hi,
I know there are rules of thumb for 2- or 3-way speakers, but how do you know how much power a driver or the active amps (in active loudspeakers) really need?
Isn't there something like a graph of power over frequency for "standard" music
, or several music styles somewhere on the web?
Or volume in dB over Frequency, then I could (still would have to figure out) calculate the power needs...
Thanks for helping!
Cheers,
Dominique
I know there are rules of thumb for 2- or 3-way speakers, but how do you know how much power a driver or the active amps (in active loudspeakers) really need?
Isn't there something like a graph of power over frequency for "standard" music
, or several music styles somewhere on the web?Or volume in dB over Frequency, then I could (still would have to figure out) calculate the power needs...
Thanks for helping!
Cheers,
Dominique
You can find some information here:Benefits of Bi-Amping (Not Quite Magic, But Close) - Part 1
Topic 1.3 is Power Distribution and SPL.
Topic 1.3 is Power Distribution and SPL.
From experience - three ways use most power in the midrange, 80 - 2000Hz.
In my system I used a 60 W for >2kHz
a 200W for 80 - 2kHz
a 200W for >80Hz
I could bring the midrange bryston 4B ST (>300W/channel) to go into thermal cutout with a two way satellite MTM layout and each M capable of > 100W (about 86% efficient). The system was equalized to deal mainly with the peaks for in room response, only slightly elevating the throughs.
The sub amp would still be running.
In my system I used a 60 W for >2kHz
a 200W for 80 - 2kHz
a 200W for >80Hz
I could bring the midrange bryston 4B ST (>300W/channel) to go into thermal cutout with a two way satellite MTM layout and each M capable of > 100W (about 86% efficient). The system was equalized to deal mainly with the peaks for in room response, only slightly elevating the throughs.
The sub amp would still be running.
Here isa little Excel spreadsheet based on info in a "Fane" PA loudspeaker booklet I have somewhere.
http://users.tpg.com.au/users/gradds/powersharing.htm
http://users.tpg.com.au/users/gradds/powersharing.htm
Hi
Here is nice tool - DrivePower
The program for definition of a parity of capacities drivers in acoustic system.
http://www.diy-audio.narod.ru/files/DP_v1.1.rar
Switch in Eng "ßçûê -> English"
Here is nice tool - DrivePower
The program for definition of a parity of capacities drivers in acoustic system.
http://www.diy-audio.narod.ru/files/DP_v1.1.rar
Switch in Eng "ßçûê -> English"
Dominique, the program was not started at all?
Probably it of that the program presence of library midas.dll is necessary (it in archive with program)
For Win98 should copy it in folder Windows\System and for winXP in Windows\System32
This program will work only for windows-family OS I guess...
Hope it will help
Probably it of that the program presence of library midas.dll is necessary (it in archive with program)
For Win98 should copy it in folder Windows\System and for winXP in Windows\System32
This program will work only for windows-family OS I guess...
Hope it will help
Hi,
the power spread through the frequency range is more of an average thing and does not relate well to peak values in the signal.
IF the speakers are of the same sensitivity, then it is usual to find that the PEAK signal is similar anywhere in the frequency spectrum.
The result of this is that the peak voltage to be fed to each speaker is similar and since we are comparing equal sensitivity we are also equalising impedance.
In other words the peak power requirement is similar all the way through the frequency range.
HOWEVER, when you start mixing speakers of different sensitivity you will find that the drive requirement becomes skewed very much away from treble towards bass.
An example could be 1000W into a 84db 4ohm bass driver, 100W into a 93db 8ohm mid range driver and just 25W into a 98db 8ohm treble driver. This skewing is so common many fall into the trap of thinking it has something to do with the spectral spread of power in the signal. Don't be fooled.
the power spread through the frequency range is more of an average thing and does not relate well to peak values in the signal.
IF the speakers are of the same sensitivity, then it is usual to find that the PEAK signal is similar anywhere in the frequency spectrum.
The result of this is that the peak voltage to be fed to each speaker is similar and since we are comparing equal sensitivity we are also equalising impedance.
In other words the peak power requirement is similar all the way through the frequency range.
HOWEVER, when you start mixing speakers of different sensitivity you will find that the drive requirement becomes skewed very much away from treble towards bass.
An example could be 1000W into a 84db 4ohm bass driver, 100W into a 93db 8ohm mid range driver and just 25W into a 98db 8ohm treble driver. This skewing is so common many fall into the trap of thinking it has something to do with the spectral spread of power in the signal. Don't be fooled.
AndrewT said:
Well, at least there is a relation between the spectral spread of needed power in the amplifier and the spectral spread of power in the signal.
But now you leave me a bit confused.
As I understand it, some factors may distort this relation, as the size of the room for example.
Unfortunately I have to go now, thanks for joining the discussion!
Cheers,
DOminique
Hi again,
{I just wrote this post offline before coming here, it was written before I read Andrew's post. (just to clarify why his comments aren't mentioned)...}
The recommendations vary a lot:
With
1) the table from http://sound.westhost.com/bi-amp.htm#power_dist,
2) the excel sheet from: http://users.tpg.com.au/users/gradds/powersharing.htm
and
3) the curve (sorry comments are in german) I found in a script on: www.ant.uni-hannover.de/fileadmin/institut/lehre/ut/Dateien/Skript.pdf
the power percentage for the Bass should e.g. be:
Bass up to 250Hz: 1) 42%, 2) 40%, 3) 30%
Bass up to 500Hz: 1) 57%, 2) 60%, 3) 40%
Those people seem to have different sources.
Possibly only the weighting of various music material is different, as people who listen a lot to modern music need more power for the bass,
while classical music needs more power in the midrange. A little (but interesting) table on the subject can be found here:
http://www.silcom.com/~aludwig/EARS.htm
(Scroll down to 1/3rd of the page)
I'll further dig for info and possibly make measurements on my own. But in this case I'd have to talk to some acoustic's experts to prevent errors when calculating from Decibels to Watts.
Cheers,
Dominique
{I just wrote this post offline before coming here, it was written before I read Andrew's post. (just to clarify why his comments aren't mentioned)...}
The recommendations vary a lot:
With
1) the table from http://sound.westhost.com/bi-amp.htm#power_dist,
2) the excel sheet from: http://users.tpg.com.au/users/gradds/powersharing.htm
and
3) the curve (sorry comments are in german) I found in a script on: www.ant.uni-hannover.de/fileadmin/institut/lehre/ut/Dateien/Skript.pdf
the power percentage for the Bass should e.g. be:
Bass up to 250Hz: 1) 42%, 2) 40%, 3) 30%
Bass up to 500Hz: 1) 57%, 2) 60%, 3) 40%
Those people seem to have different sources.
Possibly only the weighting of various music material is different, as people who listen a lot to modern music need more power for the bass,
while classical music needs more power in the midrange. A little (but interesting) table on the subject can be found here:
http://www.silcom.com/~aludwig/EARS.htm
(Scroll down to 1/3rd of the page)
I'll further dig for info and possibly make measurements on my own. But in this case I'd have to talk to some acoustic's experts to prevent errors when calculating from Decibels to Watts.
Cheers,
Dominique
Attachments
Hi,
It is worth mentioning that this AVERAGE spectrum spread will be useful if you are trying to establish the PSU and heatsink requirements for each of the amplifiers. Particularly if you were to go into production and needed to trim costs to make the active speaker affordable.
The transformer, smoothing capacitors, bridge diode, number of output devices and heatsink will all be dependant on the average power delivered to each of the speakers.
I note that the average power in that table (ears) went 1:1:0.1 but that the peak power went 1:1:0.6 (for voltage this becomes 1:1:0.78) supporting my contention that you design for peak voltage and not average power.
It is worth mentioning that this AVERAGE spectrum spread will be useful if you are trying to establish the PSU and heatsink requirements for each of the amplifiers. Particularly if you were to go into production and needed to trim costs to make the active speaker affordable.
The transformer, smoothing capacitors, bridge diode, number of output devices and heatsink will all be dependant on the average power delivered to each of the speakers.
I note that the average power in that table (ears) went 1:1:0.1 but that the peak power went 1:1:0.6 (for voltage this becomes 1:1:0.78) supporting my contention that you design for peak voltage and not average power.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.