The Use of POWER [Archive] - Audio & Video Forums

PDA

View Full Version : The Use of POWER



Bingo
03-26-2007, 06:29 AM
:confused5: I read or heard somewhere that many amplifiers do not have enough power to insure good sound quality.... that a speaker rated at less than 95 Sensitivity needed more power to avoid clipping...etc. My own Orsa speakers are rated at 88 db sensitivity...and my NAD C370 has a power rating that is progressive....120 wats and yet it can produce much more...but I don't know how that works. What I am trying to find out is whether an integrated amplifier rated at 250 Watts per channel into 8 ohms would do any harm to my Orsa speakers (88 db sensitivity) or if perhaps it might even improve my sound. The integrated amplifier I have in mind is the Musical Fidelity A-5....which has 250 watts per channel...etc. It also crosses my mind that more wattage in the amplifier might have a good effect on the overall quality of sound produced....anyone dare to tackle that? Bingo

Dusty Chalk
03-26-2007, 06:44 AM
Yeah, it should be good. Don't think of it as more power unless you actually turn up the volume to unreasonable levels, think of it as more headroom.

topspeed
03-26-2007, 10:30 AM
Yeah, it should be good. Don't think of it as more power unless you actually turn up the volume to unreasonable levels, think of it as more headroom.
Truth.

Some basics about power:
* Most speakers today are around 89dB/1w/1m; meaning they produce 89dB's of sound pressure with 1 watt measured at 1 meter. Just so you know, 89dB is pretty freakin' loud! Your Orsa's are normal at 88dB efficiency. 95dB's+ usually require horns or Lothars (although Audio Note and Von Schweikert both have dynamic speakers of 99dB+ sensitivity).

* It will require twice the power to affect a 3dB increase in output, which is hardly noticeable. Watts are overrated.

* Don't worry too much about how many watts your speakers are rated for. If they are clipping from too much power, turn it down. Trust us, you'll hear the clipping/compression. If you're going to overdrive something, let it be the speaker and not the amp. Usually, tweeters are the first to clip when overdriven, a clipping amp can eat your entire speaker.

* Focus on what kind of load the Orsa's present to an amplifier. Some speakers, planars in particular, are known to be amp killers because of the wicked impedence swings they can throw at amps. The Martin Logan Odyssey was particularly cruel IIRC, as it presented a <1ohm load at certain frequencies.

* Don't disregard synergy. Some amps will "lock in" with particular speakers yet be merely adequate with others.

* Finally, it's human nature to misinterpret increased volume as better sound. Therefore, when you are auditioning, bring a spl meter along and make sure you're comparing apples to apples.

Hope this helps.

Florian
03-26-2007, 10:58 AM
My speakers are known to be one of the most inefficent (which is total BS). Here is how it works.

My speakers are 86db efficency per 1watt measured at 4meters.

1watt=86db
2watt=89db
4watt=92db
8watt=95db
16watt=98db
32watt=101db
64watt=104db
128watt=107db
256watt=110db
512watt=113db
1024watt=116db
2048watt=119db


I use a 1200watt Krell amp on the bass, a 1200 watt Krell amp on the tweer, a 360 watt on the midrange and a 225 watt amp on the midbass. Its OVERKILL!!!! Absolutly useless power.

Thats why i sold my two 250 Krells.... my new Sphinx drives this speakers so easily with 105db peaks (which is damn loud) and only uses 60 watts on the entire array.

GMichael
03-26-2007, 11:04 AM
My speakers are known to be one of the most inefficent (which is total BS). Here is how it works.

My speakers are 86db efficency per 1watt measured at 4meters.

1watt=86db
2watt=89db
4watt=92db
8watt=95db
16watt=98db
32watt=101db
64watt=104db
128watt=107db
256watt=110db
512watt=113db
1024watt=116db
2048watt=119db


I use a 1200watt Krell amp on the bass, a 1200 watt Krell amp on the tweer, a 360 watt on the midrange and a 225 watt amp on the midbass. Its OVERKILL!!!! Absolutly useless power.

Thats why i sold my two 250 Krells.... my new Sphinx drives this speakers so easily with 105db peaks (which is damn loud) and only uses 60 watts on the entire array.

So, my receiver should be able to drive them to 105db, just before it puffs out some smoke, starts to melt down and then catches fire? Oh, that's if I can disable the protection circuits.
:idea:

Florian
03-26-2007, 12:00 PM
If it reallyhas that power then yes, we tried a 50 watt KR Tube amp and it got the Grand to
100db peaks. A perfect match if only used on the midrange. Mind you those figures are for the ENTIRE speaker but it only has to drive 1 ribbon. And since the impedance is even easier then on most commercial B&W and Paradigm speakers (from 4.2 to 5.1 ohm) and no crazy jumps its one of the easiest speakers around :)

GMichael
03-26-2007, 12:23 PM
The OEM claims 130 wpc rms at 8 ohms.

I've seen bench tests that show it at 125 wpc rms at 8 ohms. (for 2 chanels)
N/A at 4 ohms.
Complete meltdown at 2 ohms.
Bursts into flames at 1 ohm.

Carl Reid
03-26-2007, 06:20 PM
:confused5: I read or heard somewhere that many amplifiers do not have enough power to insure good sound quality.... that a speaker rated at less than 95 Sensitivity needed more power to avoid clipping...etc.

You read that on Musical Fidelity's website. It's their new controversial advertising campaign:

http://www.musicalfidelity.com/


My own Orsa speakers are rated at 88 db sensitivity...and my NAD C370 has a power rating that is progressive....120 wats and yet it can produce much more...but I don't know how that works. What I am trying to find out is whether an integrated amplifier rated at 250 Watts per channel into 8 ohms would do any harm to my Orsa speakers (88 db sensitivity) or if perhaps it might even improve my sound. The integrated amplifier I have in mind is the Musical Fidelity A-5....which has 250 watts per channel...etc. It also crosses my mind that more wattage in the amplifier might have a good effect on the overall quality of sound produced....anyone dare to tackle that? Bingo

Keep in mind though, that All Musical Fidelity's arguements rest upon one MAJOR assumption (which they freely admit) - which is that you want to have volume peaks of 105db!

Not everyone either wants or is able (due to neighbours and the police) to have concert hall level volumes in their own home.....

I love the sound of Musical Fidelity Gear (especially compared to NAD) but I'd suggest listening to the A5 with your speakers to see if the System Synergy works for you.... And seriously consider whether you really need as much power as 250 Watts... Since if you like the Musical Fidelity sound, you could easily swing the A3.5 (150 Watts) for much less money....

mlsstl
03-26-2007, 06:58 PM
You've got a lot of variables to consider besides raw power.

First take a look at the music you listen to and the volume you tend to listen at. If you listen to folk, choir, small group classical, etc. and moderate volumes, you may have little need for vast amounts of power. You may be interested in other aspects of an amp's tonal quality and how it interacts with your speakers.

If you're a real head-banger in terms of volume, you may gain more satisfaction with a more efficient pair of speakers versus a bigger amp.

Every person has a unique set of sonic cues they respond to that makes them think a particular stereo is better than another. You may respond more to dynamic impact and punch, another person may respond to impressive bass, and another to qualities in the midrange that strike that person as near-perfect voice reproduction. Since even the most expensive systems don't do everything perfect (in combination with a world full of erratic and imperfect recordings), each person needs to find the system balance that best suits them.

However, one notion I would like to dispel is the need for vast amounts of "peak" power. If you look at average recording levels (the average SPL you perceive in a room) recorded peaks generally aren't that much louder. If your recording is averaging -6 dB, you might see peaks of 6 to 9 dB above that. There is a natural limitation in the recording mechanism itself (whether digital or analog) that compresses or clips sound levels much above 0 to +3 dB. So, if you are using 10 watts at -6 dB (which would be very loud) the maximum a recording can ask your amp to deliver to the speaker is about 80 watts.

Live musical instruments can have much louder peaks, but they are difficult to record fully intact. You simply don't see many recordings with an average dB level at -12 or -20 dB or more to allow for somewhat unrestricted peaks. (Good analog tape has a range of about 60 dB and 16 bit digital a bit over 90 dB, but commercial recording practices typically don't set an extremely low average level just to allow for peaks. This is near unheard of in pop recordings and not much more common in classical.)

To summarize, it may be interesting to try a more powerful amp in your case. If you typically listen at very loud levels, you may find it worth it. If you listen at more average levels, I suspect there are other things you could put your money into other than just sheer power that'd give you more progress in your listening enjoyment.

Dusty Chalk
03-26-2007, 09:23 PM
It's not just for headbanging -- extra reserves in terms of power means that it doesn't distort, clip, or even compress when asked for particularly powerful transients (acoustic guitar plucks, piano chord hits, percussion, etc.). Extra power == prat, in my experience.

But the part about auditioning it in your own system is still good advice.

Bingo
03-27-2007, 01:50 AM
You read that on Musical Fidelity's website. It's their new controversial advertising campaign:

http://www.musicalfidelity.com/



Keep in mind though, that All Musical Fidelity's arguements rest upon one MAJOR assumption (which they freely admit) - which is that you want to have volume peaks of 105db!

Not everyone either wants or is able (due to neighbours and the police) to have concert hall level volumes in their own home.....

I love the sound of Musical Fidelity Gear (especially compared to NAD) but I'd suggest listening to the A5 with your speakers to see if the System Synergy works for you.... And seriously consider whether you really need as much power as 250 Watts... Since if you like the Musical Fidelity sound, you could easily swing the A3.5 (150 Watts) for much less money....
I think you have suggested something I had already arrived at before I read all of the great responses on this thread. I think I would go for the A 3.5 and save some bucks to put on the A-5 CD player...what do you think of those two as a combination? I think the A-5 CD player upsamples to 196 while their A-3.5 goes only to 96...if that means anything, does it? I would not like to get the 3.5 and later wish I had the 5....My present CD player, the Cary 308 upsamples to only 96...but I do wonder if the higher upsampling means better, clearer, cleaner sound?

mlsstl
03-27-2007, 05:07 AM
> xtra reserves in terms of power means that it doesn't distort, clip, or even compress when asked for particularly powerful transients...

That is why I went into the discussions about average recording levels and the recording headroom available. You may well have a microsecond burst of acoustic energy from a plucked guitar in a live setting that is +25 dB above your average recorded level of -6 dB, but your recording equipment is going to cap that peak at roughly 9 dB over that -6 dB level. So you are talking about needing power for a live situation that really doesn't exist in the recorded world unless you've got an unusually low average recorded level.

Dusty Chalk
03-27-2007, 07:52 AM
I think you have suggested something I had already arrived at before I read all of the great responses on this thread. I think I would go for the A 3.5 and save some bucks to put on the A-5 CD player...what do you think of those two as a combination? I think the A-5 CD player upsamples to 196 while their A-3.5 goes only to 96...if that means anything, does it? I would not like to get the 3.5 and later wish I had the 5....My present CD player, the Cary 308 upsamples to only 96...but I do wonder if the higher upsampling means better, clearer, cleaner sound?You're really getting into the realm of diminishing returns here. I am of the belief that 24/96 is good enough upsampling; I've also heard the opinion stated (by an electronics designer, I believe -- Mark Levinson, perhaps? I think?) that not to bother unless one is upsampling at least to 192. Probably the best assessment I've heard, though, is that I'd rather hear a well-designed non-upsampling DAC, then a poorly designed upsampling DAC, so really, the overall design is more important than whether or not it upsamples to 96 v. 192.

Now, me, personally -- I'd buy the A5 if I could afford it. I believe it has tubes in its output stage where the A3.5 does not.

basite
03-27-2007, 09:04 AM
IMO:
for a regular speaker, with a sensetivity of 89-90db/w/1m
too much power (like 3 or 4 times more than the speakers can handle at a constant rate) = total overkill, waste of power and money.

enough power (the same amount, or slightly more, max 2 times the power the speakers can handle at a constant rate) = enough, usefull, no need for more watts or current.

too less power (like 1/3 the power (or less) that the speakers could handle at a constant rate) = deadly, not advised in any way! turn up the volume a little too much and the amp will start to clip, and it could damage your speakers.

these were all for normal amps & speakers. when you don't need much more power, you should concentrate on the quality of the power.

Keep them spinning,
Bert.

hydroman
03-28-2007, 12:10 PM
^yep, i agree with Bert.

mlsstl
03-28-2007, 07:27 PM
Out of curiosity, I decided to check on the dynamic range of an LP record. I chose a Sheffield Recordings LP - "The King James Version" by Harry James and his band. This is a direct-to-disk recording so no tape equipment is involved. The music is unamplified big band jazz. Sheffield LP recordings are highly regarded in all technical aspects and pure analog.

Here are the dynamic ranges from the song "Cherokee" which has a lot of heavy duty drum action along with the brass and other instruments.

The background noise (phono off) is about -70 dB down. With the phono preamp on it is about 60 dB down. The silence in between tracks on the LP is roughly -45 dB to -40 dB. The average non-brass level of the music was -12 dB and with the horns cooking fairly well, that ups to about -9 dB. The highest typical peaks during the heavy drumming were -3 dB. There was a single peak of -1 dB with a particularly heavy drum knock. In looking at the waveform, that peak lasted about 1/100th of a second.

So, here you have an audiophile record with an average line level volume at -12 to -9 dB. The highest peak on the entire album is -1 dB. Therefore, if you were using 10 watts to play the lower of the two average volume levels (which is a LOT of average power), it take you 80 watts to deliver the -3 dB peaks and perhaps 120 watts to do the -1 dB peak.

Of course, that's be a horridly inefficient pair of speakers since 1 watt will get you in the 85 dB to 90 dB range in volume on a pair of average efficiency speakers.

As noted before, in this Sheffield example, the recording process simply wouldn't allow you to record a +10 or +20 dB peak on the record. The mike, it's preamp or the recording medium would naturally limit the peak (or just record it as overload distortion.)

While it is important that you have adequate power for your speakers and some reserve power above that is certainly desirable, the argument that you need a tremendous multiple of the average power demand gets stretched a bit thin. There are a lot of other reasons besides raw power that influence why an amp sounds good. You don't want to get myoptic about a single factor.

Bingo
03-29-2007, 06:16 AM
Out of curiosity, I decided to check on the dynamic range of an LP record. I chose a Sheffield Recordings LP - "The King James Version" by Harry James and his band. This is a direct-to-disk recording so no tape equipment is involved. The music is unamplified big band jazz. Sheffield LP recordings are highly regarded in all technical aspects and pure analog.

Here are the dynamic ranges from the song "Cherokee" which has a lot of heavy duty drum action along with the brass and other instruments.

The background noise (phono off) is about -70 dB down. With the phono preamp on it is about 60 dB down. The silence in between tracks on the LP is roughly -45 dB to -40 dB. The average non-brass level of the music was -12 dB and with the horns cooking fairly well, that ups to about -9 dB. The highest typical peaks during the heavy drumming were -3 dB. There was a single peak of -1 dB with a particularly heavy drum knock. In looking at the waveform, that peak lasted about 1/100th of a second.

So, here you have an audiophile record with an average line level volume at -12 to -9 dB. The highest peak on the entire album is -1 dB. Therefore, if you were using 10 watts to play the lower of the two average volume levels (which is a LOT of average power), it take you 80 watts to deliver the -3 dB peaks and perhaps 120 watts to do the -1 dB peak.

Of course, that's be a horridly inefficient pair of speakers since 1 watt will get you in the 85 dB to 90 dB range in volume on a pair of average efficiency speakers.

As noted before, in this Sheffield example, the recording process simply wouldn't allow you to record a +10 or +20 dB peak on the record. The mike, it's preamp or the recording medium would naturally limit the peak (or just record it as overload distortion.)

While it is important that you have adequate power for your speakers and some reserve power above that is certainly desirable, the argument that you need a tremendous multiple of the average power demand gets stretched a bit thin. There are a lot of other reasons besides raw power that influence why an amp sounds good. You don't want to get myoptic about a single factor.
Thanks Misst...I think it is misst..is it? I believe you are 150% right...there are so many other factors that have to do with producing good two channel stereo sound, that power alone is only a small part of the equation. While I did not fully follow all of the detail you gave me in your example of the Sheffield recording, I was able to get enought of it in my rusty brain to understand what you were explaining to me...I might be a C student, not an A student. The one thing I do know is that I am very very grateful for the time you took to explain the technical aspects of the quality of sound as it relates to amplifier power. As I digest all of the comments of all the guys who were kind enough to respond to me, I have once more examined my situation as it now exists with the NAD C370. And...believe it or not, I am not sure that I don't have the ideal amp in my set up for proper synergy...at least I know I have a pretty good set up. In time, someone may come up with a simple way to improve it even more. Bless you...Bingo

hermanv
03-30-2007, 12:17 PM
Subjective loudness is quite different than SPL. Poor sounding equipment sounds much louder at the same wattage as good equipment. This is because a lot of the clues the brain uses to arrive at a subjective loudness are concentrated in the those frequency ranges where poorly designed electronics produce a lot of extra energy often called glare or etch.

A local Mexican restaurant has a Mariachi band, 2 trumpets, I think I've read that trumpets can easily hit 115 dB or more especially at 10 feet away. Yes it's loud, but it's not even a little bit painfull, it's clean and undistorted. Real world music probably has a dynamic range of somewhere around 85dB. At home, if you set it loud enough to hear the lower level sounds you better have some headroom for those peaks,

In a similar vein, a good quality high power amplifier sounds more relaxed at normal levels. My Pass Labs runs class A upto near 25 watts before switching to AB. At my favorite listening level the little power supply load meter just starts to wiggle. With my 4 Ohm speakers the amp could produce 500 Watts. Seems like lots of reserve. My Radio Shack dB meter (slow and known to be inaccurate for peaks) says I'm running peaks of 95dB. So I'm probably pushing more like 100dB. I think my speakers are 87dB/watt SPL; so that calculates to only 25 watts. I used to own a 25 watt amp, there's no way it could even approach those same loudness levels without sounding severely constricted. Remember that 500 watts is only 13 dB more than my old 25 watts. (10 log 500/25)

My point is; It's not that simple. Use your ears, better some spare power than not enough.

mlsstl
03-30-2007, 01:41 PM
I don't think anything I said implies things are "simple" and I also don't think I said anything that indicated that we should all be using 17 watt per channel Dynaco SCA35's.

The first primary point I tried to make is that when it comes to dynamic range, recorded music and live music are two different animals. Many acoustic instruments have a wide dynamic range, but the question for a home stereo rig is what is in the recording?

Second is you certainly want some multiple of your typical power needs, but there is a point of diminishing returns. In your example, 500 watts may indeed be "only" 13 dB louder than 25 watts but 13 dB is a very significant volume difference. We're back to asking what could be on the recording that could demand that power swing? Also, in looking at my audio budget, if I spend the money to get a quality 500 watt amp, what might have I given up in another area that could mean more to my listening satisfaction.

Available power is simply one of many factors that get thrown into the equation when choosing an amp. A good example would be an amp that sounds good with well-behaved speakers but doesn't like "difficult" speaker loads. What happens when your most-favorite speakers in the whole world have an impedance curve from hell, back EMF and a host of other challenging behaviors? You may find that a lower power, but more tolerant amp sounds better than a less tolerant high power one. One might say there are certainly well behaved high power amps. True, but if they are out of the price range, or not available to audition, or any number of other issues, one must choose between what is available in that particular situation.

In other words, the issue becomes balance. One wants adequate power, but without chasing an obsessive amount just for the sake of it that ends up reducing quality in another part of our system.

hermanv
03-31-2007, 05:39 AM
I don't think anything I said implies things are "simple" and I also don't think I said anything that indicated that we should all be using 17 watt per channel Dynaco SCA35's.

The first primary point I tried to make is that when it comes to dynamic range, recorded music and live music are two different animals. Many acoustic instruments have a wide dynamic range, but the question for a home stereo rig is what is in the recording?
There is no general answer; 16 bit audio has a 96 dB dynamic range, very few recordings use all of this, but some do, otherwise 24 bit wouldn't sound any different. One of the Stereophile test CDs has a track recorded at both 14 and 16 bits (84 db and 96 dB dynamic range) Try as I might, I couldn't hear any difference on my old all Denon system, it is fairly easy to hear the difference on my newer, far more expensive system..


Second is you certainly want some multiple of your typical power needs, but there is a point of diminishing returns. In your example, 500 watts may indeed be "only" 13 dB louder than 25 watts but 13 dB is a very significant volume difference. We're back to asking what could be on the recording that could demand that power swing? Also, in looking at my audio budget, if I spend the money to get a quality 500 watt amp, what might have I given up in another area that could mean more to my listening satisfaction.
It takes about 10 dB for most people to "perceive" a doubling of volume, the actual doubling is at 3 dB. The ear is highly non-linear. I also said I wasn't using all available power of the amp. My point was that the reserve power contributes to ease. Of course costs need to be considered and I never meant to imply that hundreds of watts should take presidence over general quality. (I paid $2,400 for my used amp, while this is real money, it is not a high price for equipment in this industry)


Available power is simply one of many factors that get thrown into the equation when choosing an amp. A good example would be an amp that sounds good with well-behaved speakers but doesn't like "difficult" speaker loads. What happens when your most-favorite speakers in the whole world have an impedance curve from hell, back EMF and a host of other challenging behaviors? You may find that a lower power, but more tolerant amp sounds better than a less tolerant high power one. One might say there are certainly well behaved high power amps. True, but if they are out of the price range, or not available to audition, or any number of other issues, one must choose between what is available in that particular situation.

In other words, the issue becomes balance. One wants adequate power, but without chasing an obsessive amount just for the sake of it that ends up reducing quality in another part of our system.
This is oversimpliified but, an ideal amp is a pure voltage source. A pure voltage source has an infinite damping factor and can supply infinte power to a load, obviously not obtainable in the real world. All else being equal, the higher the power rating of an amp the closer it approximates an ideal voltage source.

Certainly any one system has somewhat unique needs, room size and speaker efficiency for a start. I don't think we disagree as much as you might think. I learned the hard way ($$$) to buy with my ears not the spec sheet. The spec sheet is also important, but must take a back seat to actual listening.

Bingo
04-02-2007, 09:44 AM
See what I started? I ought to be ashamed of myself...

Bingo

hermanv
04-02-2007, 10:08 AM
If it was simple we wouldn't need these questions:)