mW.

A three-resistor chain from a 9 V source: R_A 10 kilohms, R_B 12 kilohms and R_C 8.2 kilohms to ground, with a 6.0 V tap between R_A and R_B and a 2.5 V tap between R_B and R_C.
Figure 18. The worked example: RAR_A = 10 kΩ, RBR_B = 12 kΩ and RCR_C = 8.2 kΩ from 9 V, with taps at 6.0 V and 2.5 V.

Rounding is one error. Tolerance is the other: with ±5 % parts, each resistor can be off by 5 %, and the taps move with them. The worst case for a tap puts every resistor below it 5 % high and every one above it 5 % low, or the other way round. For the 2.5 V tap that gives 2.27 to 2.63 V; for the 6.0 V tap, 5.82 to 6.22 V. If a tap must be exact, two standard values in series or a potentiometer can trim it.

Two panels for the 6.0 V and 2.5 V taps. Each shows the target as a dashed line, the output with exact E24 parts as a dot (6.02 V and 2.44 V), and a band for the worst case with plus or minus 5 percent parts: 5.82 to 6.22 V and 2.27 to 2.63 V.
Figure 19. The two taps of the worked example: the target (dashed), the output with exact E24 parts (dot), and the worst-case range with ±5 % parts (band).
Practice 5

(a) Design a chain that gives 10 V and 4 V from 15 V with about 47 kΩ in total. Choose E24 values, then recompute both outputs and their errors.

(b) If the bottom resistor in (a) is 5 % high and the others are exact, which way does each tap move?

(c) What limits how small the total resistance may be? What limits how large?

6 Loading

6.1 A load sits in parallel with R2

Connect a load RLR_L to a divider's output and it sits in parallel with R2R_2. The pair has a smaller resistance than R2R_2 alone,

R2∥RL=R2RLR2+RLR_2 \parallel R_L = \dfrac{R_2 R_L}{R_2 + R_L}(4)

so the output drops:

VL=R2∥RLR1+R2∥RL VinV_L = \dfrac{R_2 \parallel R_L}{R_1 + R_2 \parallel R_L}\, V_{\text{in}}(5)
A loaded voltage divider: source V_in drives R_1 in series with R_2 to ground; a load resistor R_L connects from the middle node to ground, in parallel with R_2, and the load voltage V_L is taken at that node.
Figure 20. A loaded divider: RLR_L in parallel with R2R_2.

The loading effect measures how much the output fell:

loading effect=Vunloaded−VLVunloaded×100 %\text{loading effect} = \dfrac{V_{\text{unloaded}} - V_L}{V_{\text{unloaded}}} \times 100\,\%(6)
Worked example: two 10 kΩ resistors from 9 V

Unloaded, the output is 4.50 V. With a 10 kΩ load, R2∥RLR_2 \parallel R_L = 5 kΩ and VL=9×5/15=3.00V_L = 9 \times 5/15 = 3.00 V, a loading effect of 33.3 %. With a 1 kΩ load, R2∥RLR_2 \parallel R_L = 0.909 kΩ and VLV_L = 0.75 V (83.3 %). With 1 MΩ, VLV_L = 4.48 V (0.5 %).

Output voltage against load resistance, from 100 ohms to 10 megohms on a log axis, for a divider of two 10 kilohm resistors from 9 V. The unloaded output is 4.50 V. Marked points: 1 kilohm gives 0.75 V (83.3 percent low), 10 kilohms 3.00 V (33.3 percent low), 50 kilohms 4.09 V (9.1 percent low) and 1 megohm 4.48 V (0.5 percent low). Loads above 50 kilohms, ten times the 5 kilohm source resistance, are shaded as firm.
Figure 21. Output of the two-10 kΩ divider against load resistance. Loads above 50 kΩ, ten times the divider's 5 kΩ source resistance (§6.2), are shaded as firm.

6.2 Seen from the load, a divider is a source

Section 4 had a source with a resistance inside. A divider behaves the same way. Seen from its output, it is an ideal source VthV_{\text{th}}, the unloaded output, behind a resistance RthR_{\text{th}}, which is R1R_1 and R2R_2 in parallel:

Vth=R2R1+R2 VinRth=R1∥R2V_{\text{th}} = \dfrac{R_2}{R_1 + R_2}\, V_{\text{in}} \qquad R_{\text{th}} = R_1 \parallel R_2(7)
Left: a 9 V source with two 10 kilohm resistors in series, the output taken between them. An arrow points to the right: the equivalent source V_th of 4.5 V in series with R_th of 5 kilohms, driving a load R_L.
Figure 22. The two-10 kΩ divider from 9 V and the source it looks like from its output: 4.5 V behind 5 kΩ, driving RLR_L.

For the two 10 kΩ resistors, VthV_{\text{th}} = 4.5 V and RthR_{\text{th}} = 5 kΩ. Equation (2) with RS=RthR_S = R_{\text{th}} gives the same answers as Equation (5), and the firm rule says the divider is firm for loads above 50 kΩ. That is the design rule for a divider that must feed a load: make RthR_{\text{th}} at least ten times smaller than the load. Lab 2 shows that every circuit of sources and resistors can be reduced this way. That is Thévenin's theorem.

6.3 Even the DMM is a load

A voltmeter must draw a little current to measure, so it is a load too. Many DMMs have about 10 MΩ of input resistance on their DC voltage ranges; your meter's manual gives its value. Against kilohm resistors, 10 MΩ changes nothing you can see. Against megohm resistors, it does.

A divider of two 1 megohm resistors from 9 V, with the DMM's 10 megohm input resistance R_in connected across the lower resistor, shown in a dashed box labeled inside the DMM.
Figure 23. A divider of two 1 MΩ resistors from 9 V, with the DMM's 10 MΩ input across the lower one.
Worked example: measuring a megohm divider

Unloaded, the output is 4.50 V. The meter puts 10 MΩ in parallel with the lower 1 MΩ: 1 MΩ ∥ 10 MΩ = 0.909 MΩ. The meter reads 9×0.909/1.9099 \times 0.909/1.909 = 4.29 V, 4.8 % low. The circuit is fine; the measurement changed it.

Practice 6

(a) A divider of two 4.7 kΩ resistors runs from 10 V. Find VLV_L and the loading effect for a 47 kΩ load.

(b) Find VthV_{\text{th}} and RthR_{\text{th}} for that divider, and use Equation (2) to check your answer to (a).

(c) A divider of two 470 kΩ resistors runs from 10 V. What does a 10 MΩ meter read across the lower one? How could you change the divider so the meter's error drops below 1 %?

7 The Oscilloscope

7.1 Three groups of controls

An oscilloscope draws voltage against time. Its screen is a grid, usually 10 divisions across and 8 high. The controls come in three groups. The vertical controls set volts per division for each channel. The horizontal control sets time per division for all channels. The trigger decides when each sweep across the screen starts.

A four-channel digital storage oscilloscope showing a sine wave and a square wave, with vertical, horizontal and trigger controls to the right of the screen.
Figure 24. A digital storage oscilloscope. Photo: Wild Pancake, CC BY 4.0.

7.2 Triggering

The scope redraws the screen many times a second. If each sweep starts at a random point of the waveform, the sweeps land on top of each other out of step and the screen shows a blur. The trigger fixes this. It waits until the signal crosses a set level in a set direction, then starts the sweep. Every sweep then starts at the same point of the waveform and the trace stands still.

Two scope screens at 0.5 V per division and 100 microseconds per division. Left, no trigger: seven sweeps of a sine wave start at random points and overlap into a blur. Right, triggered: every sweep starts on a rising edge through 0 V at the center of the screen, so the traces sit on top of each other as one stable sine wave.
Figure 25. Left: without a trigger, sweeps start at random points. Right: each sweep starts where the signal rises through 0 V.

For a stable trace, set the trigger source to the channel you are measuring, the slope to rising, and the level somewhere inside the signal. A level above the top of the signal never triggers. Use edge triggering in Normal or Auto mode; Auto keeps drawing even when no trigger comes, which helps you find a signal.

7.3 Reading the screen

Count divisions, then multiply by the scale. In Figure 26, one period spans 4.0 divisions at 100 µs/div, so TT = 400 µs and f=1/Tf = 1/T = 2.50 kHz. Peak to peak spans 4.0 divisions at 0.5 V/div, so VppV_{pp} = 2.0 V. The amplitude is half of that, 1.0 V, and for a sine the RMS value is the amplitude divided by 2\sqrt{2}, 0.71 V.

A scope screen at 0.5 V per division and 100 microseconds per division showing a sine wave. One period spans 4.0 divisions, 400 microseconds, so f = 2.50 kHz. Peak to peak spans 4.0 divisions, 2.0 V. Amplitude 1.0 V, RMS 0.71 V.
Figure 26. A 2.5 kHz sine of 2.0 Vpp at 0.5 V/div and 100 µs/div, with its period and peak-to-peak value read off the grid.

Choose the scales so the signal fills most of the screen: several periods across, and most of the height. Reading is more precise when the thing you read spans more divisions. The scope's built-in measurements and cursors read the same quantities with finer resolution; say in your report which you used.

7.4 Period and frequency errors

Suppose you can place each zero crossing to ±0.1 division, which is ±10 µs at 100 µs/div. A period is the difference of two crossings, so their errors combine:

δT=102+102 μs=14 μsδTT=14 μs400 μs=3.5 %\delta T = \sqrt{10^2 + 10^2}\ \mu\text{s} = 14\ \mu\text{s} \qquad \frac{\delta T}{T} = \frac{14\ \mu\text{s}}{400\ \mu\text{s}} = 3.5\,\%

Since f=1/Tf = 1/T, the relative error of the frequency is the same 3.5 %, but its absolute error is in hertz: δf\delta f = 0.035 × 2500 Hz = 88 Hz, so ff = 2.50 ± 0.09 kHz. Keep the two apart in your report: δT\delta T in microseconds, δf\delta f in hertz.

Measure across several periods and divide, and the reading error is shared among them. To fit four periods (1.6 ms) on the screen you need 200 µs/div, where ±0.1 division is ±20 µs. The four periods then carry ±28 µs, which is ±7 µs per period, 1.8 %: half the error of reading one period.

Two scope screens of a 2.5 kHz sine at 0.5 V per division. Left, 100 microseconds per division: one period spans 4.0 divisions, read to plus or minus 14 microseconds, so T is 400 plus or minus 14 microseconds (3.5 percent). Right, 200 microseconds per division: four periods span 8.0 divisions, read to plus or minus 28 microseconds, so T is 400 plus or minus 7 microseconds per period (1.8 percent). Shaded strips mark plus or minus 0.1 division at each crossing.
Figure 27. Left: one period read at 100 µs/div, ±14 µs (3.5 %). Right: four periods read at 200 µs/div, ±28 µs in all, ±7 µs per period (1.8 %). The shaded strips are ±0.1 division around each crossing.

7.5 Two settings to check

SettingIf it is wrong
Probe factor on the scope channel (1× or 10×)Every voltage is 10 times too large or too small
Generator load setting (50 Ω or High-Z, §4.4)The scope shows twice the amplitude the generator displays

A 10× probe divides the signal by ten and presents about 10 MΩ to the circuit instead of 1 MΩ. The scope only knows to multiply back by ten if the channel's probe setting says 10×. A plain BNC cable from the generator is a 1× connection.

7.6 Export the data, then plot it

A digital scope can save the trace as a CSV file on a USB drive: one column of time and one column of voltage per channel, after a few header lines that differ from scope to scope. Open the file, check the units in the header, and plot voltage against time.

A plot of simulated exported scope data: CH1 voltage in volts against time in milliseconds from 0 to 1.2 ms, a 2.5 kHz sine of 2.0 V peak to peak with slight noise, with labeled axes, a grid and a title.
Figure 28. A report plot of simulated data: a 2.5 kHz sine of 2.0 Vpp, sampled at 500 kS/s.

A good report plot has labeled axes with units, a scale that suits the numbers (milliseconds rather than 0.0012 s), a title or caption that says what was measured, and a source note if the data are simulated.

Practice 7

(a) On a screen at 200 µs/div and 2 V/div, one period of a sine spans 5.0 divisions and the trace spans 3.2 divisions from top to bottom. Find TT, ff, VppV_{pp} and VrmsV_{rms}.

(b) If you can read each crossing to ±0.1 division, what are δT\delta T and δf\delta f for (a)?

(c) The generator says 2.0 Vpp and the scope reads 4.0 Vpp. Name two settings that could explain it.

8 Error Analysis

Every task in this course needs an error analysis. For each value you report, state its uncertainty, name the main sources of error, and compare it with the value you expected, simulated or took from a datasheet.

MeasurementMain error sources
ResistanceDMM specification for the range; lead resistance for small values; your body in parallel for large ones
Divider voltageResistor tolerance and rounding; the supply's set voltage; the meter's input resistance
Loading effectThe errors of the two voltages it is computed from
Period, frequency, amplitudeReading the grid or cursors; trace thickness and noise; the scope's own accuracy; probe and generator settings

A difference between measured and expected is only meaningful when it is larger than the uncertainty. When it is, say which source most likely explains it, and check that the size fits.

← All course materials