>> hi,
>> could you please let me know how to convert base 2 logrithms of energy to
>> dBov (dB level relative to over load ).
>>
>> thanks
>> -Lingadevaru
Converting energy to voltage requires a set of assumptions that I'm not
prepared to make for you. Once you can make that conversion and also
know the overload voltage of your system, the rest is easy.
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Jerry Avins●December 29, 20042004-12-29
devaru wrote:
> hi,
> could you please let me know how to convert base 2 logrithms of energy to
> dBov (dB level relative to over load ).
>
> thanks
> -Lingadevaru
Converting energy to voltage requires a set of assumptions that I'm not
prepared to make for you. Once you van make that conversion and also the
overload voltage of your system, the rest is easy.
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by devaru●December 29, 20042004-12-29
hi,
could you please let me know how to convert base 2 logrithms of energy to
dBov (dB level relative to over load ).
thanks
-Lingadevaru
Reply by Steve Underwood●May 22, 20042004-05-22
Ben Bradley <ben_nospam_bradley@mindspring.com> wrote in message news:<cv3ga0tibfqq6rvojar25d9evqgkans6vh@4ax.com>...
> Back to quoting Steve Underwood:
>
> >That is defined as 8 specific samples
> >which will produce a 1kHz sine wave that is considered 0dBO. The clip
> >point turns about to be about +3.14dBO for a sine wave. I would think
>
> That doesn't sound right. I'd think a sine wave should be either
> 0dBO (relative to a max sine) or -3dBO (relative to a max square).
Well, I guess they didn't care what sounds right to you, because G.711
went ahead and defined it like that anyway. :-)
I have never checked this, but I think the rather odd figure of 3.14dB
comes about because no simple pattern of u-law or A-law values they
could quote in the spec. as a reference would give a sine wave
precisely 3dB below clip.
Regards,
Steve
Reply by Jerry Avins●May 17, 20042004-05-17
Ben Bradley wrote:
...
> I've not heard of either dBO (the letter Capital O) nor dBov, and
> I didn't see anything anything relevant for "dBO" in Google's first
> 100 links. But dBov shows up, here's a definition - "ov" (and 'O') is
> apparently short for "overload":
> http://ftp.tiaonline.org/UWC136/136-250.pdf
>
>
>>dBov. Sound level in decibels with respect to 16-bit overload (the
>>maximum value in a 16-bit word, 32768).
...
Admirable! The rest of us have been arguing about what time is is, with
the likelihood that someone will call actually call for a vote, and Ben
goes and looks at a clock. Way to go!
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Ben Bradley●May 17, 20042004-05-17
On 11 May 2004 19:05:54 -0700, steveu@coppice.org (Steve Underwood)
wrote:
>Hi,
>
>I don't know about dBov, but perhaps it is similar to dBO, used in PCM
>(i.e. A-law and u-Law) telecoms.
I've not heard of either dBO (the letter Capital O) nor dBov, and
I didn't see anything anything relevant for "dBO" in Google's first
100 links. But dBov shows up, here's a definition - "ov" (and 'O') is
apparently short for "overload":
http://ftp.tiaonline.org/UWC136/136-250.pdf
>dBov. Sound level in decibels with respect to 16-bit overload (the
>maximum value in a 16-bit word, 32768).
This is the same as dBFS (though dBFS is relative to max
representation regardless of the number of bits per word - I suspect
dBov is too, and the above definition is unnecesarily restrictive),
which I've seen used in pro audio to represent the value of a digital
signal below the max of 0dBFS (for example, a good "24-bit" [yes, it's
not really accurate to 24 bits, but it does send 24 bits of
"something" to the computer for every sample time] soundcard or
outboard A/D converter may have internal noise measured at -105 dBFS.
If there were such a thing as a "true 24-bit" audio converter, its
noise would be speced at about -144dBFS (using 6dB per bit times 24
bits).
Quoting from
http://ietf.levkowetz.com/drafts/avt/profile-new/draft-ietf-avt-profile-new-02.txt
> The noise level is expressed in dBov, with values from 0 to 127 dBov.
> dBov is the level relative to the overload of the system. (Note:
> Representation relative to the overload point of a system is
> particularly useful for digital implementations, since one does not
> need to know the relative calibration of the analog circuitry.)
> Example: In 16-bit linear PCM system (L16), a signal with 0 dBov
> represents a square wave with the maximum possible amplitude (+/-
> 32767). -63 dBov corresponds to -58 dBm0 in a standard telephone
> system. (dBm is the power level in decibels relative to 1 mW, with an
> impedance of 600 Ohms.)
I don't know if some of these are typos or what (this is supposed
to be some standard, so it should have been gone over with a
fine-tooth comb), but the first line talks about 0 to 127 dBov (a
positive number) but later it uses a negative number, -63 dbov - I
would have expected all such number to be negative with a max of 0 (or
a max of about +3dbov, depending on the exact definition). And then
there's "dBm0" which I've never seen, but then it defines dBm with
which I'm familar.
I designed a modem for a POTS line (okay, this was 1996, it's
getting old on my resume), and never heard of dbov. I'm very familiar
with dBm (what the phoneline specs use), dBu, dBV (what the Tek scope
FFT displays use), and from my familarity with professional
(/semi-pro/home recording) audio, I'm used to seeing dBFS (for "Full
Scale") as a maximum representation for digitized signals.
Cool Edit (the company was bought by Adobe and product renamed
"Audition") under Analyze->Statistics has an option for displaying RMS
values with "0dB=FS Sine wave" or "0dB=FS Square wave," and toggling
between these two changes the displayed RMS values by 3.01dB, as
expected.
So the maximum RMS value a signal can be, compared to a max square
wave, is 0dBFS, but compared to a max sine wave it can be as high as
+3.01dBFS. I like the square wave comparison better, just because it
bothers me to see a positive dBFS value.
Peak and peak-to-peak values are always just 'what they are' and
don't need any more definition (and they're always negative or at most
zero in 'dBFS').
Back to quoting Steve Underwood:
>That is defined as 8 specific samples
>which will produce a 1kHz sine wave that is considered 0dBO. The clip
>point turns about to be about +3.14dBO for a sine wave. I would think
That doesn't sound right. I'd think a sine wave should be either
0dBO (relative to a max sine) or -3dBO (relative to a max square).
>it likely that dBov would be defined in a similar sine wave manner,
>but with the wave being considered 0dBov just below clip. Square wave
>are also a possibility, I guess, but engineers have a deep love of
>sine waves. :-)
>
>Regards,
>Steve
>
>
>Randy Yates <randy.yates@sonyericsson.com> wrote in message news:<xxpoeovne6i.fsf@usrts005.corpusers.net>...
>> Jerry Avins <jya@ieee.org> writes:
>> > [...]
>> > If I wanted a measure relative to overload, I'd define overload first.
>> > For DACs and ADCs, it's peak voltage. It seems reasonable (but hardly
>> > proof) that dBov is defined that way.
>>
>> I would say it's clear what an overload is (something exceeding full-scale),
>> but since there are a number of ways to specify it. The question is then
>> "Which way was it specified" (e.g., "full-scale RMS sine wave" or "full-scale
>> square wave power" are two most-probable methods).
>>
>> By the way, an SEMC colleague believes it is the full-scale RMS sinewave
>> power method.
> On Wed, 12 May 2004 08:59:56 -0400, Jerry Avins <jya@ieee.org> wrote:
>
>
>>Allan Herriman wrote:
>>
>>
>>>On Tue, 11 May 2004 13:55:40 -0400, Jerry Avins <jya@ieee.org> wrote:
>>>
>>>
>>>
>>>
>>>>A*[sin(wt) + sin(3wt)] and A*[sin(wt) - sin(3wt)] have
>>>>precisely the same RMS power
>>>
>>>
>>>I could have sworn they have the same *average* power.
>>>
>>>Regards,
>>>Allan.
>>
>>The total RMS power is the sum of the components' individual RMS powers.
>>A component's RMS power is unchanged by changing its sign. I haven't
>>(and won't) compute average power, but I guess you're right. I can see
>>by inspection that the ratio of peak powers is about 5:8.
>>
>>How do we disagree?
>
>
> I'm not sure. How do you define "RMS Power"?
>
> Ta,
> Allan.
Now I get you! My bad! Sheesh! I stepped in it and tracked it into the
house! (Thanks. Really.)
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Allan Herriman●May 12, 20042004-05-12
On Wed, 12 May 2004 08:59:56 -0400, Jerry Avins <jya@ieee.org> wrote:
>Allan Herriman wrote:
>
>> On Tue, 11 May 2004 13:55:40 -0400, Jerry Avins <jya@ieee.org> wrote:
>>
>>
>>
>>>A*[sin(wt) + sin(3wt)] and A*[sin(wt) - sin(3wt)] have
>>>precisely the same RMS power
>>
>>
>> I could have sworn they have the same *average* power.
>>
>> Regards,
>> Allan.
>
>The total RMS power is the sum of the components' individual RMS powers.
>A component's RMS power is unchanged by changing its sign. I haven't
>(and won't) compute average power, but I guess you're right. I can see
>by inspection that the ratio of peak powers is about 5:8.
>
>How do we disagree?
I'm not sure. How do you define "RMS Power"?
Ta,
Allan.
Reply by Jerry Avins●May 12, 20042004-05-12
Allan Herriman wrote:
> On Tue, 11 May 2004 13:55:40 -0400, Jerry Avins <jya@ieee.org> wrote:
>
>
>
>>A*[sin(wt) + sin(3wt)] and A*[sin(wt) - sin(3wt)] have
>>precisely the same RMS power
>
>
> I could have sworn they have the same *average* power.
>
> Regards,
> Allan.
The total RMS power is the sum of the components' individual RMS powers.
A component's RMS power is unchanged by changing its sign. I haven't
(and won't) compute average power, but I guess you're right. I can see
by inspection that the ratio of peak powers is about 5:8.
How do we disagree?
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Allan Herriman●May 12, 20042004-05-12
On Tue, 11 May 2004 13:55:40 -0400, Jerry Avins <jya@ieee.org> wrote:
>A*[sin(wt) + sin(3wt)] and A*[sin(wt) - sin(3wt)] have
>precisely the same RMS power
I could have sworn they have the same *average* power.
Regards,
Allan.