AVR power output S&V test bench - Would the results affect your decision on you next AVR purchase?

[gene]

RE: Audio Precision Test Equipment
Just for those that are unfamiliar with the company and it's products, AP revolutionized the audio test industry with their first product, which combined the electronics required to make measurements with a PC to crunch the numbers and provide output. Prior to that, all Distortion Analyzers and other related test gear were standalone devices. For output you recorded the values shown on the machine and presented them as you saw fit. The AP outputs a distinctive readout that can be saved or printed from the PC to a standard printer.

Because it is the Gold Standard of audio test equipment, it's not unheard of to print out a result made from some other device, or even some other DUT and presented as the device being marketed, and add an AP logo to the top right corner. (Note that the Sound & Vision chart does not use a genuine AP logo, but just the letters "AP"). Since every AP device offers excellent one-click printable output, you have to actually go out of your way to present it any other way.

I am not implying that other test equipment has no value (CLIO, which is less than one tenth the cost of the lowest priced AP machine, does excellent test measurements, for example) but just that it's a matter of trust and if you don't use an AP, why would you bother implying you do, and if you do use one, why would you not print out the readily available chart? It raises doubts about the authenticity of the results. Those doubts may be unfounded, but none the less they are there.

Trust is a critical component of any audio (or anything, really) review. The whole "Snake Oil" debate, or the "DBT debate", or even the "all amplifiers sound the same" debate, are perfect examples of how both audiophiles and the general public have a trust discord with audio specialty reviewers.

Wow I never realized this but always wondered how some of the smaller AV pubs got their hands on a genuine AP. I'm surprised a lawyer at AP hasn't noticed this yet and taken decisive action.

 

[SCG]

Maybe because I'm new the system wouldn't let me post multiple quotes with replies so attached is a PDF in that format

*EDIT - Ignore this post, after I hit 5 posts I was able to post the way I intended below

 

[SCG]

No way S&V did 7CH driven @ 4 ohms, nobody to my knowledge does that (including us) as it would demand too much power from the outlet and shut down virtually any multi-ch receiver. Most pubs, including us will test up to 2CH driven, 4 ohms on AV receivers.

I was wondering about that when I read your multi-channel reviews, I totally understand the laws of physics and not doing all 7 channels at 4 ohms

Many manufacturers like to tout power figures that exceed the capabilities of the wall outlet they are plugged into. It's important to note that not even Scotty from Star Trek could change the laws of physics, so don't expect an amplifier manufacturer to do so. A 120V/15A line max power output is 1800 watts while a 120V/20A line max power output is 2400 watts. The max efficiency one could expect from a Class AB amplifier is around 78%. So, factoring in efficiency, the available power from a 15A output would now be about 1400 watts and 1900 watts from a 20A line.

But with so much of the BS coming back around on inflated power into 8 ohms 7 channel AVRs :

Onkyo TX-SR383
Backed by 155W/ch (6 ohms, 1 kHz, 10% THD, 1 channel driven).
Instead of publishing unclipped full bandwidth power with two-channels driven into 8 ohm loads per FTC mandate, many of the major AV receiver manufacturers are now touting power with only one-channel driven, at 1kHz, into a 6-ohm load and 10% distortion. This type of testing scenario inflates the power rating up to almost 2X the former FTC way of rating power. Only when you search for the fine print on the manufacturer's websites or spec sheets do you find the two-channel continuous ratings, which in this case is 80 watts/ch.

Pioneer Elite VSX-LX103 You also get 170 watts of pure raw unadulterated power!
(6 Ohm - at 1 kHz - THD 10% - 1 channels)
It looks as though parent company Onkyo is rubbing off on Pioneer by artificially and erroneously inflating the power ratings just like the Onkyo TX-SR383. Further reading shows that the power rating for 2 channels is actually 80 watts (8 Ohm - 20 - 20000 kHz - THD 0.08%)! So now we have receiver companies more than doubling the power ratings of the unclipped full bandwidth power.

The new Model 7220 Amplifier is a class AB, fully balanced design. It delivers 220W per channel into an 8-ohm load with all seven channels driven and 330 watts per channel into 4 ohms (assuming sufficient available wall current) realistically with about 4-5 channels driven.

So why not test at least 4 channel at 4 ohms and hold them accountable if they publish 4 ohms specs and separate the chaff from the wheat and let the cream rise to the top? If it gives me at least 50% increase into 4 ohms with 4 channels driven (at continuous low distortion levels) then I know its got a kickass power supply and will take anything my powergrid and I throw at it.

And your just the man and this is just the site to set the Gold standard for multi-channel 4 ohm real world testing to keep them all honest!

 

[Dmantis10]

There are a lot of factors that help me decide on what AVR to purchase next. I usually only buy flagship AVR's so bench testing is less important to me unless I have 2 or 3 different models I'm comparing.
Having maximum power to me is always Ideal. I'd rather have more power and not need it then need it and not have it.
I find the best sounding AVR's are the most powerful due to the head room. Dynamic range to me is the most important thing.

 

[PENG]

No doubt the Denon AVR-5805 was the most powerful and best measuring AV receiver we've ever measured. It was an engineering marvel and elegant piece of AV gear for a much more civilized era

No way S&V did 7CH driven @ 4 ohms, nobody to my knowledge does that (including us) as it would demand too much power from the outlet and shut down virtually any multi-ch receiver. Most pubs, including us will test up to 2CH driven, 4 ohms on AV receivers.

I know, and I was wondering if that one and only 7 channel 4 ohm test was actually an 8 ohm test and S&V made a typo, but context in the paragraph quoted below did not indicate a typo. As I wanted to search for the most powerful AVR ever measured, other than the obvious candidate 5805, I bound to look at the RX-Z series, and accidentally found this one and only S&V test on the second from the top Yamaha flag ship. You reviewed the Z9 too but you did not measure the output.

Driving all seven channels into 8ohms, the RX-Z9 delivered 168Wpc at 20Hz and 165Wpc at 1kHz (to the nearest watt) before clipping (1% THD+noise). Into 4ohms, all seven channels operating, it delivered 221Wpc at 20Hz and 207Wpc at 1kHz. With only two channels operating, at 1kHz, the Yamaha clipped at 224Wpc into 8ohms and 382Wpc into 4ohms .
Read more at https://www.soundandvision.com/content/yamaha-rx-z9-av-receiver-measurements#htG0WspThSzISuEh.99

It was done in 2004. Based on the format, it must have been a S&V review, as Home Theater (owned by S&V) typically did theirs at both 0.1% and 1% THD, whereas S&V did it "at clipping" only.

 

[PENG]

And your just the man and this is just the site to set the Gold standard for multi-channel 4 ohm real world testing to keep them all honest!

Agreed, but only as far as this being the right place to set a standard, because of it's track record and the fact that Gene is an experienced real electrical engineer in the relevant field. I might have visited every website that published any sort of AVR test bench results and @gene of audioholics.com is the only one that do the full bandwidth measurements, preouts, and a few other important metrics that no one else does. There used to be another one, the AVtech of Miller Research that seemed to have stopped doing anything since 2014. Avtech did much more than S&V, but not close enough to AH's.

As for multi-channel real world testing, I believe a good compromise could be the 3 channel (LRC) driven into 4 ohm test. That would be for practical reason as illustrated by Gene in the case of something like the new and big 7 channel Outlaw amp. It is also reasonable because we all know the left, right and center channels on average do most of the heavy lifting anyway, especially the center channel. I am not saying we don't need to worry about the surround channels, but in the real world most people have smaller speakers that are not designed to take too much current anyway, even if they have some low impedance dips.

There are of course exceptions, but on balance, I think a combination of 7 channel into 8 ohms and 3 channel into 4 ohms, at 0.1% THD, full bandwidth, duration longer than a minute, would be a reasonable standard.

 

[Alex2507]

Agreed, but only as far as this being the right place to set a standard, because of it's track record and the fact that Gene is an experienced real electrical engineer in the relevant field. I might have visited every website that published any sort of AVR test bench results and @gene of audioholics.com is the only one that do the full bandwidth measurements, preouts, and a few other important metrics that no one else does. There used to be another one, the AVtech of Miller Research that seemed to have stopped doing anything since 2014. Avtech did much more than S&V, but not close enough to AH's.

As for multi-channel real world testing, I believe a good compromise could be the 3 channel (LRC) driven into 4 ohm test. That would be for practical reason as illustrated by Gene in the case of something like the new and big 7 channel Outlaw amp. It is also reasonable because we all know the left, right and center channels on average do most of the heavy lifting anyway, especially the center channel. I am not saying we don't need to worry about the surround channels, but in the real world most people have smaller speakers that are not designed to take too much current anyway, even if they have some low impedance dips.

There are of course exceptions, but on balance, I think a combination of 7 channel into 8 ohms and 3 channel into 4 ohms, at 0.1% THD, full bandwidth, duration longer than a minute, would be a reasonable standard.

My Aragon model 2007 is spec'ed as 200 watts into 3 channels simultaneously. I see why they might have done it that way now.

 

[PENG]

My Aragon model 2007 is spec'ed as 200 watts into 3 channels simultaneously. I see why they might have done it that way now.

I can tell you according to my own measurements of voltage and current (no scope) when playing various music and movies using two reasonably accurate and fast multimeter and clamp-on current meter, that while often overrated, the center channel does seem to have more current demanding contents that the left and right. People often overrated this apparent facts though, because I did find L+R draws more on average than the C. I am confident to say that the talks of the center channel account for 60-70% etc of the load is highly exaggerated.

 

[Steve81]

So why not test at least 4 channel at 4 ohms and hold them accountable if they publish 4 ohms specs and separate the chaff from the wheat and let the cream rise to the top? If it gives me at least 50% increase into 4 ohms with 4 channels driven (at continuous low distortion levels) then I know its got a kickass power supply and will take anything my powergrid and I throw at it.

I agree in principle that a simple 2ch driven measurement is not necessarily enough to determine a multi-channel amplifier's competence in dealing with 4 ohm loads. OTOH, I'd probably craft a 4 ohm multi-channel stress test a little differently: drive two (or three if you really want to push) channels to full power, with the balance being driven at 1/8th power (9dB down). IMHO, that would be sufficient to see whether or not the power supply was giving all it had in the base two channel test, while being a bit more grounded in what real world material tends to do, and better avoiding the limits of the wall outlet.

 

[SCG]

Agreed, but only as far as this being the right place to set a standard, because of it's track record and the fact that Gene is an experienced real electrical engineer in the relevant field.

There are of course exceptions, but on balance, I think a combination of 7 channel into 8 ohms and 3 channel into 4 ohms, at 0.1% THD, full bandwidth, duration longer than a minute, would be a reasonable standard.

I'd probably craft a 4 ohm multi-channel stress test a little differently: drive two (or three if you really want to push) channels to full power, with the balance being driven at 1/8th power (9dB down).

IMHO, that would be sufficient to see whether or not the power supply was giving all it had in the base two channel test, while being a bit more grounded in what real world material tends to do, and better avoiding the limits of the wall outlet.

I understand these POVs and I believe this thinking is most in line with Gene's current line of testing multi-channel amps at 4 ohms - test for what most users will use/need.

In my opinion a Gold standard is testing what the amp is capable of max outputting at 4 ohms based on standard max line input - this reveals its design, component, build and efficiency standards.

For me the Gold standard in subwoofer testing is Josh at https://data-bass.com/systems.

Do I need (or will I ever need?) a Captivator 4000-ULF that outputs a 108db at 10hz? Maybe not, but it's nice to know it exists if I ever do need it, and its also nice to know that I can confidently recommend it to friends and associates (who have more money than time to research) based on a consistent reputable testing site.

And your just the man and this is just the site to set the Gold standard for multi-channel 4 ohm real world testing to keep them all honest!

So if a manufacturer publishes 4 ohm specs I would love to have them verified - meaning real world (bench tested, not manufacturers' published specs) Gold standard testing (what the amp's max output capability is at 4 ohms multi- channel driven with standard max line input).

as AudioHolics moniker states:

"Let our rigorous testing and reviews be your guidelines to A/V equipment – not marketing slogans"

Just like Gene's reference speakers and reference amps:

XPR-1 Mono-Block Power Amplifier You’re probably wondering, “Who would need that much power?” Well, quite frankly most people don’t. I utilized my $50k Status Acoustics 8T speakers being a tough load dipping down to 2 ohms in the bass frequencies, these speakers can often be quite a problem for amplifiers not up to the challenge.

With the XPR-1, Emotiva has answered my calling to all manufacturers to build an ultra-high power, efficient and affordable mono-block amp that is robust enough to properly drive even the largest and most power hungry speaker systems on the market.

Unlike some other audiophile sites, Gene tests and recommends such down to earth (affordable) components that when he does publish gear in the upper stratosphere (though I may never need it or use them - $50K speakers) it's nice to know they exists and that someone reputably has thoroughly put them to the test - and I feel the same way about 4 ohm capable multi-channel amps, especially with some of the current BS going on with mainstream AVR manufacturers and their fine print 8 ohm power specs.

Though I suspect that most people including Gene would agree more with the quoted line of thinking above than mine - but one can dream and the great thing about this forum is we can all post our informed thoughts and opinions.

 

[Steve81]

Do I need (or will I ever need?) a Captivator 4000-ULF that outputs a 108db at 10hz? Maybe not, but it's nice to know it exists if I ever do need it, and its also nice to know that I can confidently recommend it to friends and associates (who have more money than time to research) based on a consistent reputable testing site.

Just tell your friend to reduce the channel count per chassis, and install multiple dedicated circuits to feed everything. If you're looking to maintain BIG power into low impedance loads across all channels, packing 7 channels into a single chassis is inherently a compromise (albeit one that makes sense for most of us).

 

[Alex2507]

I can tell you according to my own measurements of voltage and current (no scope) when playing various music and movies using two reasonably accurate and fast multimeter and clamp-on current meter, that while often overrated, the center channel does seem to have more current demanding contents that the left and right. People often overrated this apparent facts though, because I did find L+R draws more on average than the C. I am confident to say that the talks of the center channel account for 60-70% etc of the load is highly exaggerated.

My earlier comment only makes sense if you know it's a 7 channel amp. I think they wanted it known that they had the front 3 covered @ 8 Ohms. Their 4 Ohm number goes to 300 watts but doesn't mention the number of channels driven. It must still be three, now that I think about it. Now I'm curious what is left over for the other 4 channels if anything once it's red lined.

 

[PENG]

My earlier comment only makes sense if you know it's a 7 channel amp. I think they wanted it known that they had the front 3 covered @ 8 Ohms. Their 4 Ohm number goes to 300 watts but doesn't mention the number of channels driven. It must still be three, now that I think about it. Now I'm curious what is left over for the other 4 channels if anything once it's red lined.

If you do the math, not much left over assuming it is plugged into a 15 A outlet. That's for the so called "continuous" output measurements. For real world use, drawing more than 15 A, say 20 to even 30 A for some milliseconds from a dedicated 15 A outlet should not be a problem, though at such high current the voltage would dip a little momentarily and accordingly.

 

[gene]

I was wondering about that when I read your multi-channel reviews, I totally understand the laws of physics and not doing all 7 channels at 4 ohms



But with so much of the BS coming back around on inflated power into 8 ohms 7 channel AVRs :





So why not test at least 4 channel at 4 ohms and hold them accountable if they publish 4 ohms specs and separate the chaff from the wheat and let the cream rise to the top? If it gives me at least 50% increase into 4 ohms with 4 channels driven (at continuous low distortion levels) then I know its got a kickass power supply and will take anything my powergrid and I throw at it.

And your just the man and this is just the site to set the Gold standard for multi-channel 4 ohm real world testing to keep them all honest!

7CH driven into 8 ohms is enough to tax the available power in a 15A household outlet (220x7/.78% eff) = 1974 watts / 120 = 16.5A. If the wall outlet starts becoming the limiting factor in an amplifier test, it gets a little ridiculous to have to use a VARIAC to hold the line voltage up to do the tests. This is NOT how consumers use a product. I limit to 4 ohm 2 CH driven tests more to just show the actual per channel capability that is possible. These sets of tests give plenty of indication about the amplifiers ability to deliver power.

Outlaw isn't claiming 330 watts into 4 ohms all channels driven though they should be a bit more clear on that point that one power cord limits the ability for that rating to be ACD.

 

[3db]

I voted unsure depending on who did the testing and if the test results correlated with the reviewer's subjective findings. I dont believe S&V or Home Theater review were all that accurate in their testing methodology.
ACD tests also doesnt reflect real world situations and at best only tested the sensitivity of the AVR's protection circuitry.

mcode, Im really curious about the smart AVRs who can sense non real world application and therefore limit their power output. Do let us know what you find out. If the new Yamaha can dynamically alter the volume of the center channel depending on what the other channels are being sourced, then one doesn't need to make a leap of faith in an AVR's ability to determibe whether the load being driven is resistive or has a real impedance.

 

[gene]

My earlier comment only makes sense if you know it's a 7 channel amp. I think they wanted it known that they had the front 3 covered @ 8 Ohms. Their 4 Ohm number goes to 300 watts but doesn't mention the number of channels driven. It must still be three, now that I think about it. Now I'm curious what is left over for the other 4 channels if anything once it's red lined.

It's easy to do the math here. The Model 7220 is capable of delivering 1800 watts continuously and it has 20A fuses so it can likely up to about 2100 watts or so before blowing the fuse. Let's just assume 2kwatts to be safe. Divide that by 78% efficiency (max eff for Class AB under full load) and you're left with 1560 watts/330 or 4.7. This amp can easily deliver 330 watts/ch for up to 4CH driven into 4 ohms with a 20A line. If you're actually driving 7CH, the current limiting of the wall outlet would still limit the amp to deliver around 240 watts/ch with all 7 channels driven.

Again, this amp has a big enough power supply to use up the 20A line as indicated by the VA ratings of the transformers. There was no skimping here and the power claims appear to be legitimate.

 

[Alex2507]

It's easy to do the math here. The Model 7220 is capable of delivering 1800 watts continuously and it has 20A fuses so it can likely up to about 2100 watts or so before blowing the fuse. Let's just assume 2kwatts to be safe. Divide that by 78% efficiency (max eff for Class AB under full load) and you're left with 1560 watts/330 or 4.7. This amp can easily deliver 330 watts/ch for up to 4CH driven into 4 ohms with a 20A line. If you're actually driving 7CH, the current limiting of the wall outlet would still limit the amp to deliver around 240 watts/ch with all 7 channels driven.

Again, this amp has a big enough power supply to use up the 20A line as indicated by the VA ratings of the transformers. There was no skimping here and the power claims appear to be legitimate.

Thanks for the reply. Gene, The amp in question was an Aragon 2007 which PENG wanted to apply math to as well, but thanks anyway.

If you do the math, not much left over assuming it is plugged into a 15 A outlet. That's for the so called "continuous" output measurements. For real world use, drawing more than 15 A, say 20 to even 30 A for some milliseconds from a dedicated 15 A outlet should not be a problem, though at such high current the voltage would dip a little momentarily and accordingly.

After reading this I am ready to move heaven and hell for a couple of 20A circuits going to my 2 systems. I may have to put you on my ignore list until I get these ideas out of my head. You're a bad influence.

Setting my rec'r to have a max volume of 0 keeps me on this side of sane when I show off. It would go to like +10 or +15 but I'm sort of afraid of frying my speakers. You have to remember that I built the subs, replaced a blown tweet and rebuilt the crossovers on the ManTown system. One day I will grow up and just use the rec'r amps because of my low spl requirements. It's just more work than I'm interested in atm.

 

[SCG]

7CH driven into 8 ohms is enough to tax the available power in a 15A household outlet (220x7/.78% eff) = 1974 watts / 120 = 16.5A. If the wall outlet starts becoming the limiting factor in an amplifier test, it gets a little ridiculous to have to use a VARIAC to hold the line voltage up to do the tests. This is NOT how consumers use a product. I limit to 4 ohm 2 CH driven tests more to just show the actual per channel capability that is possible. These sets of tests give plenty of indication about the amplifiers ability to deliver power.

If I have time later I'll search for the exact review, but Audioholics did a review on a name brand multi-channel avr and it had a good review and testing results driving 8 ohm loads and the only the caveat was NOT to use it with 4 ohm speakers even though it had a 4 ohm rating because the unit became unstable and the only way to get it to pass testing was to flip the switch to the 4 ohms position which severely limited the power output of the avr.

 

[snakeeyes]

Is the amp review data to be used like a subwoofer bassaholic rating?

 

[Johnny2Bad]

Just to clarify, another aspect of the FTC rule is you cannot specify a power output into an impedance if the amplifier section cannot survive the pre-conditioning ( one-sixth power for one hour) without a fuse blowing or the protection circuit engaging. Note that the original 1974 rule specified one-third power; it was watered down to one-sixth about 20 years ago.

So, you have amps that might work with a 4-ohm load loudspeaker with music at moderate volumes but the spec sheet cannot indicate a power rating into 4 ohms so the buyer is left to wonder about the low impedance stability. The corollary is if there is a 4-ohm spec, the amp section must have survived the pre-conditioning (not usually a problem with switch mode supplies and class-D amp sections, but it's a difficult load - not impossible, just needs more money spent - for a conventional linear supply (transformer and caps) and a Class-AB amp.

Also care should be taken calculating output power and the draw from the AC socket. Although there is a relationship it's not simply a matter of using the power draw from the AC line in watts and correlating that to RMS power into an 8 ohm load. When an amplifier qualifies as requiring compliance (eg a two channel amp) with the FTC rule the testing lab (UL, CSA, etc) uses a regulated 120V 15A supply (because if the voltage sags or runs higher than 120V, power output will typically be affected even on a low power amp). It's entirely possible to get 600 WRMS x2 @ 20~20K 8 ohms at less than 0.1% THD (eg a Bryston 14B SST2) from such an outlet, yet the 14B is specified as drawing 950w from the outlet and that is what it will actually draw; in other words the relationship is not direct.

Note as well that the label specifying actual power draw in watts at x voltage from the outlet is required by law on every device (not just audio) that is designed to connect to AC power.

Gene's point is well taken, however. A Class-AB amp that suggests otherworldly output from a 15Z 120V circuit is suspicious.