I have a few questions on amp power that I'm hoping you guys can help me better understand. To my knowledge the amount of power you need for a speaker system is dependent on the following :
1)Room Size, 2)Distance seated from the speakers 3)Listening levels 4)System set up 6) Sensitivity and 5) Content
People often claim that you need to use lots of power to control the bass drivers in a big speaker. Let's take the B&W N803 speakers for example. Sensitivity is around 90 dB, which is rather high but it is nevertheless still considered by most to be a power hungry speaker for some reason.
Why is this? And does one absolutely require 200 or 300 watts to get a good "grip" on those bass drivers? Why wouldn't 100 watts suffice given a particular seated distance and/or listening level? Let's cut to the chase here.
Any engineers in the house who can shed light on this?
1)Room Size, 2)Distance seated from the speakers 3)Listening levels 4)System set up 6) Sensitivity and 5) Content
People often claim that you need to use lots of power to control the bass drivers in a big speaker. Let's take the B&W N803 speakers for example. Sensitivity is around 90 dB, which is rather high but it is nevertheless still considered by most to be a power hungry speaker for some reason.
Why is this? And does one absolutely require 200 or 300 watts to get a good "grip" on those bass drivers? Why wouldn't 100 watts suffice given a particular seated distance and/or listening level? Let's cut to the chase here.
Any engineers in the house who can shed light on this?
Give this a read and a try,
http://www.diyaudio.com/forums/mult...much-voltage-power-do-your-speakers-need.html
http://www.diyaudio.com/forums/mult...much-voltage-power-do-your-speakers-need.html
Of course. To fill a large room is going to need more "power". Not just that though, the room contents and absorbtion play a part too.
Goes without saying. If your sat at the back of a huge room or hall it isn't going to sound as loud as up at the front.
That is where the thread I linked to is so revealing. Even if you were 100% in error it still kind of puts into perspective the actual levels used.
That can mean many things to different folk
Such as lots of bass boost or playing heavily compressed music files etc etcAs reply #4 really. Certain types of music "need" or appear to need different overall amplifier power. Classical music with its high average to peak ratio for example.
Can play the biggest role of all. An 84db/m speaker and a 90db/m need approximately 4 watts and 1 watt respectively for the same level. The numbers grow big very quickly for the lower sensitivity. The higher efficiency speaker will manage around 99db for 8 watts input, the lower efficiency one needing around 32 watts.
A power hungry load might mean the speaker presents a reactive load with difficult phase angles when provoked by a suitable drive waveform. It might mean the speaker has an uneven and low minimum impedance. That's not the same as a difficult reactive load though. Values as low as 3 ohm aren't unusual even though the speaker may be quoted as 8 ohm.
Amplifiers don't really "grip" or damp the speaker no matter what high damping factors suggest.
True, when the amp is operated linearly. What happens when you do drive into clipping is another matter.. and the "effective" damping factor will be a whole lot worse then the numbers claimed. If you drive an amp with a woosy power supply or inadequate reservior caps into clipping with heavy bass it will get VERY sloppy sounding. Look at the waveform on a scope, and it's a mess. Rail voltage jumping up and down, power supply ripple being conducted through saturated outputs (and of couse, producing mixing products with the signal). An amp with more behind it, even if it sarts clipping at a lower power level, will tend to stay cleaner sounding. The waveform will look better too - just flat-topping nice and clean, producing gragefully-degrading 3rd order products as it drives harder.
Guys, thanks for the replies. I think I understand most of the theory, but it bothers me when people generalize and claim you need 250 watts or more for a particular speaker, that it will "provide more grip ... better control, etc etc etc".
I mean, if I run a 100 watt amp with a pair of N803's, jump to 200 watts, the levels should jump 3 dB's, but the mere fact that I added more power doesn't provide more control or "insert some audiophile term".
The other thing I'm not sure of (correct me if I'm wrong), is the amplifier damping. Wouldn't damping factor be insignificant compared to the damping in a speaker, generally?
I mean, if I run a 100 watt amp with a pair of N803's, jump to 200 watts, the levels should jump 3 dB's, but the mere fact that I added more power doesn't provide more control or "insert some audiophile term".
The other thing I'm not sure of (correct me if I'm wrong), is the amplifier damping. Wouldn't damping factor be insignificant compared to the damping in a speaker, generally?
Just today, I had someone tell me "B&W crave power". But why? What makes it crave power and how can one make such a claim without knowing listening levels, room size and seated distance, system set up etc etc? It doesn't make any sense to me.
90 db sensitivity. 8Ω (minimum 3.0Ω) Power handling between 50-500 watts. But again, how does one qualify that statement that they crave power? I don't get it.
An externally hosted image should be here but it was not working when we last tested it.
90 db sensitivity. 8Ω (minimum 3.0Ω) Power handling between 50-500 watts. But again, how does one qualify that statement that they crave power? I don't get it.
It's because damping factor in the real world has to take into account the whole "network" hung on the speaker terminals and not just the actual speaker drive unit. It has to include the crossover network etc. So although the damping factor of the amplifier itself may be "impressive" at the terminals on the PCB of the power amp, by the time the signal has passed through an output inductor, relay contacts, speaker leads, a fuse ?? then the damping factor is greatly reduced anyway.
Remember damping factor is the output impedance of the source (amplifier) divided by the impedance of the load. And the output of the amplifier is really the speaker end of the speaker cable.
So a high or low damping factor in reality makes little difference. Where differences can occur is where a designer deliberately wants to increase the output impedance of an amplifier by adding a small series output resistor. (this is something I favoured with my FET amp and B&W 703'S and I added a 0.22 ohm).
As to craving power and all that... well it sounds just like marketing waffle. I crave the wonderful detailed three dimensional sound stage the B&W's give yet they do that with only a couple of watts and an amplifier with very low damping factor.
I don't do bass management
I would say so yes. It is just marketing talk in the same way that amplifier classes of operation can be talked up to be something they are not.
Power doesn't "control" or exercise control over the speaker cone in the way I think you mean.
That said, amplifiers of varying circuit topologies certainly do differ in their sound presentation but that's not due to power ratings (or damping factor) being able to control the speaker better.
It's also worth noting that speakers with a genuinely large power handling capacity and high efficiency do have the very real ability to play loud, probably far louder than you would ever want in a typical domestic setting.
I had a similar thread to Panos' here,
http://www.diyaudio.com/forums/soli...er-do-you-really-need-domestic-listening.html
Mooly said:
What other way would I mean?
Unless the amplifier was clipping ... that could subjectively sound different vs one that has ample headroom to spare? Since the topic is veering off slightly (and I am guilty of this), what makes up the sonic signature of an amplifier, excluding power and/or damping. There is a case to be made that all amplifiers sound alike when kept within their operating limits, but I wouldn't know a thing about that.
Well that's the thing. You have big B&W speakers, a spec sheet with a 50-500 watt rating and we don't have any clue how they conducted their power handling tests! Which means their power handling figures are meaningless, really. 500 watts at 1kHz for 0.5 seconds, or 500 watts at 800 Hz for 1 sec, or "insert any example".
I mean if a speaker has a 90 dB sensitivity and you are seated 3 meters away, 100 watts could be more than sufficient to drive you out of the room. Years ago I was always under the impression that by simply adding more power, like doubling the power, would suddenly make your system sound much better.
But then I wonder .. how can anything sound better if I'm not making use of the extra reserves? It's just a load of nonsense. Of course there are plenty of people who would disagree, I'm sure.
Thanks! I'll take a look. I do appreciate your input in this thread!
Last edited:
One thing to keep in mind, though, is that all contributions except the amplifier output impedance and loudspeaker cable impedance are under control of the loudspeaker designer. The effect of loudspeaker voice coil resistance is already included in the Thiele and Small parameters that are usually the basis for the box design, and the loudspeaker designer can compensate for the effect of things like crossover inductor DC resistance by calculating with slightly changed QES and QTS values that are corrected for this effect. The loudspeaker designer can not compensate for the amplifier output impedance or the cable resistance if he or she doesn't know what amplifier and cable are going to be used.
That being said, it indeed doesn't matter much whether the amplifier damping factor is 50 or 50000, as long as the output impedance is totally negligible compared to the voice coil resistance.
Different sonic signatures of amplifiers is a term many designers don't like and some I think even deny it even exists.
A couple of examples of differing sonic signatures as I have found them. The first is what has become known as the "blameless" topology that on paper offers remarkable performance. Based on the classic long tailed input pair (LTP) it has been refined and optimised to a remarkable degree. It sounds clean and sharp with an almost etched quality. While it can't be faulted as such it can be described as being uninvolving to listen too. It's the amplifier where a few recordings sound great but many many more don't. Although you want to sit down and listen to the music your mind is wandering and your bored and so go and do something else instead. I have built and bought amplifiers of this topology and I would say they all had this underlying characteristic to some degree.
Then there is the other extreme, circuits of less technical ability in numbers but circuits that produce an amplifier that sounds great pretty much whatever is played. Your compelled to listen, you don't want to switch off.
There's a lot in between those two examples such as amplifiers that use lateral FET's, amplifiers that use valves and so but I find the "blameless" type always seem to produce this uninvolving sound to some degree. That is a personal view though Amplifiers such as the classic JLH69 (a 10 watt 4 transistor design) which use what is termed a "single ended" rather than differential input stage which tends to produce even harmonic distortion as opposed to the more predominantly odd order of the LTP always seem much more musical and relaxing to listen too.
I'd have to look back to Panos' thread where I posted detailed measurements and used an SPL meter but for me more than a few watts is unbearably loud in my room. You would have to put the speakers in a large hall to be able to make use of 100+ watts of power.
There's no right and wrong answer to all this. It comes down to what you want and like to listen too.
A couple of examples of differing sonic signatures as I have found them. The first is what has become known as the "blameless" topology that on paper offers remarkable performance. Based on the classic long tailed input pair (LTP) it has been refined and optimised to a remarkable degree. It sounds clean and sharp with an almost etched quality. While it can't be faulted as such it can be described as being uninvolving to listen too. It's the amplifier where a few recordings sound great but many many more don't. Although you want to sit down and listen to the music your mind is wandering and your bored and so go and do something else instead. I have built and bought amplifiers of this topology and I would say they all had this underlying characteristic to some degree.
Then there is the other extreme, circuits of less technical ability in numbers but circuits that produce an amplifier that sounds great pretty much whatever is played. Your compelled to listen, you don't want to switch off.
There's a lot in between those two examples such as amplifiers that use lateral FET's, amplifiers that use valves and so but I find the "blameless" type always seem to produce this uninvolving sound to some degree. That is a personal view though
Amplifiers such as the classic JLH69 (a 10 watt 4 transistor design) which use what is termed a "single ended" rather than differential input stage which tends to produce even harmonic distortion as opposed to the more predominantly odd order of the LTP always seem much more musical and relaxing to listen too.I'd have to look back to Panos' thread where I posted detailed measurements and used an SPL meter but for me more than a few watts is unbearably loud in my room. You would have to put the speakers in a large hall to be able to make use of 100+ watts of power.
There's no right and wrong answer to all this. It comes down to what you want and like to listen too.
Some valve amplifiers, especially those without overall feedback, have very low damping factors indeed, meaning that the output impedance is anything but negligible compared to the voice coil resistance. That can indeed give substantial errors in the bass response.
By the way, they used to do it quite differently in historic valve radios. Those often had a pentode output stage with no feedback, resulting in a damping factor smaller than unity. The resulting bump in the low-frequency response was used to compensate for the acoustic roll-off caused by the perforated back of the radio enclosure (and the roll-off of the output transformer). In the end the amplifier output impedance just needs to be close to what the loudspeaker designer expected.
By the way, they used to do it quite differently in historic valve radios. Those often had a pentode output stage with no feedback, resulting in a damping factor smaller than unity. The resulting bump in the low-frequency response was used to compensate for the acoustic roll-off caused by the perforated back of the radio enclosure (and the roll-off of the output transformer). In the end the amplifier output impedance just needs to be close to what the loudspeaker designer expected.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.