[James Tanner]

Hi James,

I have a question that bothers me for a long time .
Experience tells me that when you increase the max power of an amplifier from 50W to 300W on a dynamic full range speaker the sound will improve considerably. That was the main reason I upgraded my B60 to 4Bsst.

However, the numbers do not add up. Suppose we have an amplifier that outputs 1 watt to 8ohm, and a speaker which has a constant impedance of 8 ohm over all frequency ( I know it's unrealistic, but for the sake of argument).
If the speaker has efficiancy of 90db, it means that it will output 90db from 1w, to a distance of 1 meter. 90db is considered enough for most listening sessions.
According to this argument 10W is enough for every practical purpose. A 300W amplifier seems ridiculess.

Obviously, reality taught me differently . But I cannot find the error in the previous argument. Can you please explain to me why is this argument invalid?


Hi

   Thank you for your recent inquiry about amplifier power. The reason your system will sound better with more power is because of the large dynamic range of musical sources today.  Your example assumed a 10dB range between average power and peak power requirements, but the actual range is quite a bit larger, as much as 20dB and even more.  Naturally, an amplifier with 60 Watts available will reach overload on peaks of 20dB beyond an average power of 1W. If the ratio between average and peak power is 23dB, a 200W amplifier would theoretically be required to play music with a 1W average level.

   Keep in mind also that speakers' efficiency ratings are generally established at a 1M on axis.  If you are located at a more normal distance of 10-12 feet, the output will measure as much as 5-7dB less, requiring at least 2-5 times the stated power to produce the same 90dB level in the room.

   From the above examples, it's easy to see that an amplifier delivering 'only' 1W per channel of average power to a set of speakers in a normal room might easily need 300W of peak power to avoid overload distortion in many circumstances.

   I hope the above information is satisfactory, but please let me know if you have further questions.  Thanks for your support for Bryston audio components!

Sincerely,
Christopher W. Russell
CEO, Bryston Ltd.

[Phoenix]

That's why my speakers have an efficiency of 98 dB/W/m!
Amp: B60

[dan_lo]

I have two questions:
1. I'm not sure I understand your math . I usually do:

ratio in db = 20*log(P1/P0)

That's how we reach the number 96db for the maximum dynamic range of CD.
you multiply by 10 instead of 20. Is this the normal convention?

This is more than a technicality to me . If 23DB is the dynamic range and we use 20 as the multiplier, we reach 14W instead of 200W.  

2. Is the  amplification of the listening room itself  irrelevant here, and how much is it roughly?

[perose]

you multiply by 10 instead of 20. Is this the normal convention?

Yes, small signal db is multiplied by 20 while power db is multiplied by 10.

[James Tanner]

Hi Dan;

   Thanks for your follow-up inquiry.  When calculating dB, we use 20 log Vout/Vin.  Thus, an increase of 20dB means an increase of 10X the Voltage, but 100X the power. That's why 20dB above 1W is not possible for a 60W amplifier, you require a 100W amp.

   Again, I hope this is helpful, and thanks for your interest!

Sincerely,
Chris Russell
Bryston Ltd.

[dan_lo]

Got you !
Thanks again.