If we look at the Fletcher Munson curves the bass sensitivity of the ear is shown to be very poor. So wouldn't this mean that bass signals are typically pretty high in amplitude in normal music. Even the softest bass might be fairly large relative to the higher range.
Wouldn't this enable us to use power amps run almost in classB ( under biased at say 10 mA or less bias). So that we are out of the critical ' thermal bias issues' region. Cross over harmonics should be pretty much reduced due to the driver HF roll off and possibly any passive crossover .
Worth exploring ?
Has anyone tried this ?
Wouldn't this enable us to use power amps run almost in classB ( under biased at say 10 mA or less bias). So that we are out of the critical ' thermal bias issues' region. Cross over harmonics should be pretty much reduced due to the driver HF roll off and possibly any passive crossover .
Worth exploring ?
Has anyone tried this ?
You can use any class you like. However, if we are talking high powered amplifiers, Class H & Class D are used as they are more efficient in terms of power consumption than Class B, Class A and/or Class AB.
You miss the point !
I was asking about an amp that has power devices that are practically off under quiescent conditions. So it will have potentially quite a bit of cross over distortion when there is a signal passing through it. However it will be used only to reproduce signals below 100Hz . So under these conditions would the crossover distortion be heard ?
I was asking about an amp that has power devices that are practically off under quiescent conditions. So it will have potentially quite a bit of cross over distortion when there is a signal passing through it. However it will be used only to reproduce signals below 100Hz . So under these conditions would the crossover distortion be heard ?
It's not about Fletcher Munson. Your ear is actually more sensitive to the harmonics than the fundamental down that low. Distortion is actually more audible. But you get so much less distortion 20-100 Hz than you do higher in the spectrum because most amplifier topologies have more available distortion-reducing feedback at the low end. You can therefore afford to run your bias low if you don't care about 20k.
The particular topology I use for my PA amps has 3 voltage gain stages, and an incredible amount of loop gain at 100 Hz. With compensation it's no better than a Blameless at 20k, but no worse either. These amps sound fine even run full range, with only 3-5mA of bias per output pair. At 100Hz, the amps should theoretically have ppm distortion levels but secondary effects like electromagnetic coupling and thermal feedback in resistors will far exceed that.
If you really want to reduce heating, class B won't help much. Class H is a huge improvement and easy to implement onto your favorite 'host amplifier' topology (unlike class D which can result in a fireball for the inexperienced). If you're worried about bias drift on the other hand, that's why I only run 3-5mA.
The particular topology I use for my PA amps has 3 voltage gain stages, and an incredible amount of loop gain at 100 Hz. With compensation it's no better than a Blameless at 20k, but no worse either. These amps sound fine even run full range, with only 3-5mA of bias per output pair. At 100Hz, the amps should theoretically have ppm distortion levels but secondary effects like electromagnetic coupling and thermal feedback in resistors will far exceed that.
If you really want to reduce heating, class B won't help much. Class H is a huge improvement and easy to implement onto your favorite 'host amplifier' topology (unlike class D which can result in a fireball for the inexperienced). If you're worried about bias drift on the other hand, that's why I only run 3-5mA.
The mention of the Fletcher Munson curves was to only support the assumption that bass levels are generally higher than mid frequencies.
Yes you are correct when you say that the ultimate aim is to make the amp completely dc bias current stable even at high operating wattage's and therefore temperatures. Nothing to do with over all efficiency of the amp.
Note that the harmonics generated by the crossover will not pass through any other part of the system as it is an active sub amp and only connected to the bass driver. But yes, it also depends on the sub driver's HF response.
But you bring up a good point. Feedback being highest at LF it might just be possible to have little or no bias at frequencies below 100Hz and still have inaudible distortion artifacts ! Or low enough to be masked by the level of the upper range?
Yes you are correct when you say that the ultimate aim is to make the amp completely dc bias current stable even at high operating wattage's and therefore temperatures. Nothing to do with over all efficiency of the amp.
Note that the harmonics generated by the crossover will not pass through any other part of the system as it is an active sub amp and only connected to the bass driver. But yes, it also depends on the sub driver's HF response.
But you bring up a good point. Feedback being highest at LF it might just be possible to have little or no bias at frequencies below 100Hz and still have inaudible distortion artifacts ! Or low enough to be masked by the level of the upper range?
No.
As I mentioned previously it will not matter so you can use any class you like.
Your biggest issue will be power consumption. If you do not have enough current to feed a high-powered amplifier properly, you will be faced with distortion emitting from the output terminals.
I own lots of amplifiers from various classes and, using a class H or class D is a better alternative for a high-powered amplifier than class B for bass applications.
In general, most recorded music envelopes have similar bass/midrange levels, that is fairly even power response from around 60 Hz to 1000 Hz or so, often with mids peaking above bass, and a general reduction of level above 1000 Hz (or so).
Lower LF may extend at a reduced level, or in some genres higher levels below 60 Hz or so.
Some music is mixed with as much as 20 dB more level in the 40-100 Hz range, and many playback systems are voiced with a big 5-15 dB "haystack" boost in that region (a practice I am not a fan of except for very low SPL playback).
At any rate, whether crossover distortion due to amp topology would be audible would depend on the level listened at, the genre of music and the sensitivity and raw frequency response of the sub.
Last edited:
- Status
- Not open for further replies.