doubt in 3dB level calculation

[arava prakash]

hi friends,
may know the reasons for the following
1. why 3dB only considered in frequency response?
2. why half power(or 0.707 voltage) is considered during calculation of 3 dB?

 

[chuckey]

If a signal is half power then its 3dB down. The half power points are defined to be the bandwidth of circuits.
Frank

 

[kishornakul]

Re: doubt in 3dB level calculation
1) 3dB value is taken only for the purpose of original Bandwidth calculation.
2) 0.707 value represents RMS(Root Mean Square) which is the value of 1/square root of two

 

[ravindragudi]

1. 3dB point or half power point is considered as reference in most of the amplifier systems - audio amplifiers. Human ears can't detect any difference till the audio power changes by 50%. Hence for the design of circuits this half power point is taken as reference.
2. On the log scale 50% power reduction is equivalent to -3dB.

 

[arava prakash]

If a signal is half power then its 3dB down. The half power points are defined to be the bandwidth of circuits.
Frank

i mean only " why half power is considered"

 

[arava prakash]

Re: doubt in 3dB level calculation
1) 3dB value is taken only for the purpose of original Bandwidth calculation.
2) 0.707 value represents RMS(Root Mean Square) which is the value of 1/square root of two

yes ,"0.707 value represents RMS(Root Mean Square) which is the value of 1/square root of two" is correct
but what i mean "why only half power (or 0.707 voltage) is considered and why don't we take 1/4th of power(or half of voltage"

 

[volker@muehlhaus]

but what i mean "why only half power (or 0.707 voltage) is considered and why don't we take 1/4th of power(or half of voltage"

There is no technical reason why -3dB instead of other values. People just agreed to specify this value. The -3dB bandwidth is widely used, but you can also specify -10dB bandwidth.

 

[kenambo]

Actually RMS value is considered to be the equivalent of DC voltage for an AC response...and after this -3dB the frequency rolls off more...

 

[volker@muehlhaus]

Actually RMS value is considered to be the equivalent of DC voltage for an AC response...and after this -3dB the frequency rolls off more...

Sorry, but this makes no sense. Transfer functions like S-parameters are ratios of output/input, so it makes no difference if you use peak or RMS as long as you use it consistently.

 

[arava prakash]

1. 3dB point or half power point is considered as reference in most of the amplifier systems - audio amplifiers. Human ears can't detect any difference till the audio power changes by 50%. Hence for the design of circuits this half power point is taken as refaerence.
2. On the log scale 50% power reduction is equivalent to -3dB.

ok, thank u
but Human ears ability of detection is the only reason for taking half power?

 

[FvM]

but Human ears ability of detection is the only reason for taking half power?

The explanation is wrong. Acoustics text books say the differential detection threshhold is 1 dB.

An obvious explanation refers to a first order low- or high-pass filter. It has 3 dB attenuation at it's characteristic frequency. In so far it's a "natural" way to determine the cut-off frequency of a system with first order frequency response. But as volker@muehlhaus explained, the 3 dB criterion is also convention, an arbitrary setting.

 

[vfone]

The decibel unit was invented by Bell Telephone Labs, for telephone amplifiers which were driving speakers, to analyze how much power is available, and how much power was lost in getting the signal to the speaker, and producing further a pressure change.
In audio, 1dB is the smallest volume change which can be heard in a quiet environment. 1dB represents an approximate change in pressure of 12%.
In audio systems was used 6dB, which represents twice (half) of pressure increase.
By analogy with this "half/twice" from audio, engineers agreed to use 3dB (half power) in electric circuits.

 

[arava prakash]

There is no technical reason why -3dB instead of other values. People just agreed to specify this value. The -3dB bandwidth is widely used, but you can also specify -10dB bandwidth.

thanks for reply
In which application -10dB bandwidth is used

 

[volker@muehlhaus]

https://www.edaboard.com/threads/125885/

 

[ravindragudi]

The explanation is wrong. Acoustics text books say the differential detection threshhold is 1 dB.

An obvious explanation refers to a first order low- or high-pass filter. It has 3 dB attenuation at it's characteristic frequency. In so far it's a "natural" way to determine the cut-off frequency of a system with first order frequency response. But as volker@muehlhaus explained, the 3 dB criterion is also convention, an arbitrary setting.

I had read about the differential threshold of of sound detection by humans long time back and i remember it as 3dB, that too in the context in which this thread had started
After the reply from you I did some search and found out that the JND (just noticeable difference) in sound level by human is not a fixed value. It depends on frequency and sound pressure level. The JND is about 1 dB for soft sounds around 30-40 dB at low and midrange frequencies. It may drop to 1/3 to 1/2 a decibel for loud sounds.

Refer: http://hyperphysics.phy-astr.gsu.edu/hbase/sound/db.html , and try out yourself with http://highered.mheducation.com/sites/dl/free/007312387x/334868/webers_mx.swf

The 3dB level as standard, has certainly decided based on the telephone systems being developed at that time mainly by Bell labs.

 

[chuckey]

It gives better bandwidth figures for amplifiers and loudspeakers.
Frank