Audio Applications
Secondary Breakdown
their much smaller size, no filaments and greater reliability with reduced vulnerability to physical damage.
When used in an audio power amplifier, the advantages of the power MOSFET over the power transistor are much more difficult to describe and would require greater complexity than can be gone into here. However, they can be summarised as follows – the most important point is that the power MOSFET has a negative temperature coefficient whereas the power transistor has a positive temperature coefficient. This means that when a power transistor is handling power it heats up further and consumes more power. This characteristic, called thermal runaway, will result in the destruction of the power transistor if some means is not provided to control it. The power MOSFET on the other hand, although heating up due to the power flow through the device, does not continue to draw more and more power just because its temperature has risen. But in fact has a tendency to stabilize itself – provided adequate head sinking is available to remove the heat generated during normal operation. Incidentally this is less heat sinking than is required for a similarly power rated standard transistor.
Then there is the appearance of secondary breakdown and ‘hot spots’ in a power transistor. This is related to thermal runaway. In order to understand this, one must imagine that the chip silicon inside the power transistor is in fact many smaller transistors connected in parallel. Now, if one of these smaller transistors or a spot on the chip has a greater gain (or amplification factor) than the rest, then that spot will heat up faster and to a greater temperature than the remainder of the transistor chip. This means that whole power dissipation capability of the transistor has been severely reduced and is a major cause of these unexplained output stage failures in large power amplifiers, i.e. over 80Wrms.
The power MOSFET is largely immune to this problem because if a small part of this chip has a higher gain than the rest then its temperature will rise slightly causing that spot to reduce gain and hence stabilization occurs. The power is more evenly distributed throughout the chip and therefore reliability is maintained.
It can be seen from the above that the transistor power amplifier has to have a much larger margin of power dissipation capability and heat sinking in its output stage than the power MOSFET amplifier.
The transistor power amplifier of 100Wrms output into 8Ω can require a driver stage capable of delivering 10W at 1kHz and up to 20W at 20kHz into the input of the output device. The power MOSFET only requires a maximum of 0.01W so a major saving in driver stage componentry and associated noise and distortion can be eliminated.
23
