On power

[cequalspid]

I hope this thread is okay to start. Having recently bought (but not yet received) a Benchmark AHB2 I was wondering about the question, how much power do speakers generally need. Say from low volume (evening) to normal/loud (sunday chores around the house) but not room shaking levels.

Say for a pair of 3-way, bass reflex JBL L112 speakers:

maximum recommended power: 300, nominal impedance: 8 ohm, sensitivity: 89dB​

or say JM Reynaud Offrande speakers,

peak power: 260W, 4 ohm, 89dB​

or say Source Technologies 1.7 Excel (2-way monitor grade bookshelf flare vented - 3rd order) monitors
power min/max: 30W/300W, 8 ohm, 87dB

 

[RayDunzl]

With the above sensitivity numbers, as a general estimate...

1 watt or less - casual listening

10 watts - loud

100W - for very loud peaks

When you have 2 speakers (stereo) add 3 to 6db SPL from what you would calculate for one speaker.

Right now, watching TV, someone talking on the phone in the room;

A watt peak or so...

Maybe a 1/10th watt averaged.

 

[NTK]

Dr. Floyd Toole has written a series of articles on how to design a home theater that are available for free download. Here is the website to his book.
https://www.routledge.com/Sound-Rep...f-Loudspeakers-and/Toole/p/book/9781138921368

Click on the "Companion Website" (to the right of the Google Preview button). When a new tab opens up, click on the "Designing a Home Theatre" at the top, and you will find his 3 part article. Part 3 talks about amplifier power requirements in great detail.

 

[Speedskater]

Instructions for measuring power in your system. (rather involved)
https://www.diyaudio.com/forums/multi-way/204857-test-voltage-power-speakers.html

 

[headshake]

Get an SPL meter and you can have some idea of where you sit.

https://myhometheater.homestead.com/splcalculator.html

A spl calculator will let you see how much a watt can give you w/ your speakers.

This marketing is a nice read: https://www.crownaudio.com/how-much-amplifier-power

 

[Willem]

Those are quite efficient speakers. My Quad 2805s are some 83 dB, and so are my Harbeth P3ESRs, so I need rather more power, and all the more so for highly dynamic symphonic music in my quite large listening room. So a few years ago I upgraded from a 2x45 watt Quad 303 to a 2x140 watt Quad 606-2, and later an additional 400 watt B&W PV1d subwoofer. The upgrades gave a much cleaner sound with less compression. Even so, I sometimes hear that I am pushing the envelope. I could buy a pair of Quad QMP monoblocks, but my current idea is to high pass the Q606-2 with some Harrison Labs Fmods and add a second sub. My hunch is that this is not only cheaper but also more effective, if only because the main speakers have their limitations as well.
I saw a video of Harbeth's Alan Shaw demonstrating a pair of their M40.1s in Hilversum in the Netherlands. The power meters on the amplifiers showed that at times more than 500 watt per channel was needed.
The conclusion (also form the Crownaudio paper) is that with efficient speakers listening in near field in a small room you need almost nothing, but with less efficient speakers and dynamic music in a large room you need massive power. Insufficient power is one of the most effective ways to degrade the sound.

 

[Julf]

Insufficient power is one of the most effective ways to degrade the sound.

How does insufficient power degrade the sound?

 

[Julf]

how much power do speakers generally need.

Pretty much stated here:

sensitivity: 89dB

That means that to achieve 89 dBSPL at a distance of 1 m you need 1 W. Double the power for each additional 3 dB you need. Quadruple the powr for each doubling of distance.

 

[Willem]

Insufficient power is responsible for clipping distortion and dynamic compression.

 

[Julf]

Insufficient power is responsible for clipping distortion and dynamic compression.

Only if you play so loud you get clipping/compression.

 

[Willem]

Sure, by definition. The point is that the relation is not linear. So play dynamic symphonic repertoire at fairly realistic levels in a large room using quite inefficient speakers and you need a lot of power.

 

[Julf]

Sure, by definition. The point is that the relation is not linear. So play dynamic symphonic repertoire at fairly realistic levels in a large room using quite inefficient speakers and you need a lot of power.

No matter how dynamic the music, the peaks won't be any higher - it is just the average that will be lower.

 

[Willem]

Yes but if you play a loud average like a symphony and you have large peaks on top of that as you would with that repertoire you need massive power.

 

[Julf]

Yes but if you play a loud average like a symphony and you have large peaks on top of that as you would with that repertoire you need massive power.

Right - bringing us back to "if you need to go really loud, you need enough power".

 

[Speedskater]

Insufficient power is responsible for clipping distortion and dynamic compression.

If this refers to AC line power, then low line voltage will lower the max power of an amp with a linear power supply. But the audible difference between 100 Watts @ 1% THD and 90 Watts is trivial.
Now if tha amp has a SMPS then there will be no power differences.

 

[Willem]

And the good news is: watts are cheap.

 

[sergeauckland]

Insufficient power is responsible for clipping distortion and dynamic compression.

I understand how insufficient power can cause the amplifier to clip, but I see no dynamic compression, just increased distortion. Dynamic compression (and limiting) is not the same as clipping. Dynamic compression, which limiting is the extreme setting, causes no increase in distortion.

S.

 

[Julf]

And the good news is: watts are cheap.

Indeed, especially with modern class D designs.

 

[Julf]

I understand how insufficient power can cause the amplifier to clip, but I see no dynamic compression, just increased distortion. Dynamic compression (and limiting) is not the same as clipping. Dynamic compression, which limiting is the extreme setting, causes no increase in distortion.

S.

If the highest peaks get attenuated because the amp runs out of power or even clips, it is compression. Conversely, compression implies non-linearity, and non-linearity always causes distortion.

 

[sergeauckland]

If the highest peaks get attenuated because the amp runs out of power or even clips, it is compression. Conversely, compression implies non-linearity, and non-linearity always causes distortion.

Then you have a very different understanding of how a compressor works than I have.

All compressors have an attack and release time, so don't act instantaneously. Clipping is instantaneous, and acts on any signal that exceeds the maximum available voltage output. Clipping always increases distortion, compression and limiting doesn't.

S.