Interesting... I never knew that.
I was just idly searching for something (as you do
) and came across this.Speaker Impedance (Z)
Not sure how accurate all the info is but its an interesting read.
All those reviews that state that adding a monster power amp provided all kinds of extra control is meaningless. But then that also means that people are mistaken in their experiences too. How many times have you heard a guy buying a Krell.. or equivalent and it transformed the bass etc?
If damping is irrelevant especially with solid state then all those claims are basically made up? Or generated in the mind? I suppose if claims of high-end cabling can make supposedly large transformations to system performance then the same can happen with super high priced amplification.
If damping is irrelevant especially with solid state then all those claims are basically made up? Or generated in the mind? I suppose if claims of high-end cabling can make supposedly large transformations to system performance then the same can happen with super high priced amplification.
Cabling is a strange one because the effects can be real.
As we are talking of damping imagine the effect of a few metres of extortionately priced thin "super wonder audiophile" cable that adds say 0.4Ω to the total speaker lead resistance. Taking my B&W's and their 3Ω minimum impedance the damping factor would now be only 7.5 at that impedance even assuming an amplifier with a damping factor of 1000000. That would certainly be audible and might even be sonically pleasing. As ever so much depends on each individual case, the speakers your personal taste, the listening room characteristics and so on. The extortionatley priced cable might thus get a great review as it "transformed" the system. A couple of £$0.10 resistors would achieve the same effect.
If you ever want to try an experiment then just add something like a 0.33 ohm resistors in series with your speakers. You might like it, you might not.
As we are talking of damping imagine the effect of a few metres of extortionately priced thin "super wonder audiophile" cable that adds say 0.4Ω to the total speaker lead resistance. Taking my B&W's and their 3Ω minimum impedance the damping factor would now be only 7.5 at that impedance even assuming an amplifier with a damping factor of 1000000. That would certainly be audible and might even be sonically pleasing. As ever so much depends on each individual case, the speakers your personal taste, the listening room characteristics and so on. The extortionatley priced cable might thus get a great review as it "transformed" the system. A couple of £$0.10 resistors would achieve the same effect.
If you ever want to try an experiment then just add something like a 0.33 ohm resistors in series with your speakers. You might like it, you might not.
That piece exaggerates the difference between SS and valve amps, unless the author is only thinking about SE valve with no overall feedback. He also says that there is no 'second input' point in a valve amp for applying feedback; wrong, the input cathode works quite well. The piece is well written so easy to understand, but a newbie would not know which bits to believe and which to ignore.Mooly said:
Would I be correct in saying that if my speakers only need 100 watts to reproduce dynamic peaks without clipping in my room, at my seated distance and at my preferred listening levels that adding more power is really just redundant? It would seem so.
People have told me the importance of headroom, how the average music/listening level could vary wildly (perhaps -12db below peaks) so in some instances you might need 16x the power than at average listening to reproduce the peak properly.
But it all sounds like there are too many "what if's" to give a correct answer.
People have told me the importance of headroom, how the average music/listening level could vary wildly (perhaps -12db below peaks) so in some instances you might need 16x the power than at average listening to reproduce the peak properly.
But it all sounds like there are too many "what if's" to give a correct answer.
Another thing. Correct me if I'm wrong but the bass sections of a speaker usually require more power than the mid and/or highs all things being equal. But a large speaker or larger (all things being equal again) usually has improved sensitivity, so then doesn't that mean that the bass drivers are easier to "drive" with a given amount of power at a particular volume level?
Last edited:
A larger woofer will only have a higher sensitivity if it is designed to work in a large *box*. People don't like large boxes, so the result is low sensitivity woofers. Mids and tweets usually have decent sensitivity without any special measures being taken. So usually in a 3 way hi-fi system the woofer ends up with a lower sensitivity than the mid and tweet, and EQ is applied in the passive crossover. I'm sure you've heard lots of "newbie" 3-way builds that sound hot as hell on top.
And since the bass does require the majority of the power, it's surprisingly easy to clip the tar out of a domestic system if you boost the bass.
And since the bass does require the majority of the power, it's surprisingly easy to clip the tar out of a domestic system if you boost the bass.
Today I heard someone tell me that their Bryston monoblock amps (600 watt 8 ohm/900 watt 4 ohm) clipped while using Tannoy 95 dB sensitivity speakers. To make matters worse he says he wasn't showing off.
I can't even fathom how such powerful amps could possibly clip with such high sensitivity speakers.
I can't even fathom how such powerful amps could possibly clip with such high sensitivity speakers.
There are to many what ifs for sure.
One area of confusion is actually talking in terms of power. When you say "my speakers only need 100watts to reproduce dynamic peaks without clipping in my room" what do you really mean
Think carefully before you answer
Lets say you apply a frequency sweep going from 20hz to 20khz across your speaker. Your amp is rated at 100 watts RMS into 8 ohms so we set the output voltage to be 28.3 volts RMS. If we connect an 8 ohm resistor across the output it dissipates 100 watts RMS irrespective of the frequency. If we now swap the resistor for a speaker what happens. Assuming a perfect amplifier the voltage is still 28.3 volts RMS but the load is different for every change in frequency. It might be around 3 ohm at perhaps 30 hz and as high as 20 or 30 ohm at the top end and with dips and valleys in between. So the "power" dissipated by the speaker is constantly varying as the frequency varies.
That 100 watt amplifier might really be 100 watts into 8 ohm. If the load impedance is halved the "power" in the load should double but very few amplifiers manage that and go into current limiting.
So there is a case for amplifiers to rated in terms of voltage output into a given load rather than a nominal "power" figure. At least one forum member firmly believes that should be the case and it's a view I am coming to agree with.
So now lets imagine the case of an active speaker with three drivers and three power amps. The bass amp might need to supply 28.3 volts RMS into as low as 3 ohm. So that's a 266 watt amp. The mid driver might be an easier load of say 12 ohms (totally guessing on that
) and so needs an amp that can still put out 28.3 volts but this time in to 12 ohms. So a 66 "watt" amp is sufficient (but it still puts out 28.3 volts RMS) and the tweeter might be no lower than 20 ohms and so a 40 watt amp is OK as long as it can put out 28.3 volts RMS.The two tests, Panos' and mine linked to earlier are well worth having a play at. They give you very relevant results to help determine what power (voltage... voltage
) you need.
Classical is more dynamic than other types of music so I generally work on 20dB of headroom to be on the safe side.
This would give an rms input of a mere 6W to use 600W on peaks with very dynamic works.
Tannoy drivers are famously current hungry, they just love that.
I've come across a few people who bought Tannoy DMT215s for their SET valve amps as they do 101dB 1W @ 1m only to find out that they only really come to life when driven by huge SS amps. Tannoys also sound better the higher the amps DF.
Don't know why and also I don't really want to get in an argument about that but that is the general consensus on the Tannoy Yahoo group and since I own Tannoys I agree.
I've tried a number of amps on mine but they do sound better the more power is available and the higher the amps damping factor.
I now run mine active with a total of 900Wrms into 8Ohm per channel and couldn't be happier.
This would give an rms input of a mere 6W to use 600W on peaks with very dynamic works.
Tannoy drivers are famously current hungry, they just love that.
I've come across a few people who bought Tannoy DMT215s for their SET valve amps as they do 101dB 1W @ 1m only to find out that they only really come to life when driven by huge SS amps. Tannoys also sound better the higher the amps DF.
Don't know why and also I don't really want to get in an argument about that but that is the general consensus on the Tannoy Yahoo group and since I own Tannoys I agree.
I've tried a number of amps on mine but they do sound better the more power is available and the higher the amps damping factor.
I now run mine active with a total of 900Wrms into 8Ohm per channel and couldn't be happier.
Mooly, thanks for your explanations, you have been a great help. Thanks to everyone. You brought up very good points on amplifier damping. The damping of the speaker components themselves are the predominant factor in terms of total system damping.
I think I read somewhere online that more power (however you define "power") equals more stopping power in a speaker. For those who aren't knowledeable or are not EE's, people will read this stuff and probably believe it. For the most part anyway.
I think I read somewhere online that more power (however you define "power") equals more stopping power in a speaker. For those who aren't knowledeable or are not EE's, people will read this stuff and probably believe it. For the most part anyway.
I know this topic is probably veering off completely and it's my fault, but this is getting interesting. The stopping and starting of a speaker .... that is related to VC inductance? What other variables are at play? Suspension? Damping in the enclosure, enclosure type?
Sorry, I realise this isn't the correct thread. Hopefully we can still continue this here but if not, moderators just let me know.
Sorry, I realise this isn't the correct thread. Hopefully we can still continue this here but if not, moderators just let me know.
90dB sensitivity is less than 86dB/W into a 3Ω minimum loudspeaker.
At 10 foot into a room that would be about 75dB/W, so 100W would be 95dB peak, and on dynamic program material maybe 80dB.
Classical music can hit 105dB (peak), so allowing for 3dB of power compression you would need 2KW to reproduce the full dynamic range of an orchestra. Figuring a pair of speakers in the room you would only need 1KW per channel.
At 10 foot into a room that would be about 75dB/W, so 100W would be 95dB peak, and on dynamic program material maybe 80dB.
Classical music can hit 105dB (peak), so allowing for 3dB of power compression you would need 2KW to reproduce the full dynamic range of an orchestra. Figuring a pair of speakers in the room you would only need 1KW per channel.
Well I'm not an expert on speaker design and their mechanics so I can only try and reason this out...
A large bass driver cone has considerable mass and so if a "step function" type voltage were applied it would take a finite time to move to its new position. Much like an analogue voltmeter doesn't instantly have the pointer deflect to a new voltage... it takes time. When the voltage is removed, again it takes time to return to zero. An interesting observation is that the voltmeter returns to zero far more slowly if the meter terminals are instantly shorted. The meter is heavily damped. So the same must happen with the speaker and the impedance across the terminals (the crossover network, amp and cables) must influence what happens when the speaker is driven instantly to different cone positions by a DC source.
I would also suggest though that there is nothing in music that comes even close to that sudden voltage change and so the cone is able to "follow" the audio signal correctly. The higher frequency components of a transient don't move the bass driver quickly enough but they are reproduced by the lighter mid and hf units.
I'm sure all the things you mention play a part. Sealed enclosure or ported etc, all must play a part in the theory and practice.
The inductance of the VC is an interesting one because its very hard to instantly change the level of current flowing in a large inductor "instantly". Again, in audio nothing really comes close to doing that.
djk said:
Sure, but what speakers can actually handle 1 KW per channel for any given amount of time before reaching thermal failure? It's ridiculous. Apparently the Tannoy speakers can reach down to 12 Hz or thereabouts. But still, if 900 watts into a 4 ohm load is enough to clip them, I can't see how adding a 1.5 kw amp is going to rectify matters if the voice coils are burnt to a crisp. At some point there will be too much heat and the coils will melt.
Sorry for dragging this thread up ... just another question. I know we covered amplifier damping and I know that it was claimed that wattage does not provide better control over the cone, but why is that? Why doesn't higher wattage provide better control? What does provide the control? Just the speaker, primarily?
I'm not disagreeing with anyone, just would like a more detailed explanation if possible.
I'm not disagreeing with anyone, just would like a more detailed explanation if possible.
It's hard to put into words.
As long as the amplifier output is within its design limits it doesn't matter.
Imagine you are sat listening to an amp and its turned up loud and the peak voltage across the speaker is say 20 volts (40v peak to peak). That would be a realistic figure.
And we'll say the amp runs on -/+30 volt rails. The amp can actually swing say -/+25 volts maximum on those rails. Depending on the speaker impedance that figures defines the max power.
If we now take the same amp and increase the PSU to say -/+50 volts we have greatly increased the voltage swing available... lets say to -/+45 volts.
The two different powered amps are the same at the original listening level. There's no difference.
As long as the amplifier output is within its design limits it doesn't matter.
Imagine you are sat listening to an amp and its turned up loud and the peak voltage across the speaker is say 20 volts (40v peak to peak). That would be a realistic figure.
And we'll say the amp runs on -/+30 volt rails. The amp can actually swing say -/+25 volts maximum on those rails. Depending on the speaker impedance that figures defines the max power.
If we now take the same amp and increase the PSU to say -/+50 volts we have greatly increased the voltage swing available... lets say to -/+45 volts.
The two different powered amps are the same at the original listening level. There's no difference.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.