PA Amplifier Distortion at High Output with a 4-Ohm Load
A 100 W public-address amplifier is used in a school hall with two 8 Ω speakers connected in parallel, giving a 4 Ω load. At low volume the sound is clean, but as the volume is advanced past about 60 W output, the sound becomes harsh and the heatsink becomes very hot within ten minutes. The amplifier uses a ±45 V supply and a class-AB output stage with a preset bias control. On the bench with a 1 kHz sine input and a 4 Ω dummy load, the output clips at about 28 V peak. The DC offset at the output is 75 mV. The idle current measured across one 0.22 Ω emitter resistor is 3 mV, which corresponds to about 14 mA. The bias preset is at its minimum end. The customer complains that the amplifier 'cannot deliver its rated power' and wants the bias increased. The workshop has a distortion analyzer, an oscilloscope, and a thermal probe.
1. Is the amplifier clipping because of insufficient bias, supply limitation, or load mismatch? Support with calculations.
Read the worked answer
The amplifier is rated 100 W into 8 Ω, but it is being asked to drive 4 Ω. For 100 W into 8 Ω, the required RMS voltage is sqrt(100 × 8) = 28.3 V, peak 40 V, and RMS current 3.54 A. Into 4 Ω, the same 28.3 V RMS would demand about 7.07 A and 200 W, which is beyond the amplifier’s rating. The measured clipping at 28 V peak corresponds to 19.8 V RMS, and into 4 Ω that is P = 19.8² / 4 ≈ 98 W, with about 4.95 A peak current. The output stage is therefore near its current and safe-operating-area limit, not simply starved of bias. A 14 mA idle current is low for class AB and can cause crossover distortion at low levels, but it does not explain harshness only at high output. The dominant problem is the 4 Ω load and the amplifier’s protection/current limit, aggravated by a hot heat sink.
2. What is the risk of simply increasing the bias to cure the complaint?
Read the worked answer
Increasing the bias raises idle collector current and idle dissipation in the output devices. The heat sink is already very hot at 60 W output, so extra bias will make thermal drift worse and can lead to thermal runaway, output-device failure, or damaged emitter resistors. Bias adjustment sets the class-AB operating point to reduce crossover distortion; it does not increase the amplifier’s maximum current, safe operating area, or rated load impedance. Turning up the preset may make the amplifier sound slightly different at low volume, but it will not make a 100 W/8 Ω amplifier deliver 200 W into 4 Ω. It may also invalidate the protection calibration and cause premature shutdown. The correct approach is to restore the specified bias and fix the load mismatch.
3. Recommend a test and adjustment plan, including what to tell the customer.
Read the worked answer
First verify the supply rails under load and confirm that the protection circuit is not tripping early. Set the bias to the manufacturer’s specified idle current, typically measured as 20–50 mA per output pair depending on design, and check that the heat sink reaches a stable temperature. Measure DC offset after warm-up; 75 mV is high but not necessarily destructive, and it should be reduced if the preset is available. Test with a 1 kHz sine into 8 Ω, measure distortion at 1 W, 10 W, and rated power, and repeat into 4 Ω only briefly with thermal monitoring. Tell the customer that two 8 Ω speakers in parallel are not suitable for this amplifier. Recommend connecting them in series for 16 Ω, using a single 8 Ω load, or adding a suitable 100 V line transformer/amplifier rated for 4 Ω. Do not advertise more power by increasing bias; the repair is a load and thermal solution, not a bias tweak.