Large vs. Small midrange

[David Gatti]

Simulation is one thing, Have you heard scanspeak drivers playing > 110dB at all? I have and its not pretty.

All depends on your crossover frequency. Keep cone excursion under control and they'll do fine, and better than most in a domestic situation.

I really wonder about guys who are listening at such extreme levels. Most of my listening is done at 0.1 - 1 watt (80dB average, 100dB peaks).

 

[Brett]

I really wonder about guys who are listening at such extreme levels. Most of my listening is done at 0.1 - 1 watt (80dB average, 100dB peaks).

And that sub 1W level is your average and Doug is talking about peak levels. I seriously doubt he listens at 110dB average.

 

[alspe]

/Offtopic/

I don't think that too much sensitivity is good because most of transistor amplifiers are not designed to be used at their lowest side of power. Volume knob is hard to handle and also many amplifiers makes more distortion when driven with almost zero watts.

 

[Brett]

Then you have the wrong amp and/or gain structure. I've had horns with 104dB+ sensitivity and driven them with all sorts of amps with none of these horrible effects.

 

[mondogenerator]

i think Alspe is referring to most hifi amps, class AB, and he has a point. Under a watt, they arent optimal but most run class A, over a watt the transistion to class B may have audible faults in some amps.
If u have 100dB/W speakers, then most people would scale the amp accordingly. Eg a 20watt class A S.S. Or valve amp would be all you need practically. After all peak watts is only root2 times rms, and there is seldom more alot more than 40db dynamic range in modern media. Ie. Eliminate amp clipping with more headroom, yes good plan, but if transients are THAT expanded then its likely the phono pre, cartridge or slew rate of cd output will clip too.

 

[Brett]

I know exactly what he was talking about and know the classes and specs of the amps I used.

But I agree on the point that you can scale to smaller class A amps if needed with very high efficiency, I just don't think it is needed most of the time.

 

[doug20]

All depends on your crossover frequency. Keep cone excursion under control and they'll do fine, and better than most in a domestic situation.

I really wonder about guys who are listening at such extreme levels. Most of my listening is done at 0.1 - 1 watt (80dB average, 100dB peaks).

I do not listen that loud at all. 85dB @ 1m, probably 75dB @ LP. Its all about 100% clean peaks though so even if my listening levels is .1 Watt the rule of thumb is 100x power for clean peaks.

If you listen even at 80dB you wil have 110dB peaks during movies in the 10Hz to 300Hz range.

 

[doug20]

/Offtopic/

I don't think that too much sensitivity is good because most of transistor amplifiers are not designed to be used at their lowest side of power. Volume knob is hard to handle and also many amplifiers makes more distortion when driven with almost zero watts.

I have 110dB sensitivity in my TAD Tweeters. Its all about gain structure and chooosing the right amp which isnt always easy.

Are you really talking about amp noise floor?? We can definitely agree about noise floor differences in many amps.

I do have issues with noise floor with all my high sensitivity designs. Its a pain to deal with, even with proper gain structure the "Hiss" is > 50dB. I found the ATI amps to have so far the best SNR for my designs. I also have a DBX SNR-1 that Im going to try out. I moved away from the DCX because it added about 6dB gain which increased the "HISS" by 6dB to my system, the MiniDSP adds zero.

 

[jb0194]

Simon,

Though, perhaps, less ideal than some ribbon tweeters, do you believe a larger dome mid-tweeter such as the old Scan 3806 (cognizant of its top end limitation), the Audax TW034 or Seas Exotic T35 may marry successfully with an 8 inch mid? Again, merely a satisfactory combination.

 

[Charles Darwin]

/Offtopic/

I don't think that too much sensitivity is good because most of transistor amplifiers are not designed to be used at their lowest side of power. Volume knob is hard to handle and also many amplifiers makes more distortion when driven with almost zero watts.

As anecdotal evidence I always found that any transistor system I had starts sounding really good when the amp delivered about 1Wrms regardless of SPL.
That goes for Klipschorns just as well as for those old british ones that produced 82dB/w (KEFs?) and everything in between.

 

[cotdt]

Back to the original topic. I've heard big 12" mids crossed high like 2kHz and they sound very grainy. Obviously a poor choice of crossover point, but there will be cone breakup past a certain frequency even if the frequency response seems reasonably smooth. In my general observations the smaller mids have better clarity and lack of grain. Larger midranges do throw a bigger soundstage, and the speakers seem to just disappear. I attribute this to the more pistonic behavior of smaller cones. The larger cones dissociate at higher frequencies, especially cones with softer material and lossier surrounds.

None of the pro systems I've heard at large concerts sound as good as a hifi speaker played at low output in a quiet home. I measure my normal listening levels and it is less than 1W. If I am hosting a party, I use up to 300W but my ears hurt and the police come within 20 minutes each time for disturbing the peace. I don't see any reason to go any louder than with a hifi speaker into pro speaker territory. That's just asking for more fines. I'm sure pro speakers will play clearer at high outputs but it's not that useful inside a home.

I agree with matching directivity. Due to room reflections, the total energy at all frequencies should be flat to sound right. That's why I'm so interested in the new 1.5" dome tweeters paired with 10" midranges. I would have to experiment with optimal crossover points.