Is the volume actually logarithmic or is it linear? Here is what I want to know...Until I get better speakers, everything but my mains can only handle 100W (mains 150W) and I am also getting a 770, so if the scale is actually logarithmic, I should only have to worry about going from +7db to +10db that would possible damage my speakers (3db is a doubling of power). Anyone have a good answer?

It somehow seems to me that you might have the wrong end of the 'rope.'

The actual SPL is what you are interested in and at what levels it can be produced so that your speakers don't distort - which will more than likely be coming from a clipping amp as you crank it up. So paradoxically enough the safest is to use an amp that is as big as you can reasonably get and the 770 would seem to fit that bill.

Basically all quality speakers should handle CLEAN (non square-wave, clipping) signals up to a 'run you out of the room' level in a decent sized room so that you are still hearing 'GOOD' sound. Bass units might 'bottom' out on some peaks but well-made bass drivers will 'recover' and rarely have permanent damage. (Clearly - one 'learns' what those peaks are 'carefully' and one avoids levels which produce them - but a good speaker should be 'forgiving' if you occasionally bottom it out 'reasonably')

But taking a usual Japanese amp - and cranking it up - will VERY often result in overloading and get you 'bad' signals which will come out as damn bad sound - and burned coils - from your speakers.

(Actually speaker engineers avoid giving 'watt' measurements for the 'durability' of speakers - and the only place where watts are used by professionals is in giving sensitivity measurements - so many dB at 1 watt, 1 kHz, 1 meter.)


[This message has been edited by Jason Kent (edited September 09, 2002).]

Actually engineers often give watt measurements for the individual drivers, esp. mids and tweeters as this is generally an easily discovered thermal limit. bass drivers have to contend with enclosure design and while they also have thermal limits their excursion limit usually comes into play.

Where it gets tricky is, as you say, once the drivers are integrated into a loudspeaker system.

Most tweeters can only dissipate a hand full of watts in a continuous tone test - 10 - 15 would not be unusual. So how can they survive? In music they only get a fraction of the signal and that signal typically peaks and dips, giving plenty of time to cool between peaks. the same is true of the other drivers, to a lesser extent.

To put it simply, I've NEVER blown a tweeter (or anything else) from having an oversized amplifier.

Match the loudspeaker to the room, feed it from a decent amp, all should be well.

I've had very nice amps from the oreint, North America and Europe that gave no trouble, and also not so nice stuff from most of those places too. Country of origin is no assurance of quality.

I guess you could calculate the 100 watt point if you're using a 770, knowing that it is a 28dB voltage gain amplifier, and if you knew the impedance of your speakers and the voltage of your source. (Is'nt line level about 1.7 volts max? I don't remember the exact number but I think it's somewhere around there.)

------------------
Matthew J. Hill
matt@idsi.net

It would still be rough, since most speakers are impedence variable with frequency, and amps are supposed to be voltage sources (mostly). So you'd have to know the exact impedence the speaker would display to the specific signal.....

Plus, I bet if you sent a 99 watt 17kHz sine wave into them you'd smoke the tweeter in no time.

The power ratings on speakers are pretty much meaningless - what I'd like to see would be a set of curves, the decibels output at thermal limit vs. frequency and the decibels output at physical limit vs. frequency. Maybe a 'music duty cycle' (1 second on, 4 off or???) decibels output at thermal limit vs. frequency would be good too.

But that would be too much truth in advertizing I bet.

I agree that power ratings of speakers are sometimes meaningless but I don't know if we'd be helped by having power ratings of simple sine waves, or not. Somehow that doesn't seem like a real world solution. Most music is not a simple sine wave at a single frequency. Music is so much more complex than that, and speakers in the real world need to play music pleasingly.

If a speaker that plays music well even when played loudly, were to smoke its tweeter if only one watt of a 17k Hz sine wave played through it (with a 'music duty cycle'), would that be meaningful information?

[This message has been edited by Will (edited September 09, 2002).]

All audio waveforms are composed of sine waves. If a 1 watt sine wave @ 17kHz would blow the tweeter then either the speaker is very efficient or it cannot play very loud.

This is why I would like to see decibels rather than watts - watts are not what we hear. If a 17 kHz tone @ 90 decibels would blow the tweeter (similar to a typical driver in above scenario) then the speaker is very fragile and unlikely to be able to reproduce music well.

Make sense?