Does this peak relate to transient response requirements from music playback?
I thought it you were listening to something at an average volume of one watt and then a 20dB peak showed up, you'd need to have 100 watts on tap to cover it.
Yup. That's it. I normally don't use win media player and didn't know
that was on by default.
Dennis
That's true. IF you are listening to a digital recording that has an average level of -20dBFS (full scale) at an amplifier output of 1 watt, you'd need 100 watts for the peaks. That's 28.3 volts peak into an 8 ohm load.
CDs with good dynamics can be 20-24dB average below peak. If you need 1 watt to get the average level you want, then you'll need 100 watts or more for the peaks. Say about 30 volts for 8 ohm speakers.
It all depends on your speakers, your room and how loud you want it, average. The poll above gives you a good idea of the voltages some people here use. For example, my system uses a fraction of a volt at average listening levels. Someone else with a bigger room and/or less efficient speakers will need higher voltages for those same levels.
This test allows you to determine the voltage ranges you might use.
I analyze a track and look at RMS value of the track. That way I know how far below peak the average is. -18dB is typical for some older rock/pop CDs. -24dB is common in classical works.
OK, so it seems that the DR rating given by things like JRiver is the crest factor DR -- the # of dB for a peak above the average that is present on the recording.
So yes, if there is a DR of 10 ( fairly common ), you'll need 10x average power to ensure your amp does not clip. A DR of 20 will need 100x average power, and a DR of 30 will need 1000x average power.
I've never seen a DR of 30, but would probably be there at a live event.
The highest I am aware of is a DR of 22.
So it seems that if you listen to music with a 1 watt average, you'll need a 100 watt amp.
So yes, if there is a DR of 10 ( fairly common ), you'll need 10x average power to ensure your amp does not clip. A DR of 20 will need 100x average power, and a DR of 30 will need 1000x average power.
I've never seen a DR of 30, but would probably be there at a live event.
The highest I am aware of is a DR of 22.
So it seems that if you listen to music with a 1 watt average, you'll need a 100 watt amp.
Yes, and this would go hand in hand with the popularity of 100W/ch in the pre-home theatre era. If you're working on 1W base and 20dB DR, that would have permitted over 80dB/1m for most stereo speakers at the time. 80dB/1m wouldn't have permitted a club party (and so many found out when their tweeters didn't work the day after) but it would have been more than enough for regular listening.
So I did the test and got 2.23V with the 120hz test tone. Room is 8m x 5.5m x 2.4m, about 43sqmeters. Speakers are Rega rs3 6ohm, 89db)
So I need 8.92V peak voltage from my amp. The power calculation you propose is for 8ohm speakers, how would I go and calculate for 6 or 4 ohm ones?
If I understand correctly then I take the 2.23*4= 8.92 then I square that: 79.57 and divide by the impedance of speakers: 79.57/6= 13.26W needed.
Is my calculation reasonable?
So I need 8.92V peak voltage from my amp. The power calculation you propose is for 8ohm speakers, how would I go and calculate for 6 or 4 ohm ones?
If I understand correctly then I take the 2.23*4= 8.92 then I square that: 79.57 and divide by the impedance of speakers: 79.57/6= 13.26W needed.
Is my calculation reasonable?
Last edited:
Your calculation looks reasonable to me.
These days, I'm doing quite a lot of PA work - a Behringer NU6000DSP is feeding a Beyma 15P1200Nd a side, for somewhere around 1.5kW peaks per side. Sometimes I have to mash into the limiters a bit to keep up with drummers.
Some subwoofers ought to help...
Chris
These days, I'm doing quite a lot of PA work - a Behringer NU6000DSP is feeding a Beyma 15P1200Nd a side, for somewhere around 1.5kW peaks per side. Sometimes I have to mash into the limiters a bit to keep up with drummers.
Some subwoofers ought to help...
Chris
Hello Santos. Thanks for posting. 
Since you measured 2.23 volts, I figure that you need about 6.3V RMS for the loudest signal, or about 9 volts peak, as you said.
How do you know if an amplifier can supply that? Well, you could measure it while looking for clipping or distortion. Or you could calculate backwards from the specifications. How would an amp that can supply 6.3V RMS be rated? Use Ohm's law.
Since you measured 2.23 volts, I figure that you need about 6.3V RMS for the loudest signal, or about 9 volts peak, as you said.
How do you know if an amplifier can supply that? Well, you could measure it while looking for clipping or distortion. Or you could calculate backwards from the specifications. How would an amp that can supply 6.3V RMS be rated? Use Ohm's law.
- 10 Watts RMS into 4 ohms
- 6.6 Watts RMS into 6 ohms
- 5 Watts RMS into 8 ohms
Post #994 gives a worked calculation based on a measure 0.6 volts RMS,
http://www.diyaudio.com/forums/mult...-power-do-your-speakers-need.html#post4143172
yes adding that to post1 would help all those that can't find the actual post.
There are probably other posts that give the method, but I tried searching the first 200 and missed finding the answer.
When post 994 came along, I then remembered how simple it was. But that was a week after I had already searched for the method.
There are probably other posts that give the method, but I tried searching the first 200 and missed finding the answer.
When post 994 came along, I then remembered how simple it was. But that was a week after I had already searched for the method.