A Test. How much Voltage (power) do your speakers need?

[Bksabath]

Hum which of those would you use as reference?

Nuclear Tests to the 1812 Overture - YouTube

Tchaikovsky "1812 Overture" with 105mm Cannons 20101017 (2/2) - YouTube

Peter Tchaikovsky 1812 Overture (real cannons) - YouTube

Same again for Floyds
Original or American version of the same records
Or the digitaly and much compressed Pulse in 180 G or even worst the CD version?

If you use a record as reference could you please give us the catalog number?

[Pano]

I don't have catalog numbers, but here are a couple I looked at for dynamics.

• Frank Shipway conducting the London Symphony Orchestra
• Andrew Litten conducting Dallas Symphony Orchestra 1996 (Telarc, I think).

[a.wayne]

Originally Posted by Bksabath

Hum which of those would you use as reference?

Nuclear Tests to the 1812 Overture - YouTube

Tchaikovsky "1812 Overture" with 105mm Cannons 20101017 (2/2) - YouTube

Peter Tchaikovsky 1812 Overture (real cannons) - YouTube

Same again for Floyds
Original or American version of the same records
Or the digitaly and much compressed Pulse in 180 G or even worst the CD version?

If you use a record as reference could you please give us the catalog number?

Use only the mercury recordings for the 1812 .. i have others they are compressed and horrible ..

[Pano]

Originally Posted by a.wayne

I used no.7 from that album I deliberately picked it due to it's hot avg and low dynamcis, the second recording is much better dynamically

Then you are deliberately skewing the test to make a point that does not have to be made. It has been addressed several times in this thread by me and others. If you listen to dynamic recordings, then use dynamic test tracks to set your level.

I will freely admit that I should have been more clear about using dynamic recordings in the first posts. My mistake. I will go back and edit that so as to be more clear.

On the other hand, what if someone listens only to dynamically compressed recordings? Should he go out of his way to pick a style of recording he never listens to just for the sake of the test? 5th Element has already addressed that in a recent post.

[5th element]

Originally Posted by a.wayne

My point ... they will need more than 25 watts unless they plan on listening to the same song forever ....

Yes but that was only an example, the user is supposed to do this test when using the most demanding music he or she has. Ergo, if it can replay 1812 without clipping then it will probably replay any other music without clipping too.

[Tom Danley]

Hi 5th
Not refusing, just not seeing the strong relationship to the title which seems to suggest it was about what is required to drive the speakers.
I am talking about reproducing the signal which is different in that it is also not subjective.
For example, you CAN’T hear instantaneous clipping as a discrete flaw it’s only audible when you can compare to not clipped / limited in that it is less dynamic sounding. Audible clipping / compression only happens well into gross non-linearity, well past where I am interested.

Part of the issue too is I have many different signal sources that I can pass to the system “volume control”, for me it is only adjustable gain knob, not a indicator of power or Voltage and doesn’t even have markings or readout.

I did the test, with 101dB 1w1M sensitivity speakers it takes about ½ VRMS mid band sine to reach a very decent loudness.

How much Voltage does it take to reproduce the variety of things I like to listen to in their dynamic entirety?

The answer to me seems that it depends on the dynamic peak to average ratio in the music and the dynamic nonlinearity of the speakers and not just 4X the .5VRms margin.
That margin IS plenty for a lot of recordings, especially modern pop but well short of others, way short of reproducing real life events in a normal noise floor.
My goal is to be able to fool people with sound, to make it seem like you are somewhere else, capturing / preserving dynamics is part of it.
Best,
Tom

[Mooly]

I still think is an interesting angle on all this,

Originally Posted by AndrewT

I propose to carry out a test for you all to consider.

I will turn the test back to front.
I normally listen sensibly with an average voltage to the speaker of 100mVac to 1Vac.

Average voltage, that's 0.637 vpk for a sine. So peak voltage here is what, about 1.56 volts. Average voltage is all we can realistically measure here when talking of a music waveform.

Originally Posted by AndrewT

If I turn these speakers up louder I can hear the sound becoming "loud" due to the increase in distortion. I can turn it louder yet, but it is not nice to listen to.

Distortion ? as viewed on a scope or, distortion due to room/accoustics or just too loud for comfort. It doesn't matter, you are the one setting the level that's right for you.

Originally Posted by AndrewT

............I can turn it louder yet, but it is not nice to listen to. The average level is now 5Vac to 6Vac. This is more than 12dB below the maximum amplifier output, i.e. the amp can push more than 22Vac into these 89dB/W speakers.

Average voltage again (6vac), so that's around 9.4 volts peak.

Originally Posted by AndrewT

The average level is now 5Vac to 6Vac. This is more than 12dB below the maximum amplifier output, i.e. the amp can push more than 22Vac into these 89dB/W speakers.

6 volts ac average level and and 22 volts (average or rms ?). That's 10.5 db to nearly 12 db as you say.

Originally Posted by AndrewT

That's the background.
The test.
What power of amplifier do you suggest I use for the test tone? 1W or 2W or 5W or 10W or 20W? I will build the amplifier to suit the majority view.

Then I will run the music at the level I think is nice but not distorted and give you both the average level and the test tone level. Then I can turn it up and measure the new average and test tone levels.

So 5 to 6 volts is not good to listen to for you. So lets be generous and say 7 volts average which is 11 volts peak. Rounding up that's 8 volts RMS

I vote Andrews amp should deliver 8 volts RMS into, ohh that's another can of worms... lets just call it a constant voltage amp

So I make that an 8 watt RMS amplifier.

Originally Posted by AndrewT

From there we can see what conclusions we can come to.

Interesting !

Are you thinking along the lines "that a 20 watt amp always sounds better than a 10 watter and that a 40 watt amp will be better than both and so on even if the 10 watt is never clipped ? Just wondered... if so I think that is a different issue.

--------------------------------------------------------------------------------------------------------------

Another angle to try and get this across (to all of you, this isn't aimed at anyone)...

Any amplifier... your amplifier that you listen too has gain. Yes ?

And if we say to set the volume to a given position, a position that is as loud as you can stand on the music you typically listen too, then we can say that the gain is Vout/Vin. Yes ! But we don't know Vin or Vout with music.

So we play Panos test track and measure the constant voltage at the speaker. We can then calculate the max voltage that your amp will ever put out on that volume setting you have just dialled in.

How can we know that for sure, and that it's realistic and reliable ?

It is because Panos track is at a known absolute level below the maximum your CD player can put out. Knowing how much voltage that track puts across the speaker allows us to say for sure what the voltage at 0db would be (0db being the the max level attainable off CD).

So we can say with absolute confidence what the maximum voltage could ever be and thus from that calculate the maximum RMS power needed to deliver that. We assume 8 ohms as a load although it's best to think in terms of the amp as a constant voltage source and ensure it can deliver that across your speaker.

Pano... think I'd have another week off if I were you

[5th element]

Okay, lets keep this simple for the time being. Where do you see the below comment as being described in any of the tests outlined here?

Originally Posted by Tom Danley

and not just 4X the .5VRms margin.

That is the 4x Vrms value as being some kind of 'margin'?

[MarcelvdG]

Originally Posted by jacco vermeulen

25.46V would normally correspond with 105dB (~19dB + 86dB/W/m), which is something the ESL63 is not capable of.
(so much for the oldy 303 and it's ~22V max )

I've actually cheated a bit by disconnecting the loudspeakers before doing the measurement, measuring with test CD's with -10 dBFS and 0 dBFS sine waves and calculating the equivalent Pano test levels.

25.46 V only corresponds to 102 dB because it is a peak value. Still, as an ESL-63 can handle 10 V RMS continuous and 40 V programme peak, a 0 dBFS sine wave could indeed have damaged it. For music there is still 4 dB of headroom to the 106 dB that an ESL-63 can produce (extrapolated to 1 m).

[Pano]

Originally Posted by Tom Danley

Not refusing, just not seeing the strong relationship to the title which seems to suggest it was about what is required to drive the speakers.

Well OK, maybe my title was too brief - maybe it should have been "How Much Voltage do you need to get your speakers to the loudest level you ever use in your room at your listening position with your music?" But that seemed a bit long.

In my defense, I did not title the thread "A Test. How much power will you ever need to reproduce the real level of fireworks, jet engines and cannon in a big room?" That is a fun and enjoyable goal, but not the goal of most music lovers with a home system.

I am talking about reproducing the signal which is different in that it is also not subjective.

Yes, we know. But almost all music listeners set their system volume subjectively. Are they wrong to do so? Should they buy an SPL meter and set the level to some specific SPL on a reference recording to comply with objective standards? Even if they don't like the results? I've tried that several times over the years and found it does not jibe with the levels I typically use. It's good to know what that level is, but I don't let it dictate my own volume settings.

Part of the issue too is I have many different signal sources that I can pass to the system “volume control”, for me it is only adjustable gain knob, not a indicator of power or Voltage and doesn't even have markings or readout.

Yes, absolutely. I've seen live level signals come out of a kick drum mic. You could run a mixing console in the red the whole time, or back it way down. That's not the same as the typical music lover sitting at home listening to CDs. He has a maximum level "carved in stone". From a typical CD player that maximum level will be 2V RMS. From portable players it is usually lower. Once the volume control on the amp is set, then the gain is set. The maximum output of the amp can never go above Vin*gain. How could it?

I did the test, with 101dB 1w1M sensitivity speakers it takes about ½ VRMS mid band sine to reach a very decent loudness.

Thanks very much for doing the test and reporting your results. Unfortunately, we don't know how your mid band sine relates to your peak value. Is it -12dB, -6dB, -25dB? If you were to set your volume control to a very loud setting - whatever that is, then measure a tone of known value (like -12dB) then we could compare it to the results that others have gotten.

The answer to me seems that it depends on the dynamic peak to average ratio in the music and the dynamic nonlinearity of the speakers and not just 4X the .5VRms margin.

Again, this says to me that you do not understand the test. Simply set your volume control to wherever it needs to be for you to get the level you desire - but it 92dB or 106dB, it doesn't matter. Once you have set that level, you figure out the maximum voltage your digital source and amplifier will deliver. Analog sources are not nearly so tidy, tho they can be measured.

You may interested to know that my main system is a pair of Altec A5 speakers with a passive crossover of my own design. I need very little voltage to play as loud as I ever care too. In a bigger space, I'm sure those levels will go up.