- Feb 12, 2006
- 61,034
- 5
- 220,925
- Disclosures
- Development Engineer-Mesa Boogie, Development Engineer-Genzler (pedals), Product Support-Genz Benz
I would like to jump in and make a few general comments with regards to this particular subject/topic so it doesn't turn into an issue where players don't trust "D" class from any/all manufacturers.
Speakers do not present themselves as purely resistive loads to amplifiers. In fact, they can be highly complex networks of resistive, capacitive and inductive components. The reason why inductive and capacitive components are of particular interst is that the voltage and current are not in phase with each other, whereas with a pure resistance they are.
Why does this matter? Because most amplifiers (regardless of class) use global feedback to improve real-world performance. For audio amps, generally they are voltage feedback, meaning that a portion of the output voltage is fed back to the input (ideally) 180 degrees out of phase, to correct for non-linearities like rising output impedances, frequency response and undesireable distortions.
When the load is not resistive, the phase shifts between voltage and current can cause instabilities within the amp's global (and local) feedback networks. This is something that all designers are aware of but because of the wildly possibilities in speaker system designs, we do our best to cover reasonable possibilities of such loads as it's impossible to anticipate everything. These kinds of loads also store and release energy back into the amplifier at various times in the waveform's period, and releasing energy at the wrong time is the most common cause for instability.
We have seen several instances of amplifiers becoming marginally unstable due to loads that are considrably outside the industry norms, either because of defective components, incorrect wiring of the crossovers, broken or damaged parts or poor/incorrect designs. Since each power amp design is different, some amps may appear to work ok, some will blow up or in the event of those with very sophisticated protection circuitry and robust design will protect themselves.
On class AB (traditional) power amps, what happens is that the amp tends to oscillate when connected to the load and driven with signal, getting worse the harder the amp is driven and when clipped may stick to the rails, burst into catastrophic oscillation, etc. On a class B amp where the bias current goes to zero at the signal zero crossing point, this discontinuity creates a particularly vulnerable opportunity for oscillation as it generates an "impulse" at this point.
With class D amplifiers, there is always a high frequency modulation carrier present, though it's filtered, a portion of it remains. Take this high frequency carrier and drive it into the right load (incuctive, capacitive and resistive) where the network stores and releases energy at the right times and the amplifier could fail spectacularly. This is why the class D amp goes into protect without signal, it's due to interactions with the residual carrier. Designers of (quality) class D amplifiers are well aware of this and incorporate sophisticated monitoring and protection mechanisms to shut the amps down in the event any of this is detected. In my experience, it is usually caused by a fairly large, under-damped capacitace in either an impedance correcting Zobel position or by an incorrectly calculated crossover filter where the capacitance becomes dominant and the transfer function looking into the crossover from the amp looks very LC with to little R. Since there is always L (inductors and the speakers themselves) generally it's an incorrectly calculated C in the crossover or undervalued R in the Zobel.
I openly encourage everyone to read through this, understanding how complicated it really is (I do appologize how tech-oriented it is but there's also a substantial group of folks on this forum that will understand every single word), to see what we as amp manufacturers consider when designing amp products for you guys and how much effort goes into the analysis of what's out there in the real world and how we can failure-proof designs so they are even more and more reliable. Used to be that am amp would burst into flames when presented with an illegal load, but now, most do a pretty darn good job of protecting themselves.
Hope this helps
Speakers do not present themselves as purely resistive loads to amplifiers. In fact, they can be highly complex networks of resistive, capacitive and inductive components. The reason why inductive and capacitive components are of particular interst is that the voltage and current are not in phase with each other, whereas with a pure resistance they are.
Why does this matter? Because most amplifiers (regardless of class) use global feedback to improve real-world performance. For audio amps, generally they are voltage feedback, meaning that a portion of the output voltage is fed back to the input (ideally) 180 degrees out of phase, to correct for non-linearities like rising output impedances, frequency response and undesireable distortions.
When the load is not resistive, the phase shifts between voltage and current can cause instabilities within the amp's global (and local) feedback networks. This is something that all designers are aware of but because of the wildly possibilities in speaker system designs, we do our best to cover reasonable possibilities of such loads as it's impossible to anticipate everything. These kinds of loads also store and release energy back into the amplifier at various times in the waveform's period, and releasing energy at the wrong time is the most common cause for instability.
We have seen several instances of amplifiers becoming marginally unstable due to loads that are considrably outside the industry norms, either because of defective components, incorrect wiring of the crossovers, broken or damaged parts or poor/incorrect designs. Since each power amp design is different, some amps may appear to work ok, some will blow up or in the event of those with very sophisticated protection circuitry and robust design will protect themselves.
On class AB (traditional) power amps, what happens is that the amp tends to oscillate when connected to the load and driven with signal, getting worse the harder the amp is driven and when clipped may stick to the rails, burst into catastrophic oscillation, etc. On a class B amp where the bias current goes to zero at the signal zero crossing point, this discontinuity creates a particularly vulnerable opportunity for oscillation as it generates an "impulse" at this point.
With class D amplifiers, there is always a high frequency modulation carrier present, though it's filtered, a portion of it remains. Take this high frequency carrier and drive it into the right load (incuctive, capacitive and resistive) where the network stores and releases energy at the right times and the amplifier could fail spectacularly. This is why the class D amp goes into protect without signal, it's due to interactions with the residual carrier. Designers of (quality) class D amplifiers are well aware of this and incorporate sophisticated monitoring and protection mechanisms to shut the amps down in the event any of this is detected. In my experience, it is usually caused by a fairly large, under-damped capacitace in either an impedance correcting Zobel position or by an incorrectly calculated crossover filter where the capacitance becomes dominant and the transfer function looking into the crossover from the amp looks very LC with to little R. Since there is always L (inductors and the speakers themselves) generally it's an incorrectly calculated C in the crossover or undervalued R in the Zobel.
I openly encourage everyone to read through this, understanding how complicated it really is (I do appologize how tech-oriented it is but there's also a substantial group of folks on this forum that will understand every single word), to see what we as amp manufacturers consider when designing amp products for you guys and how much effort goes into the analysis of what's out there in the real world and how we can failure-proof designs so they are even more and more reliable. Used to be that am amp would burst into flames when presented with an illegal load, but now, most do a pretty darn good job of protecting themselves.
Hope this helps
Naw....just funnin'....we don't even have subways around here. 
