The old "does my amp have enough power?" question revisited

Mr Wolf

Distinguished Member
Joined
Sep 23, 2016
Messages
3,034
Solutions
1
Reaction score
3,475
Points
1,085
Location
Wiltshire, UK
spl_chart.jpg
Based on comments I see on many threads, it's clear that many people still fear they may need more amplification power in their systems when they probably don't. I once suffered from this but got over it in two ways.

First I cranked up the volume to an uncomfortable level (a level at which it compromised my enjoyment). At the this level the system still played cleanly and the amp volume setting was typically between -5db and -15db, depending on the source. I guess this practical test is the ultimate test but...that wasn't enough for me. I remained envious of other amps' THX reference status badging and I needed to know that I could achieve the coveted "reference level" of 85db continuous and 105db peaks. I therefore tried to calculate my own system's SPL capabilities.

I read up on the subject and sought to estimate the SPL at my listening position of a single 90db sensitive speaker as my L/R mains have that rating. I learned that SPL rises 3db as input wattage doubles and falls 6db as distance from the speaker doubles - please correct me somebody if I'm wrong! Anyway, with this know-how I created this chart on a spreadsheet.

View attachment 1008708

My Yamaha RX-V3900 amp (rated at 140W - [email protected]%THD) was bench tested by www.soundandvision.com years ago at 190.4 and 62.9 watts per channel for 2 and 7 channels driven respectively ([email protected]%THD). As my listening position is almost exactly 4 metres from the speakers the two highlighted figures of 96db and 99db are the most useful ones for me.

Remember, the chart is for one speaker only. I learned that as the number of speakers doubles, you add 6db so with both the L/R main channels driven, 128 watts should give 105db (99+6).

So what about the 7 channels driven rating? Well, 7 channels plus subwoofer makes 8 speakers (and I'm sure my sub can hold its own against one amp channel) so I used the 64 watts line to calculate 114db (96+6+6+6). An over-simplification I know due to different distances, speaker sensitivities room gain etc. but this is just an estimate afterall.

Both these estimates are based continous power ratings, not peaks, and still exceed the 85db continuous reference level by some margin. In fact, the two L/R channels can hit 105db on their own. Continuously!

The bottom line?

I think that, provided you don't have woefully inefficient speakers and/or sit miles away from them, most quality AVRs out there have plenty enough power for home cinema applications. I think we're all influenced by how manufacturers always throw in a bit more power for the next model up in the range. This is just the marketing men trying to convince you to spend up to your maximum budget...yes, yes...a little bit more power...just to be sure.

Of course, all other things being equal, more power is always better than less but if you don't actually need it then it has little or no value to you in practice. Certainly, an extra 10 or 20 watts isn't going to make an appreciable difference to your system.

Just hoping this might help some people out there achieve peace of mind and maybe save you a few quid on your next amp.

Any thoughts, comments welcome...and please, please, please tell me if any of this theory is wrong.
 
Last edited:
Interesting, and you may be right, that all is well for your set-up.

The subwoofer does not count as an extra speaker, the opposite in fact. As it is an active subwoofer, with it's own amplifier, it actually helps the AVR out, as any audio below the speakers crossover point will be amplified by the subwoofer not the AVR.

What you are not taking into account is that many people are now running 9, 11, 13 or more speakers from their AVR's, and the power supplies within those AVR's haven't really got any bigger.

I mentioned in the other thread today, that I have VU meters on my power amps, and I can see how much they are working, which is a lot when listening to cd's and when watching movies with a very dynamic HD audio soundtrack.

For music I have an integrated amp with a power amp, and just use my main speakers, no subwoofer, and I can see that the VU meters are jumping around and peaking a lot.

However, when watching TV at around -15ish, which is usual for every day watching, the VU meters are hardly moving at all, just registering a signal and ambling along without being stressed.

There is this calculator here, that is not going to be exact, but does show some interesting figures, it doesn't take account of room reflections or the audio below the crossover point that is handled by the subwoofer.

Amplifier, Speaker & SPL Calculator - Geoff the Grey Geek

In my room, with speakers at 88db sensitivity, if I want to listen at -5, which is about how high I would go on a 'movie' night, at 4 metres away, I need 254 watts to give me the 20db of headroom. If I turn it down a couple of db then it drops to 160 watts. Up a couple of db and it goes to 402 watts.

So, I think it depends on what the listener wants to achieve. If you want to power multiple speakers, in a medium to large sized room, and have a truly dynamic, cinematic experience, then power amps will almost certainly help.

But if you only have a 5.1 / 7.1 set-up, in a small room, and you don't like things too loud, or you don't really watch anything with very dynamic soundtracks, then the AVR on its own may be fine.
 
I would add- don't trust the advertised WPC figures from AVR manufacturers.
Power amp specs are much more reliable however.
 
Use an SPL meter or your Mobile Phone plus a suitable App to give you a good indication of the SPL levels you are achieving at your 'in room' MLP (Main Listening Position).

Joe
 
Note the comment he makes about 'open air' - that makes a huge difference vs the usual enclosed space of a living/music/cinema room.
Yes. That's why I said 'it doesn't take account of room reflections'. ;)
 
The subwoofer does not count as an extra speaker, the opposite in fact. As it is an active subwoofer, with it's own amplifier, it actually helps the AVR out, as any audio below the speakers crossover point will be amplified by the subwoofer not the AVR.

What you are not taking into account is that many people are now running 9, 11, 13 or more speakers from their AVR's, and the power supplies within those AVR's haven't really got any bigger.
I was saying that the subwoofer counts as an extra speaker for my "all channels driven" total SPL calculation. The reason I did this is because it is powered to a level at least that of the other 7 speakers that are powered by the AVR. I really don't understand why you wouldn't count it.

I understand your power supply concern for additional Atmos channels, but I'm yet to be convinced it's an issue in practice. All channels would never be driven at once at full power even if the PSU could cope with it as the resultant SPL would be intolerably high. In fact our ears may just be the most important factor in determining how big the PSU's output actually needs to be, not the total speaker count.

That said, I'm still planning on using the multi-channel analog inputs on my trusty old Denon AVR-3802 to power the 4 Atmos channels when I eventually move up to 7.1.4. only because I won't be suckered in by the AVR marketing into paying the silly entry price for having 11 channels onboard when a 9 channel amp plus my old amp will have more headroom at half the price.

By the way, I should have mentioned in my original post that my current 7.1 system is in a 5.9m x 5.4m room.
 
I was saying that the subwoofer counts as an extra speaker for my "all channels driven" total SPL calculation. The reason I did this is because it is powered to a level at least that of the other 7 speakers that are powered by the AVR. I really don't understand why you wouldn't count it.

I understand your power supply concern for additional Atmos channels, but I'm yet to be convinced it's an issue in practice. All channels would never be driven at once at full power even if the PSU could cope with it as the resultant SPL would be intolerably high. In fact our ears may just be the most important factor in determining how big the PSU's output actually needs to be, not the total speaker count.

That said, I'm still planning on using the multi-channel analog inputs on my trusty old Denon AVR-3802 to power the 4 Atmos channels when I eventually move up to 7.1.4. only because I won't be suckered in by the AVR marketing into paying the silly entry price for having 11 channels onboard when a 9 channel amp plus my old amp will have more headroom at half the price.

By the way, I should have mentioned in my original post that my current 7.1 system is in a 5.9m x 5.4m room.


The AV receiver isn't powering the sub. The sub is active and has its own integral power supply and onboard amplification. Only passive speakers are driven and powered by the AV receiver.
 
The AV receiver isn't powering the sub. The sub is active and has its own integral power supply and onboard amplification. Only passive speakers are driven and powered by the AV receiver.
Very true indeed but I think we already knew that.

The sub in question has its own 750W (1500W Dynamic) Class D amp. This is why I conservatively added 6db to my total SPL calculation. 7 AVR 62.9 watt powered speakers + 1 active 750W subwoofer = 8 powered speakers in total. For 8 speakers vs. 1, you can increase SPL by 18db.
 
Very true indeed but I think we already knew that.

The sub in question has its own 750W (1500W Dynamic) Class D amp. This is why I conservatively added 6db to my total SPL calculation. 7 AVR 62.9 watt powered speakers + 1 active 750W subwoofer = 8 powered speakers in total. For 8 speakers vs. 1, you can increase SPL by 18db.

I guess the answer is within this topics title:

"does my amp have enough power?"

So what happens if you change your sub for one with less power, would your appraisal still stand? You're calculating the combined abilities as opposed to the ability of the amp regardless of the sub it is paired with.
 
I really don't understand why you wouldn't count it.
As has already been said, because it isn't drawing any power from the AVR.

I understand your power supply concern for additional Atmos channels, but I'm yet to be convinced it's an issue in practice. All channels would never be driven at once
It depends what content you listen to. I have listened to (watched) action films, where the peak audio of explosions, car chases, war battle scenes, loud music in clubs, gun fights, etc, is all kicking off at the same time, and the VU meters on my power amps are peaking like crazy.

If the amp has run out of power, those scenes may sound distorted. With additional headroom, all of the audio, including the atmospheric audio, sounds clearer and more detailed. Instead of it all sounding like a distorted explosive mess, I can make out detail, such as glass breaking and falling on the pavement, someone relaoding a gun, a scream, for example.

EDIT: Of course, you don't have to be convinced. But I bought a power amp to experiment with, was convinced, and bought another one ;)
 
Last edited by a moderator:
The sub operates at a different frequency, so it won't increase the volume of the frequencies the speakers operate at.
Also, the adding 6db per speaker rule would only apply if all the speakers were outputting the exact same signal.
 
The sub operates at a different frequency, so it won't increase the volume of the frequencies the speakers operate at.

No but by taking a load (LF) off the AVR it will allow it (the AVR) more headroom at those frequencies.
 
No but by taking a load (LF) off the AVR it will allow it (the AVR) more headroom at those frequencies.


Actually it would result in the frequencies the AVR is dealing with having more headroom as opposed to the frequencies being diverted. The AV receiver wouldn't be handling those lower frequencies whilst its bass management is engaged. The headroom associated with those frequencies would be dependant upon the subwoofer's capabilities and not the AVR's. This is of cause as long as you are using the AV receiver's bass management and crossovers to redirect lower frequencies to the sub? It is generally suggested that people do this, but not everyone does so.
 
Last edited:
Actually it would result in the frequencies the AVR is dealing with having more headroom as opposed to the frequencies being diverted. The AV receiver wouldn't be handling those lower frequencies whilst its bass management is engaged. The headroom associated with those frequencies would be dependant upon the subwoofers capabilities and not the AVR's. This is of cause as long as you are using the AV receiver's bass management and crossovers to redirect lower frequencies to the sub? It is generally suggested that people do this, but not everyone does so.

That's what I meant, should have been clearer. :)
 
As has already been said, because it isn't drawing any power from the AVR.


It depends what content you listen to. I have listened to (watched) action films, where the peak audio of explosions, car chases, war battle scenes, loud music in clubs, gun fights, etc, is all kicking off at the same time, and the VU meters on my power amps are peaking like crazy.

If the amp has run out of power, those scenes may sound distorted. With additional headroom, all of the audio, including the atmospheric audio, sounds clearer and more detailed. Instead of it all sounding like a distorted explosive mess, I can make out detail, such as glass breaking and falling on the pavement, someone relaoding a gun, a scream, for example.

EDIT: Of course, you don't have to be convinced. But I bought a power amp to experiment with, was convinced, and bought another one ;)
I can't belive what you're saying here but if you think adding a powered subwoofer doesn't count towards increasing a system's total SPL output then I happy to disagree with you.

I very much agree though that the most demanding scenes on a AVR's power supply are likely to be those long explosions that have every speaker "kicking off at the same time" - I'm thinking Sarah Connor's playground dream scene in Terminator 2.

Let's say you wanted such an explosion to peak at 105db (THX peak reference level) for the listener. Frankly this a silly high SPL level as most people suffer physical ear pain at 110db but let's assume it anyway.

One system has 7 speakers, another has 11 speakers. All speakers are identical in both systems. The same AVR total wattage is shared between all the speakers so the 11 speaker system's speakers all receive less power. The scene arrives. All speakers kick-off playing the full explosion so that the SPL reaches the listener at the same volume level - 105db.

Now here's the big question. Which system places the greatest demand on the AVR's power supply, the 7 speaker or the 11 speaker one?

Answer. The 7 speaker one. Why? Because every time you double the speaker count you halve the available wattage which decreases output by 3db. But each time you double speaker count (provided they not in the same location) you increase SPL by 6db. Taxing your amp by doubling speakers increases SPL by 3db for the same total fed wattage.

The thing is those power demanding scenes naturally tend to be bass heavy so a system is more likely to limited by being "under sub-woofered" than lacking AVR power. Maybe this why some (admittedly budget end) professionally designed home cinema rooms still use a regular AVR for amplification but have at least two massive 1000W ported subs in support.
 
@Mr Wolf I am struggling to understand your point. Is there some other way that you can explain it, that might help it to make more sense to me?

I don't understand why you think that an AVR will find it easier to power 7 speakers, than 11 speakers?

I think you are thinking somewhere along the lines that if you need 105 db at the main listening position, and you have 11 speakers, then you only need each speaker to produce 9.54db?

I don't believe it works like that. I went to the internet for some help and found this:

You only have to think about it for a few seconds: 60dB is 'normal conversation’. According to your theory, two normal conversations would generate 120dB, equivalent to a 'loud rock concert'. Three normal conversations would be 180dB, equivalent to a gunshot, and cause permanent hearing loss in a matter of minutes.
 
@Mr Wolf
I don't understand why you think that an AVR will find it easier to power 7 speakers, than 11 speakers?
.
For the AVR to achieve the same total SPL output, it will find it easier to have access to more speakers as it is simply a more energy efficient way to generating sound pressure (SPL). This is why those nut-jobs that do SPL competitions in their cars have a speaker in every available area.

Two normal conversations vs one increases SPL by only 6db, it doesn't double it. Double it again (to four) and you get another 6db. But only if these conversations are not co-located but are equal distant from the listener. If co-located, you only get a 3db rise. e.g. like if you stacked two bookshelf speakers on top of each other.

I'll do a test this evening to try and prove my theory. I have one of those eco-energy consumption mains plug tester things and a Radio Shack SPL meter. I'll sit at my listening position and play a test signal through 2 channels to say 80db and through 7 channels to reach 80db and see which uses the most power at the socket. Would that prove my point to you if the 7 channel one uses less power?
 
I'll do a test this evening to try and prove my theory. I have one of those eco-energy consumption mains plug tester things and a Radio Shack SPL meter. I'll sit at my listening position and play a test signal through 2 channels to say 80db and through 7 channels to reach 80db and see which uses the most power at the socket. Would that prove my point to you if the 7 channel one uses less power?
I also have one of those kill-a-watt meters, if you mean these:

https://www.amazon.co.uk/dp/B071NZ5...t=&hvlocphy=9045037&hvtargid=pla-425446954692

They are useful for measuring how much power your devices are using, but they are not fast enough to measure peaks, and of course, don't take into account the amplifier's ineffeiciency. That is that the power that is drawn from the wall socket, doesn't all make it to the speakers.

Another look at the internet finds this:

If you have 10 50dB loudspeakers, they will have a volume equivalent to one 60 dB loudspeaker.
2 loudspeakers with 50dB output are equivalent to one 53 dB speaker
 
— As an Amazon Associate, AVForums earns from qualifying purchases —
Yes, it's gadget just like that one.

It doesn't need to be fast as I'll be running a continuous signal (with my F1 watching ear defenders on!). It doesn't matter that it's not a peak signal. We trying to establish the relationship between speaker count, SPL and the demands on an AVR's power supply.

Amplifier inefficiency is also irrelevant for this test as my AVR has 7 identical amp channels. A lower consumption will reflect a lower demand on the PSU - that is an undisputable fact.

The only thing messing teh waters a little are the fact that my speakers are not identical, nor equal distnat to the MLP. The mains L/R are the most efficient speakers which tips lowest power consumption in favour of the 2 channel test but they are 4 metres away, whereas the less efficent sides are 3 metres away and the backs 2 metres away.

I think your internet findings are spot-on, but only for co-located sound sources. Not if they are in different location equal distant from the listener which would give 70db and 56db for the examples you mention.

Starting to get excited about my test now...

The sub will be off by the way to ensure a more level playing field. I guess white or pink noise will be the best test signal.
 
Two normal conversations vs one increases SPL by only 6db, it doesn't double it. Double it again (to four) and you get another 6db. But only if these conversations are not co-located but are equal distant from the listener. If co-located, you only get a 3db rise. e.g. like if you stacked two bookshelf speakers on top of each other.

Pretty sure you have this the wrong way round... Co locating will give 6, separate 3, at least with a sub that's the rule.
Will be interesting to see your results.
 
Pretty sure you have this the wrong way round... Co locating will give 6, separate 3, at least with a sub that's the rule.
Will be interesting to see your results.

Yes, I believe you are correct, it is 3 db for uncorrelated sound sources, there is a mathematical formula for calculating it. There is an online calculator here : Adding acoustic levels summing sound levels 10 combining addition summation sum decibel levels or SPL of up to ten incoherent sound sources audio logarithmic decibel scale identical summing 1/3 octave spl full octave sum sound pressure level noise sound pressure acoustic pressure volts - sengpielaudio Sengpiel Berlin
 
OK guys, the results are in and you may or may not be pleased that I proved my theory wrong - @Rambles - I hope you're reading this.

I'll try and be scientific here for the benefit of future scholars.

TEST EQUIPMENT

Amp -
My 8 year old 7 channel Yamaha RV-V3900 - rated power 140w [email protected]%THD (2 channels driven), pre-calibrated via YPAO at the MLP.

Speakers - All B&W 8ohm speakers with the following roles, sensitivities and distances to the main listening position (MLP).

Speaker Sensitivity Distance
Mains B&W 603 S3 90db 4m
Centre B&W LCR600 89db 4m
Sides B&W 685 88db 3m
Backs B&W 686 84db 2m

Signal source - Youtube test signals, tones and music streamed via a HDMI connected NowTV box.

Sound level meter - Radio Shack analog SPL meter (Model 33-4050), set to slow response/C weighting

Power consumption monitor - Efergy EMS-UK

Ear defenders - red ones. From Silverstone in 1987 (famous Mansell win!)

APPROACH
I plugged the amp into the energy monitor, sat in the MLP with the SPL meter held upright in front of my face, played various things and edged up the volume via my iPad controller App until the SPL reached 80db. Once this level was reached I took the average current power consumption reading in Watts from the energy monitor.

The subwoofer was kept off at all times although all the speakers were set to small and crossed at 80hz.

Readings were taken for both "2 channel Stereo" and "7 Channel stereo" playback settings.

RESULTS

Once turned on, the amp consumes about 88 watts, even with the volume completely off! [Note to self - don't forget to turn amp off when not in use)

Signal Power consumption Amp volume

White Noise - 2CH 148.8W -25.5
White noise - 7CH 158.5W -21.0
Pink noise - 2CH 115.4W -16.5
Pink noise - 7CH 120.3W -21.0
Bassy music - 2CH/7CH 120W (Sexy and I know It - LMFAO)

CONCLUSIONS
Playing in 2CH stereo required slightly less power to generate the same SPL. What was evident though was that the main speakers were dominant, being more sensitive and benefitting from being at the exact level as the SPL meter.

Overall, I concluded that it's a zero sum game. If all the speakers were the same and positioned at the same height I think the same power consumption would be achieved. All other things being equal, power consumed correlates to SPL output irrespective of the number of speakers.

Also, high SPL levels can be achieved with very little power. The maximum incremental power consumed to achieve sound level of 80db was only 70 watts (158-88). (Amp's maximum rated power consumption per sticker on the back says 500 watts.)

I therefore have no concerns about the extra demands that additional Atmos speakers will place on an amp's power supply to achieve a given SPL level at the MLP, provided they have at least average sensitivity.
 
Hi. Interesting experiment :)

Did you push it any higher than 80db, to imitate soundtrack peaks?

Also, it would be interesting to repeat the exercise with the speakers set to large, as the bass frequencies use a lot of power.

Have you seen the Harbeth power usage test video? I'll see if I can find it.

 
I have been playing around with the Geoff the Grey Geek's calculator, which I know has it's limitations, but it can at least do the basic maths.

Putting in some figures from your set-up, I can replicate a situation with no headroom and the required amp power is 20 watts. Which is about the average difference you were finding between power consumed when AVR was idle, and power consumed when you turned up the volume to get 80db at the MLP.

If I then add in 20 db of headroom, the required amplifier power increases from 20 watts to 1,986 watts :eek:

It's a bit crazy :D

I expect you don't want to turn the volume up to get 100db at the MLP! I wouldn't recommend it as you are more than likely going to damage your speakers / amps / ears. But an interesting number though.

I think it does come down to what content the consumer wants to watch and listen to and what type of cinematic experience they want in terms of dynamic peaks, and volume levels.

If I am watching standard terrestial TV at around -20, which I do often, the power amps are only barely registering a signal. If I put a film on, with an HD audio soundtrack, and turn the volume up to -5, it's a different story!
 
Last edited by a moderator:

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom