pre-school or anything.
I am! - I’m a perennial student and am a card-carrying member for life!

You’re right as far as SH talking about his speaker’s HIGH sensitivity. I do recall the almost double the power rule of thumb for 3db more @ etc. (Not watching my details on this but you get my drift).

What he caused me to contemplate is (since I hate math really a bother trying to deal with the science behind much in AV) It ‘sounds right’ to me that drawing from a amp near its idle point could create ‘grunge’. Don’t grasp the ‘why’ behind it…just makes sense. (Guess I’m digging for the why explanation behind this also…why does running at low-draw on high-current interject more noise).

So allowing for that how do you match power to your sensitivity level (speakers specs) and in a split HT/music system where you have 7 installed and the potential for a large request of power for short transients set up the system so that your not running the high-current large reserves of power at idle all the time for the majority of your listening (interjecting ‘grunge’ on your average listening). Yet protect your system from being underpowered lowering the clip threshold dangerously low? I can see if your goal is to run your amp near say 80% draw most of the time …where this would be much easier to figure out for a 2-channel system.

I assumed that often systems are set up for 2-channel with mono-blocks etc with a 5-channel added later due to the way consumers purchase….their separate system is begun with a 2-channel amp so they just purchase minimum required channels as they add speakers for a Surround setup. SH’s statement sends me to think if you expect less draw out 2-channel and try to more closely match your power range to the front channels. (so that your not running in grunge range when listening to 2-channel) and place a 5-channel on the rest with larger reserves for the sometimes more demanding draw of HT. That’s a plus for using a multi-amp configuration Vs the convenience of running one 7-channel .
Although I see many 2-channel guys putting some hefty power on the fronts so that adding a multi-channel setup will not draw too much away from these if they don’t keep the systems separate.

I’m trying to match my recent formed conclusions regarding power (you can’t have too much - and underpowered is always more dangerous than overpowered for speaker life) and my recent listening experiences when I ran into mediocre sound more often when the amps run were barely adequate to the system. – and include SH's intriguing comment about not interjecting grunge by having amplification that is (on average) running at a low percent draw rarely tapping huge reserves.

Lena:

The effect I was talking about is really about the efficiency of the speaker verses the power output of an amplifier. Picture if you will an amplifier that is putting out a constant level of hiss. Place speaker "A" on it with an efficiency of 86db/w, typical of a speaker with low to medium efficiency. Lets say that, just for sake of comparason it produces, with your ear next to the tweeter, 50db SPL of hiss. Now place speaker "B" on the same amplifier, not changing anything else except that speaker "B" has an efficiency of 96db/w, which is not unheard of in HT speakers. If you were to put your ear to the tweeter as before, the noise would measure 60db SPL : a 10 db increase, or subjectively twice the level of hiss. Now hang an efficient horn speaker on this amplifier with a sensitivity of 106db/w, as mine are. The hiss you would measure would be 70db SPL - twice again as much!.

Now, every solid state amplifier creates a certain amount of electronic noise (hiss), distortion, and other undesirable artifacts such as crossover distortion. These are usually a small percentage of their total power output . Some of these get worse as the power is increased (distortion), some are worse at the very lowest levels of an amplifier's output. Crossover distortion and hiss are the ones that are worst at the lowest levels.

We don't want an amplifier to operate near either of those extremes: full power and beyond with it's increased distortion, and clipping of it's output, and at the lowest end, with it's noise and crossover distortion. Generally, we want the amplifier to operate most of the time somewhere in the middle, well away from the detrimental effects at either extreme. It's that lowest end of the amplifier's output that I am concerned with here.

Now, power the speakers in the first paragraph with a 200 watt amplifier. Which speaker do you think will be using the least of this power? If you chose the 106db/w one, you were right. If you put that 200 watt amplifier on the 106db/w horn speaker, the amplifier would be asked to deliver a tiny fraction of it's available power almost all the time. In other words, the amplifier would be operating at at the lowest extreme of the power amplifier's output where the noise and crossover distortion is at it's worst.


Now, if you put a 50 watt amplifier on that horn speaker with it's 106db/w efficiency, you would see that relatively speaking, more of that 50 watt amplifier's power is used. In other words, the amplifier is operating more in the middle area of the amplifiers total output. It is farther away from the lowest extreme where the noise and crossover distortion is at it's worst. And since the speaker is very efficient, it is way less likely to need the full power output of the 50 watt amplifier.

Now if you put the 86db/w speaker on the 200 watt amplifier, you can see that more power would be asked of the amplifier most of the time and as a result, it would be operating more in the middle area of the amplifier's total output. And since the speaker is less efficient, it would be more likely to need 200 watts in order to avoid clipping. If you used the 50 watt amplifier with this speaker, you would likely run into clipping.

In the days of 2 channel (back when dinosaurs roamed the land, and soundhound was a young lad with a cow lick in his hair) almost all speakers were of low efficiency. Those speakers required amplifiers with large amounts of output power in order to drive them effectively, so that the amplifier would not clip when relatively high sound pressure levels were asked for (like when cranking Led Zeppelin ) In those days, a big amplifier was better than a small one.

But something happened during the morphing of 2 channel stereo into Home Theatre. It was found that speakers needed to have higher efficiency in order to deal with the loud explosions, gunfire, and car chases (and that's just in romantic movies - blockbusters are worse). It was also necessary to raise efficiency of the speakers so that multi-channel amplifiers could be built that had practical power outputs (a 7 X 1000 watt amplifier would be impractical!) However, the "bigger is better" mentality remained. To paraphraise Porgy and Bess: that ain't necessairly so.

Another consideration is that in a typical home theater setup, the really low bass is re-directed to the subwoofer. That low bass is where the amplifier is asked to deliver it's most power. If the amplifier is relieved of this requirement by bass management, then the amplifier is going to be working again nearer that lower end of it's power range, relatively speaking.

As you can see, it is not as simple as you would think. There are a lot of variables to take into consideration. Generally speaking, if you have speakers that are more in the typical range of efficiency, in the 80's to mid 90's, you really need a more powerful amplifier. If you have really efficient speakers, in the high 90's and above, I would look at what I've said here, and re-consider your power requirements.

I hope this explains the situation better!

[This message has been edited by soundhound (edited January 24, 2003).]

Or put in yet another way, if I have 105db/1w speakers and a 30 watt amp, and I listen at 85 db with 115 db peaks my amp will be running between about 0.01 watt average with short trips to 10 watts. Pretty reasonable range.

In a second case, where I mated a set of 90db/1w speakers with a 200 watt amp, listening at the same SPL, the amp would average around (a bit under) 0.5 watt with peaks about 300 watts, or deep into clipping.

So in the second case, 200 watts isn't as loud, or 'big enough', as the first case. This is where (IMO) the 'bigger is better' amp generalization comes from, because it's almost always true.

In short, one size doesn't fit all at all. The majority of speakers I personally know of that fit most needs run anywhere from 85-95 db/1watt, or at least 10x less sensitive that the ones 'hound uses. His system, although almost certainly excellent, is also atypical.

In the more typical range I'd use anywhere from 100-1000 watts, which not coincidentally is where many amps land powerwise. So I guess I'm saying IMO you've done fine so far.

Oh yeah, one more thing - speakers SAY their sensitivity is in db/watt, but in reality its almost always db/volts referenced to 2 x the square root of 2 volts (~2.83).

[This message has been edited by charlie (edited January 24, 2003).]

As an aside, and in the intrests of confusing the matter more, class "A" single-ended amplifiers (they can be tube or solid state) do not exhibt crossover distortion at all. There is no division of the 'positive' and 'negative' portions of the audio signal as in a class "A/B" push pull amplifier, it's one unbroken wave. There's no low level crossover distortion to be had, so this type of amplifier is especially useful for very high efficiency speakers such as mine. I use a class "A" single ended triode tube amplifier for my high frequency horns: it puts out only five (!) watts, but I would never even need that much power, given the efficiency of the horns.

Then there's class "A" push-pull amplifiers, but I'm not going to confuse matters too much

[This message has been edited by soundhound (edited January 24, 2003).]

SH, you are hereby nominated to author “AV Electronics for Dummies”
What a crystal explanation even for me on a first read. Betweens Charlie’s math and your explaination.
I have calmed down from a knee-jerk, What? more power isn't always better?

I do forget (early days yet) to think backwards from speaker sensitivity. And how incredible the difference is in 10 dB increments of these ratings when calculating power required. And then there’s impedance and as you point out …class of amp.
So at 91 dB 4 Ohms on FL/FR, and 91 6-Ohms center and 90dB on small satellite surrounds. 200watts per channel are not going to leave me ever hovering in grunge territory.

Also believe it or not (I don’t) I end up being quizzed by fellow shoppers when I’m out, (including a lot of men which I find incredible as I thought they were all born with AV knowledge) Asking me questions when a sales person has walked off after they hear me talking to sales. (I always preface my answers with …”I only know enough to be dangerous” and…”I really just have bits of this down, -Get some professional help!”) But I have talked to 2 guys already about what amps they’d want to look at while in stores thinking of beefing up their old receivers or looking into separates etc (Don’t want to steer anyone wrong…but I very clearly tell them I’m a neophyte)
Its too easy at my stage to spout genralalities without the slightest grasp of all variables involved and I try not to do it.

Thanks - I guess I'm just used to explaining things to my wife

A few comments, questions (and equations) about dB output, speaker efficiency, amp power, noise and that grungy noise floor...

For the purpose generating further discussion I submit the following:
As SoundHound has very aptly described, he has incredibly efficient speakers and as a result can use a less powerful amplifier to achieve reference level dB SPL's (volume). As opposed to more typical audio systems using less sensitive speakers, SoundHound also points out that efficient speakers produce the wide dynamic range for music and movies with a far smaller range of amplifier power (in direct terms). I believe it is also true, however, that no matter how efficient the speaker, there is a proportional relationship between power and sound level (dB's), which has some very important implications.

As we are concerned about amplifiers operating in the extreme lowest and highest power ranges where distortion becomes problematic, the goal of matching the amplifier design and the speaker is critically important. Furthermore, as the sensitivity of the speakers increase, it appears that the total system becomes more susceptible to noise generated by the other components, signal interference and the like. For those who can not customize the amplifier to these high sensitivity speakers, it seem like a very delicate and potentially expensive road.

Signal Level (dB @ 1m) as defined in my text as:
dBm = 10 x log (Amp Power in watts x (10 to the exponent (Speaker Sensitivity /10) ) )

Thus, speakers with a sensitivity of 106dB/watt, at one watt amplifier power, will produce a signal level of 106dB. Because of the high sensitivity, it would take only 0.000795 watts to produce a signal level of 75 dB. I am confident that SoundHound has the right amp for the job, but WOW, these are some really small amp power numbers. A signal level of 65 dB would require only 0.00008 watts of amp power... and so on... So, I am curious; where is the grunge floor on the tube amp that can make this system sound great when the music gets soft, and movie is at a whisper?

I have B&W N803 mains, with a sensitivity of 90dB/watt. These are bi-amped with two channels of a OA Model 770 dedicated to each main speaker. . According to the formula, it takes 0.032 watts to produce a signal level of 75 dB. Although this seems low, it is 40 times more power than SoundHounds speakers require for the same 75dB signal level. I have no idea where my grunge floor is, as I do not have the diagnostic equipment, or the expertise to find it on my own. All I know is that I love what I hear; when the music is as delicate as a Chopin nocturne, or when Jimmy Page tears up the Stairway to Heaven.

Speaker preferences are what they are, and all of us can enjoy our own, but I am very interested in how the gunslingers view the interrelationship between Speaker Sensitivity, Amp Power, Amp Type, Noise and the Grunge.




[This message has been edited by AGAssarsson (edited January 26, 2003).]

I think I described the situation pretty well in my long winded post, but I'll try to 'splain further.

First of all, this whole exchange came about because of a post by Hoots, who also has very efficient speakers. Not as efficient as mine, but more efficient than 'normal'. I gave him my recommendation that he keep his lower power amp to drive them, as it probably has enough power for the job.

The post was really directed at those crazies like me who have really sensitive speakers: then Smart Little Lena (bless her heart) picked up on the post and a thread-fest ensued. If you had ever seen my setup, you would know that it is not even remotely typical of what a sane person would do, but that's just me

So, the bottom line is that it simply does not applly to most of everybody else out there: they have components that are matched pretty well, not to worry!

Now to answer your question on how much "grunge" is too much, well, the short answer is that ANY solid state push-pull amplifier on speakers as sensitive as mine will sound really bad. It's that noticable. The "grundge" floor is not a "constant" that you can measure, as I was specifically seperating out crossover distortion as a main culprit. Not the only one, but the most significant one in this situation. This is a dynamic function, not a static one: it is essentially being modulated by the audio signal.

In a typical solid state amplifier, there is a standing bias current that is flowing through the output stage at all times. If the signal stays within this amount of current, the output stage is operating essentially class "A" push-pull. At the point where the signal JUST TRANSITIONS from class "A" to class "A/B", one transistor is cutting off, and another one is taking over in the push-pull output stage (the other transistor has been conducting since just before the 'zero' transition). This is where that crossover distortion "grundge" resides. This class "A" window is exceedingly narrow in a typical solid state amplifier. Unfortunately, this also occurs right at a point where the program material is soft, and the distortion is not swamped by the signal in speakers with sensitivities as high as mine.

The more bias current that flows, the broader this class "A" area is, before it transitions from the 'positive' set of output transistors to the 'negative' set of output transistors. The catch is that the more bias current that flows, the more static heat is generated. High end stereo solid state amps tend to have a very broad transitional area. Therefore, they have to have huge heatsinks to dissipate the heat, and this costs money.

Anyway, normally this not as bad as it sounds, since this form of distortion is reduced by the negative feedback that all such amplifiers have. With speakers such as mine, it can't be reduced enough to not to be audible.

I don't know if you caught it in a post I had later in this thread, but I get around the problem entirely with the use of a single-ended triode tube amplifier for my high frequency horns, 1,200Hz and above. A single ended amplifier has a single output device, and operates in pure class "A" all the time. The micro-transitional effects of crossover distortion are avoided simply because there is no transition from one device to another. One output tube: very simple and basic. There is also no global negative feedback. Therefore a lot of potential problems are just side steped by the use of this amp for high frequencies. For the woofer, I use a 50 watt traditional class "A/B" tube ampilifier. This tube amplifier, in common with all tube amps, has a very broad transitional area between the time it is operating in class "A" push-pull and when it goes into class "A/B". Since these woofers are quite sensitive, I am not usually getting out of this class "A" area under normal listening conditions. For those occasions when more power is required, the crossover distortion is way below the signal level.

I hope this gives you the information you need!

[This message has been edited by soundhound (edited January 26, 2003).]

I gave him my recommendation that he keep his lower power amp to drive them, as it probably has enough power for the job.

And that’s what startled me. Due to the first of the two guys I mentioned recently asking if I felt added amperage would make a difference (worth the expenditure) to a purchase of a HK8000 which is rated around 110 watts x 5 channels. He had mentioned his speakers sensitivity (maybe Polk at 89). And I had give him a hearty yes, that I thought he would benefit and that I would 'hear' a difference in that sort of setup from adding watts and using the HK as a pre. Same scenario with the second guy who was looking at receivers and asked what I thought about the advantages of separates.

Now I’m prob in no danger of being asked these type questions by guys who own anything with sensitivity above 98 (as you point out it’s rarer than not in most setups) not knowing enough to match his speakers to his amps. Unless he inherited some gear, or just purchased used from someone and has never been into AV before. But I know in those 2 recent moments my brain was not prioritizing, - figure back from the speaker sensitivity (first). I DON’T LIKE being asked (too much pressure) I like talking about AV but not giving someone advice who appears to know less than I without enough experience to take me seriously when I state I’m clear on this fact (ex: you HAVE to purchase a 2-channel amp to use the HK8 as a 7.1 setup regardless that the box states it’s a 7.1 receiver)…but check into the parameters of, -some other subject were discussing because I’m vague on details….as I mentioned I do state empathathicly when I get in these boats, I am new to this and have fringe knowledge in many areas…..but I hate to think if the guy just got his Dad’s old set of speakers @ 106! And never paid attention before, that I might have neglected to factor this spec! I wanted clarification on how to figure it for the variables in each system.
And now I see (better) why tube amplification (single stage) is not even just a ‘type sound’ preference in a system like yours its practically a necessity to reduce interjecting added distortion.

Bless her heart. (I almost heard a sigh in that one SH ).

You poor thing…. You probably come here to get away from … (if you have one like me), whom is always asking…”But WHY does it work that way?”…….and run into…( 20 questions - SLL)


[This message has been edited by Smart Little Lena (edited January 26, 2003).]