Can a 80 watt receiver put out more than 80 watt?

[ski2xblack]

Close, but no cigar. What is the characteristic of a fully clipped signal, and what would its voltage level be relative to the peaks? (Hint: not 2x RMS.)

The characteristic of the signal would be the sine wave (we're not complexifying things with asymetric signals are we?) and a whole slew of odd harmonics that extend to high orders, resulting in a square wave with the flat tops determined by the supplies' rail voltage. The rms value of a square wave is the same as it's peak value. Practically speaking, I don't think many amps have supplies which can deliver twice their design voltage when driven into gross clipping, the rails would sag. But the power applied to the speaker would increase dramatically [power = (rms value)^2/resistance of the load...and in a fully clipped signal rms=peak, thus greater power than the unclipped signal].

Closer? I'm rather enjoying the brain calisthenics this beautiful Saturday morning, so thanks Irv!

 

[PENG]

As mentioned by others many times in the past, it is not very meaningful to compare power outputs if sound quality is important. Unfortunately, manufacturers of AVRs, and even audiophile class amplifier manufacturers put a lot more weight on "power" (based on their marketing info),than voltage and current.

Consider the following math (simplified, and based on 1 channel driven):

Amp A is rated 128 W into 8 ohms at 0.1% THD+N
Amp B is rated 32 Vrms, 6A rms

We can tell amp B is voltage limited to 32 Vrms, or 42.25 Vpk, and 5.656 Apk, so the maximum average power output rating for amp B driving an 8 ohm load = 32*32/8 = 128 W, but since is is rated 6 Arms we know when driving a lower impedance load, it could be rated 32 X 6 = 192 W into 5.3333 ohms, or 4X4X6 = 96 W into 4 ohms.

For amp A, we don't know what its rated output is, when driving a load with impedance lower than 8 ohms. Under the worst case scenario, the amp could be current limited to 4 A, then its output rating would drop to 4x4x4 = 64 W. It would have been much better if amp A is at least specified with rated output for both 8 ohm and 4 ohm load (most higher end integrated amp and power amps do), still not as good as specifying both max voltage and current, but will be good enough practically speaking. As it is now, we are left to rely on bench tests.

So in addition to M Code's $0.02 , i.e. "One needs to also ask @ "What THD level and for How long..." , I would add my own $0.02, i.e. One needs to ask, what is the maximum output voltage and current...

 

[AcuDefTechGuy]

So to answer the OP’s question, YES, an 80WPC amp or AVR can output more than 80 watts.

How much more power will the 80WPC amp be able to output? It could be 160 watts or twice the power. Some people say it could output even more than twice the rated power.

Is it safe to use an 80WPC AVR with speakers rated for 20-100 watts? YES, it is safe as long as you don’t listen to deafening loud volume, which applies to all amps.

2 other factors to consider are listening distance and sensitivity of your speakers. If your listening distance is farther than 18FT, you might need more power. If the sensitivity of your speakers is less than 85dB/2.83v/m, you might need more power.

 

[Irvrobinson]

The characteristic of the signal would be the sine wave (we're not complexifying things with asymetric signals are we?) and a whole slew of odd harmonics that extend to high orders, resulting in a square wave with the flat tops determined by the supplies' rail voltage. The rms value of a square wave is the same as it's peak value. Practically speaking, I don't think many amps have supplies which can deliver twice their design voltage when driven into gross clipping, the rails would sag. But the power applied to the speaker would increase dramatically [power = (rms value)^2/resistance of the load...and in a fully clipped signal rms=peak, thus greater power than the unclipped signal].

Closer? I'm rather enjoying the brain calisthenics this beautiful Saturday morning, so thanks Irv!

Bingo, assuming the clipping is so bad you actually approach a square wave. In the pro audio *design* world clipping is using defined as reaching 10% THD, which is a very weak square wave; it's still more of a sine wave. The pro audio user world seems to define RMS = 1x, "program" = 2x, and "peak" = 4x power, which is never true for Class A amps, seldom true for consumer Class AB amps, but appears to be roughly claimed sometimes for Class D amps. It's difficult to judge the efficacy of these claims for Class D amps without measurements, since Class D amps vary in topology more than Class A and Class AB amps do, so I suppose it's possible. And reading about the Pioneer example posted above, as with a lot of AVRs these days, it's a Class D design.

As M-Code mentioned, time is a critical factor for amp measurements, and especially so for clipping. The problem with clipping is that it will almost certainly occur at about 1KHz or below (and usually well below), and the harmonics occur at high frequencies, but at the power of the fundamentals, hence the potential of clipping to blow some tweeters. But even a cheap tweeter can handle clipping power for a few cycles, so to burn out a tweeter takes some rather extended clipping, which in my mind means you have to be pretty stupid to burn out a driver from clipping in consumer audio. This also lends efficacy to the often derided theory that too little power burns out more tweeters than too much power, but I'm not wading into that muck. ;-)

If you look at the measurements for a lot of Class AB amps, their THD curves hit 10% at only 1-2db above their actual maximum low-distortion power (rated power is often just a marketing number, and power is often under-rated for high quality amps, so I avoid it and look at the measurements, if available). I've often wondered why this is true, but I'm not going to speculate here. For Class D amps it seems every case is different.

 

[PENG]

If you look at the measurements for a lot of Class AB amps, their THD curves hit 10% at only 1-2db above their actual maximum low-distortion power (rated power is often just a marketing number, and power is often under-rated for high quality amps, so I avoid it and look at the measurements, if available). I've often wondered why this is true, but I'm not going to speculate here. For Class D amps it seems every case is different.

Just curious which site has class AB amps THD curves that show 10% THD at whatever output level above rated? I don't remember seeing any, anywhere..

I would like to add that if the THD is for say 100 W 0.1% THD average, then it is for 200 W peak if measured using sine wave test signal, THD will still 0.1%. I may be stating the obvious, but I believe some people may think that an amp 100 W average/rms rated and verified by measurement would clip badly if the peaks in the signal results in power output exceeding 100 W.

 

[Irvrobinson]

Just curious which site has class AB amps THD curves that show 10% THD at whatever output level above rated? I don't remember seeing any, anywhere..

I've been spending some spare time reading audiosciencereview.com, since my allergies are waking me up lately.

I would like to add that if the THD is for say 100 W 0.1% THD average, then it is for 200 W peak if measured using sine wave test signal, THD will still 0.1%. I may be stating the obvious, but I believe some people may think that an amp 100 W average/rms rated and verified by measurement would clip badly if the peaks in the signal results in power output exceeding 100 W.

I haven't doubted that math one bit. I'm only in this discussion because ADTG mentioned my question of how he concluded that clipping wouldn't occur until output exceeded 2x RMS power, and I thought that for Class AB amps that assertion wasn't supported by measurements. I've also mentioned in some previous posts that I tend to discount consumer amplifier output ratings for power amps, because they tend to be understated, for whatever reasons. Outside of bench testing and utter stupidity, I'm wondering if anyone really clips their amps in home audio scenarios.

Your concern that some people believe an amp will clip whenever you exceed the rated output reminds me of a silly situation that happened to me in 1999. We were selling a house, and it happened to be purchased by a former work colleague, so I knew the buyers pretty well. He had the house professionally inspected, and once the inspection was complete we all met in the kitchen to go over the inspection results with the inspector. In that house I had three 120v/20 amp isolated ground circuits with 10ga Romex added in the family room some years back, because I was using Krell Class A monoblocks that sometimes tripped the 15amp breaker on the single circuit that served that room. The inspector said he paid special attention to whether the added circuits met the building code, and they did. The wife in the buying couple asked why I added the circuits and why the outlets were orange. I started explaining how the amplifiers I chose at the time had huge transformer and capacitor in-rush current requirements, probably exceeding 50amps for a few seconds, and how they blew the 15amp breaker even when powered on sequentially. The inspector interrupted me as soon as he heard about in-rush current requirements, and exclaimed loudly and in an animated way that a 20 amp circuit can't possibly carry more than 20 amps. I opened my mouth to use the example of how lights often dim in the whole house when central AC comes on and why that is, but my spouse immediately gave me a look like "Are-you-really-going-to-argue-with-the-buyer's-inspector?", and I just closed my mouth and swallowed my pride. So I believe you that the notion of current limits varying with duration is not broadly understood.

 

[AcuDefTechGuy]

Just curious which site has class AB amps THD curves that show 10% THD at whatever output level above rated? I don't remember seeing any, anywhere..

I've been spending some spare time reading audiosciencereview.com...

Yeah, like PENG asked, did they show how much power was being output at 10% THD?

 

[Irvrobinson]

Yeah, like PENG asked, did they show how much power was being output at 10% THD?

The graphs show up to 100% distortion on the y-axis, but not all amps will allow output like that (most amps with soft clipping circuits) so some measurements only go to 3% (which is typical for consumer equipment test reports). Read audiosciencereview reports for yourself. For Stereophile reviews, which limited their graphs to 3% THD, it is easy to extrapolate output, since for a Class AB amp a reduction in the slope of the curve after 3% is very unlikely.

 

[AcuDefTechGuy]

The graphs show up to 100% distortion on the y-axis, but not all amps will allow output like that (most amps with soft clipping circuits) so some measurements only go to 3% (which is typical for consumer equipment test reports). Read audiosciencereview reports for yourself. For Stereophile reviews, which limited their graphs to 3% THD, it is easy to extrapolate output, since for a Class AB amp a reduction in the slope of the curve after 3% is very unlikely.

I was looking at the THD vs Watts for this PassLab 200W/8 ohm amp.

https://www.stereophile.com/content/pass-laboratories-xa2008-monoblock-power-amplifier-measurements

So at 0.3% THD it's 200W/8 ohms. At 3% THD, it's 300W/8 ohms. The output line shoots up almost vertically, so it doesn't look it it's going to even reach 400W (twice rated power) any time soon.

I assumed this 3% is clipping. But is it?

 

[Irvrobinson]

I was looking at the THD vs Watts for this PassLab 200W/8 ohm amp.

https://www.stereophile.com/content/pass-laboratories-xa2008-monoblock-power-amplifier-measurements

So at 0.3% THD it's 200W/8 ohms. At 3% THD, it's 300W/8 ohms. The output line shoots up almost vertically, so it doesn't look it it's going to even reach 400W (twice rated power) any time soon.

I assumed this 3% is clipping. But is it?

Many people have different definitions of clipping. Stereophile uses 1% THD as their threshold definition for clipping. Other designers and testers use 3%. The pro-audio industry (especially IC amp designers) often use 10%. My definition of the onset of clipping for Class A and ABs amplifiers is when the THD versus power curve begins to have an exponential increase. (The so-called "knee" in the curve.)

The Pass Labs amp you referenced is actually a little weird, in two ways. First, it is claimed to be a Class A, but as with most Pass Labs Class A amps the AC power input specs and the measurements point to them really being very high static bias level Class AB amps. Second, Nelson Pass, who designs all Pass Labs products, admits that he likes to "voice" his amplifiers by adjusting negative feedback in the various amplification stages in a power amp for more or less 2nd and 3rd harmonic distortion. That might mean that his amps could exhibit clipping behavior that's atypical compared to mainstream Class AB amplifiers. Note how different the curves look for the Benchmark Media amp:

https://www.stereophile.com/content/benchmark-media-systems-ahb2-power-amplifier-measurements

Personally, I think all of this discussion of clipping behavior is speculation anyway, as I highly doubt sane people clip amps in their homes.

 

[AcuDefTechGuy]

Many people have different definitions of clipping. Stereophile uses 1% THD as their threshold definition for clipping. Other designers and testers use 3%. The pro-audio industry (especially IC amp designers) often use 10%. My definition of the onset of clipping for Class A and ABs amplifiers is when the THD versus power curve begins to have an exponential increase. (The so-called "knee" in the curve.)

The Pass Labs amp you referenced is actually a little weird, in two ways. First, it is claimed to be a Class A, but as with most Pass Labs Class A amps the AC power input specs and the measurements point to them really being very high static bias level Class AB amps. Second, Nelson Pass, who designs all Pass Labs products, admits that he like to "voice" his amplifiers by adjusting negative feedback in the various amplification stages in a power amp for more or less 2nd and 3rd harmonic distortion. That might mean that his amps could exhibit clipping behavior that's atypical compared to mainstream Class AB amplifiers. Note how different the curves looks for the Benchmark Media amp:

https://www.stereophile.com/content/benchmark-media-systems-ahb2-power-amplifier-measurements

Personally, I think all of this discussion of clipping behavior is speculation anyway, as I highly doubt sane people clip amps in their homes.

Oh, wow, that Benchmark THD-vs-Watts line beyond 1% THD is like a vertical line. So even when this amp clips, it's not outputting much beyond the rated power.

Yeah, we both have used a 60WPC ATI amp to power the Revel Salon2.

I bet most people don't even need more than 50WPC for their home speakers.

 

[PENG]

Your concern that some people believe an amp will clip whenever you exceed the rated output

The point I tried to make earlier was more effectively made by the author of the article linked below:

http://www.acousticfrontiers.com/2013322spl-calculator/

He said:

"Many people also make a simple mistake which effectively doubles the size of amplifier required. Using the online calculators they enter follow a process of trial and error to determine the amplifier size required for 105dB SPL. The issue is that our THX requirement is not for 105dB continuous output but 105dB peak output. More on this later. "

I haven't doubted that math one bit. I'm only in this discussion because ADTG mentioned my question of how he concluded that clipping wouldn't occur until output exceeded 2x RMS power, and I thought that for Class AB amps that assertion wasn't supported by measurements.

Regarding ADTG's, it could simply be a matter of interpretation. Under the following assumptions, ADTG would be correct.

1. Define clipping as when THD exceeds a defined % level, example: 0.1%.
2. By RMS power, he meant average power the amp is rated for at the defined THD % level mentioned in 1.
3. By 2X RMS power, he meant the peak power that is 2 X the so called continuous RMS power.

If by 2X RMS power he meant whatever the amp is rated in RMS/average power output at the defined THD level considered, clipping wouldn't occur until the RMS/average output exceeded 2X the rated output, then I would disagree with him too.

Sorry I keep using /average, because I don't agree with the RMS power term, but I accept this common error seems fast becoming an accepted term. I am still waiting for that term to be adopted by EE,/physics text books though.

 

[ski2xblack]

Irv, don't be so sure that sane people never clip their amps. Based on just back of the napkin guesstimate, I suspect clipping occurs far more often than many people realize. Keep in mind that music has (generally speaking) a relatively high crest factor, with far lower average levels than a sine wave, but may have +20db or higher peak demands, which can rather easily exceed a lot of amp's capabilities. But the thing to remember is that the occasional clipping of transient peaks is typically inaudible, it's just too brief to notice. It also won't apply enough power for a long enough amount of time to damage a speaker. But the math, typical speaker sensitivity, and limited dynamic power of the typical amp (AVR) suggests that clipping happens frequently.

 

[Irvrobinson]

Sorry I keep using /average, because I don't agree with the RMS power term, but I accept this common error seems fast becoming an accepted term. I am still waiting for that term to be adopted by EE,/physics text books though.

You're just arguing for more precision in the discussion, and I'll always fully support that.

 

[Irvrobinson]

Irv, don't be so sure that sane people never clip their amps. Based on just back of the napkin guesstimate, I suspect clipping occurs far more often than many people realize. Keep in mind that music has (generally speaking) a relatively high crest factor, with far lower average levels than a sine wave, but may have +20db or higher peak demands, which can rather easily exceed a lot of amp's capabilities. But the thing to remember is that the occasional clipping of transient peaks is typically inaudible, it's just too brief to notice. It also won't apply enough power for a long enough amount of time to damage a speaker. But the math, typical speaker sensitivity, and limited dynamic power of the typical amp (AVR) suggests that clipping happens frequently.

I'm still a skeptic...

I have a large room (>8000 cuft) and speakers of just average sensitivity, and I can only hear (and see) the AT602 soft-clipping on some selected bass-heavy material at quite high volume (about 90db average at the listening seat, C-weighted). In more normal use I could live with that amp. Most people have more modest power needs, don't you think?

 

[mtrycrafts]

...

I would like to add that if the THD is for say 100 W 0.1% THD average, then it is for 200 W peak if measured using sine wave test signal, THD will still 0.1%. I may be stating the obvious, but I believe some people may think that an amp 100 W average/rms rated and verified by measurement would clip badly if the peaks in the signal results in power output exceeding 100 W.

People may think this but without seeing that sine wave on a scope over 100 watts and clipping at the peaks it is just incorrect thinking.
Power output is predicated on one's criteria for THD, not when the sine wave begins to flatten into a square wave.

I was looking at the THD vs Watts for this PassLab 200W/8 ohm amp.

https://www.stereophile.com/content/pass-laboratories-xa2008-monoblock-power-amplifier-measurements

So at 0.3% THD it's 200W/8 ohms. At 3% THD, it's 300W/8 ohms. The output line shoots up almost vertically, so it doesn't look it it's going to even reach 400W (twice rated power) any time soon.

I assumed this 3% is clipping. But is it?

All these THD & power graphs only shows power at certain THD, not the sine wave starting to flatten at the peaks. Perhaps you need 100% distortion for this to happen?

Don't think we know when the sine wave begins to flatten only that we don't want much distortion so certain definitions is made when clipping starts based on THD.

 

[Ernie Schmuntz]

Also depends on the vendor that made the receiver. I had a new Onkyo that could not do 100 watt without a lightning strike.

 

[PENG]

People may think this but without seeing that sine wave on a scope over 100 watts and clipping at the peaks it is just incorrect thinking.
Power output is predicated on one's criteria for THD, not when the sine wave begins to flatten into a square wave.

Agreed, that's why I tried to emphasize the need to define "clipping", expressed as a THD%. Without pre, or re-defining it for such discussion, clipping should simply mean as soon as the sine wave cannot reach it's theoretical peak, and that could be low enough that it cannot be seen visually on a scope, without looking at the displayed value.

 

[AcuDefTechGuy]

Agreed, that's why I tried to emphasize the need to define "clipping", expressed as a THD%. Without pre, or re-defining it for such discussion, clipping should simply mean as soon as the sine wave cannot reach it's theoretical peak, and that could be low enough that it cannot be seen visually on a scope, without looking at the displayed value.

Would you disagree with the statement that clipping generally causes power output that is 2 x Rated Average Power, but sometimes more and sometime less?

 

[PENG]

Would you disagree with the statement that clipping generally causes power output that is 2 x Rated Average Power, but sometimes more and sometime less?

Would you disagree with the statement that clipping generally causes power output that is 2 x Rated Average Power, but sometimes more and sometime less?

I tried to spell that out in my post#32, that it basically depends on the correct interpretation of what you are saying. Would you mind reading that post again and let me know which part I need to clarify further? Thank you for your patience.