SPL, watts, speaker efficiency and distance Excel calculator

[garyrc]

I've been following this thread because I wanted to see how many folks who have larger Heritage speakers feel the need to add external amps to AVRs to drive them. From what I understand is that they are actual 8 ohm rated speakers and don't have a jagged Frequency Response that dip much below 6 Ohms....

 

If that's true, what benefit does adding an external SS amp bring?

 

I think the Khorn dip might be about 6 Ohms, but I don't remember.  I do remember that the very early Klipschorns had an peak of 40 Ohms!

 

The power output of AVRs tend to be overrated by some manufacturers.  In the advertising and spec sheets they tend to be rated with 2 channels operating, which asks less of the AVR's power supply than 5, 7, or a greater number of channels running.  I think someone figured that the real power available to each of 5 channels averages about 80% of the published power.   If the AVR has not been rated with the same number of channels running as you will use, then only bench tests by someone (e.g., a magazine) will let you know what the real power available is. 

 

Now how much do you need?  I used to have a power meter on one of my amps running, and most of the time my Klipschorns (rated at between 98 dB and 105 dB, depending on who's doing it, and under what circumstances) needed 1 or 2 watts per channel, but that was most of the time, with most loud passages.  Once in a while, I remember, the meter would read up to 6.5 watts, and rarer still, about 12 watts.  Once, at the end of a Mahler symphony, the meter hit 25 watts.  That was in a small/medium size room (2,000 cu.ft.??).  If I doubled the room size (my current room), the power needs would approximately double.  So that brings us to 50 real watts needed per channel.  With modern movies, there are some very loud effects.  True, they go through the self powered subwoofer, if they are below the usual LPF for LFE of 120 Hz, but I've found myself panicking at a few effects that were way higher than 120 Hz -- metallic screeches, and the like.  Wouldn't want any clipping there!  So, I mentally put in a safety factor by doubling the wts needed, and came out at 100 w.p.c., more than many AVRs rated at 100 w.p.c. will put out, if 5 or more channels are running.  

 

Don Keele worked up a famous Dope from Hope (vol 16, #1, January 1977) that reported a need for 63 watts to produce 115 dB from one K.Horn in a 3,000 cu. ft. room.  Don thought that for two channels one would only need 1/2 the power to each channel, but that's not necessarily true, even with music, and, with modern movies, blasting away with all channels equally loud (the incident toward the beginning of The Grey), those days are over.  Don also thought that good amps would pass peaks 10 dB louder (10 times the power) without clipping, but I think those days are also over, if that ever was true.

 

PWK said something like, "To reproduce the blood stirring levels of a full symphony orchestra, you need 115 dB."  Once again, in our modern world, most of that would go through a powered subwoofer, but I don't know what frequency the impact sound of a cymbal would be, or what the SPL of someone pounding away at the keys higher than two or three octaves up during a piano concerto would be.  

Edited May 20, 2015 by garyrc

[mustang guy]

 

I think I've been misunderstood.

My thinking is as follows regarding one microwatt, or actually 60 dB down from maximum level.

If 16 bits are on we get the greatest voltage output of the playback unit.

1111111111111111 = 65536

so

-60 = 20 log (x/65536)

Solving for x we get 65.536

The value of 65 in binary is 0100001

Therefore a sine wave at -60 dB is using only 6 bits, and for the most part 5 bits. So it seems to me that the signal is going to sound like it has a bit depth of 6 or 5 bits, and most 5 bits. That seems pretty low. That is what I meant by sounding.

Am I doing something wrong here?

WMcD

 

Bits are digital reference with no regard for voltages, amps or watts.  The digital information is parsed real time by the digital to analog chip to provide an analog signal which can be processed by the analog stages of the preamp. I cannot think how gain which is stored digitally can be calculated like that unless you know that PCM stores its gain information for a particular tick in 6 bits of the word or string. I am going to look at parse PCM and see what the goog has to say.

[WMcD]

I could be wrong . . . but I don't think so.  As the song says.

 

WMcD

[mustang guy]

OK, I'm back...

 

It's confusing! There are so many PCM versions out there, but one thing they all have in common is mathematics of amplitude. I think I understand that the amplitude bits tell the DAC how loud or less loud to make the sample. The amplitude is logarithmic in some PCM derivatives and linear in others. I can see why a 16 bit sample is inferior to a 24 or even a 20 bit sample. I can't for the life of me parse a PCM even though I have read up and even analyzed code for a linear PCM. I know you need to know whether the stream of samples is little-endian, the sample rate, signed or unsigned, and finally the bitrate. Once you know that, you can start parsing. The sample tells the frequency and the amplitude. I am not sure if they are seperate parts of the sample, or an equation of the frequency and the amplitude. I assume your math above is considering that the latter is true.

 

Am I on the right track?

[Quiet_Hollow]

So it seems to me that the signal is going to sound like it has a bit depth of 6 or 5 bits, and most 5 bits. That seems pretty low. That is what I meant by sounding.

Strength of idea. Suffice to say, by that method you've got some things confused there. Edited May 22, 2015 by Quiet_Hollow

[mustang guy]

I do have things confused, but I did not say the things you quoted. That was William. 

[Quiet_Hollow]

Fixed. Thanks for the catch.  I can't imagine why the quote button picked up your name instead.