Electrical Engineering › Section F · Network theorems (dc) › Topic F.4

Maximum power transfer

Connect different loads to the same source and the power they receive rises, peaks and falls again. The peak comes when the load resistance equals the source's Thevenin resistance, but that peak costs half the power, which is why engineers match some systems and deliberately mismatch others.

Too small, too large, just right

Picture any linear network reduced to its Thevenin equivalent, \(E_{Th}\) in series with \(R_{Th}\), feeding a load \(R_L\). If \(R_L\) is zero, a short, the current is as large as it can be but the load voltage is zero, so the load receives no power. If \(R_L\) is enormous, an open, the voltage is nearly \(E_{Th}\) but the current is almost zero, so again the power is almost nothing. Somewhere between those extremes the load power must reach a maximum. The maximum power transfer theorem says exactly where.

The theorem

The load current is \(E_{Th}/(R_{Th} + R_L)\), so the load power is

P_L = \frac{E_{Th}^2 R_L}{(R_{Th} + R_L)^2}

Differentiating with respect to \(R_L\) and setting the result to zero gives a single answer: a load receives maximum power from a linear dc network when

R_L = R_{Th}

This is called matching the load to the source. Note what is fixed and what varies: the network, and so \(R_{Th}\), is given, and you choose the load. Because the Thevenin and Norton resistances are equal, the same condition applies to a Norton equivalent: \(R_L = R_N\).

How much power is that?

At the matched point the current is \(E_{Th}/(2R_{Th})\) and the load voltage is exactly \(E_{Th}/2\), since the source voltage splits equally. The load power is then

P_{L,max} = \frac{E_{Th}^2}{4R_{Th}}

In Norton terms, half of \(I_N\) flows in the load, so \(P_{L,max} = I_N^2 R_N / 4\). For example, a source with \(I_N = 40\,\text{mA}\) and \(R_N = 50\,\Omega\) can deliver at most \(0.04^2 \times 50 / 4 = 20\,\text{mW}\), and only into a 50 Ω load.

Worked example: choosing the load

A 20 V source, − to ground, feeds node a through 3.3 kΩ. Another 3.3 kΩ joins a to ground, and 1.0 kΩ runs from a to terminal b. What load between b and ground receives most power, and how much? 1. Thevenin voltage: \(20 \times \frac{3.3}{6.6} = 10.0\,\text{V}\). 2. Thevenin resistance: \(1.0 + 3.3 \parallel 3.3 = 2.65\,\text{k}\Omega\). 3. So \(R_L = 2.65\,\text{k}\Omega\) and \(P_{L,max} = \frac{10^2}{4 \times 2650} = 9.43\,\text{mW}\). 4. Check by tabulating \(P_L\) for nearby loads:

\begin{array}{c|c|c} R_L & P_L & \eta \\ \hline 1.0\,\text{k}\Omega & 7.51\,\text{mW} & 27\% \\ 2.65\,\text{k}\Omega & 9.43\,\text{mW} & 50\% \\ 4.7\,\text{k}\Omega & 8.70\,\text{mW} & 64\% \\ 10\,\text{k}\Omega & 6.25\,\text{mW} & 79\% \end{array}

The peak is broad: the nearest standard value, 2.7 kΩ, still receives 9.43 mW to three figures.

Efficiency

In the Thevenin model the same current flows through \(R_{Th}\) and \(R_L\), so the fraction of the source's power that reaches the load is

\eta = \frac{R_L}{R_L + R_{Th}}

At the matched point \(\eta = 50\%\): exactly as much power is lost in \(R_{Th}\) as is delivered. Raising \(R_L\) above \(R_{Th}\) raises the efficiency towards 100% while the load power falls; lowering it below \(R_{Th}\) loses on both counts. The \(\eta\) column of the table shows the pattern. This efficiency belongs to the equivalent circuit; the original network's internal losses can differ, because a Thevenin equivalent is exact only at its terminals.

Worked example: power against efficiency

A source has \(E_{Th} = 12\,\text{V}\) and \(R_{Th} = 2.0\,\Omega\). Compare three loads. 1. Matched, \(R_L = 2.0\,\Omega\): the current is 3.0 A, so \(P_L = 3.0^2 \times 2.0 = 18\,\text{W}\), and another 18 W heats the source; \(\eta = 50\%\). 2. \(R_L = 18\,\Omega\): the current is 0.60 A, so \(P_L = 0.60^2 \times 18 = 6.48\,\text{W}\) at \(\eta = 90\%\). 3. \(R_L = 0.5\,\Omega\): the current is 4.8 A, so \(P_L = 4.8^2 \times 0.5 = 11.5\,\text{W}\), but \(\eta = 20\%\) and the source dissipates 46.1 W. 4. Conclusion: matching wins on load power alone, a larger load wins on efficiency, and a smaller load is worse on both.

When the load is fixed instead

The theorem answers "which load?" for a given source. It does not answer "which source resistance?" for a given load. If \(R_L\) is fixed and you can choose \(R_{Th}\), the load power \(E_{Th}^2 R_L/(R_{Th} + R_L)^2\) is largest when the denominator is smallest, that is when \(R_{Th} = 0\). For a 24 V source and a fixed 50 Ω load, a source resistance of 0 Ω delivers 11.5 W, 10 Ω delivers 8.00 W, and a "matched" 50 Ω delivers only 2.88 W. Adding series resistance to a source never helps a fixed load.

Where matching is used

Matching suits systems where the available power is tiny and every microwatt counts, such as an antenna feeding a receiver, or where reflections on cables must be avoided; radio-frequency equipment is commonly built around a 50 Ω standard so that sources, cables and loads all match. Where power is large, efficiency matters more. The mains supply, a car battery and an audio power amplifier all have source resistances far smaller than their loads, so most of the energy reaches the load and the source stays cool. Running a power station matched would waste half its output as heat.

Common mistakes to avoid

First, setting \(R_L\) equal to one resistor in the network instead of the full \(R_{Th}\) seen at the load terminals. Second, writing \(P_{max} = E_{Th}^2/R_{Th}\) and forgetting the factor of 4. Third, thinking maximum power means maximum efficiency; at the match it is only 50%. Fourth, adding series resistance to a source so that it "matches" a fixed load; for a fixed load, less source resistance is always better. Fifth, assuming a short-circuited load receives the most power because the current is largest; its voltage, and so its power, is zero.

Key takeaways

Load power is \(E_{Th}^2 R_L/(R_{Th} + R_L)^2\), and for a given source it peaks when \(R_L = R_{Th}\) (or \(R_N\)). The maximum is \(E_{Th}^2/(4R_{Th})\), or \(I_N^2R_N/4\), with half the source voltage across the load. Efficiency in the Thevenin model is \(R_L/(R_L + R_{Th})\), only 50% at the match. Communication systems match to capture weak signals; power systems keep the source resistance small for efficiency. For a fixed load, the best source resistance is as low as possible.

Flashcards

Flashcard

Maximum power transfer

A load gets most power from a given source when R_L = R_Th

Called matching.

Test your understanding

Question 1 of 5

A network has E_Th = 12 V and R_Th = 6.0 Ω. Which load receives the most power, and how much does it receive?

Question 2 of 5

A load resistance is three times the Thevenin resistance of the source driving it. What efficiency does the Thevenin model give?

Question 3 of 5

A 12 V source must drive a fixed 10 Ω heater. You may choose the source's internal resistance from 0.5 Ω, 5 Ω, 10 Ω or 20 Ω. Which gives the heater the most power?

Question 4 of 5

A Norton equivalent is 20 mA in parallel with 1.0 kΩ. What is the most power it can deliver to any load?

Question 5 of 5

A 9.0 V source, − to ground, feeds node a through 1.0 kΩ; 1.5 kΩ joins a to ground; and 680 Ω runs from a to terminal b. What load resistance between b and ground receives maximum power?

← Previous topicNext topic →