The truth about decibels

[Pooh Bear]

Don Bowey wrote:

On 6/28/05 1:32 PM, in article
9tiwe.35546$Vo6.4966@fe3.news.blueyonder.co.uk, "Kevin Aylward"
see_website@anasoft.co.uk> wrote:

(snip typical Kevin blustery posturing)

But I strongly suspect Einstein stole his best ideas from his wife. You
should get a wife to help you get on track.

By the way 10log P1/P2 is still much in use, your authoritative declarations
not withstanding.

I'm not aware that *anyone* suggested that power dB ratios were ever 'wrong' !

I, for one, took offence at the 'decibel Jihadists' attempts to show that *only*
power related decibels were meaningful.

I have an open mind - you have a closed one - it closed apparently circa 1929 @
the time of the Bell Labs paper !

Graham

 

[Pooh Bear]

Don Bowey wrote:

On 6/28/05 1:52 PM, in article 42C1B8A8.42C55756@hotmail.com, "Pooh Bear"
rabbitsfriendsandrelations@hotmail.com> wrote:

tlbs wrote:

Pooh Bear wrote:
1. A measuring system first used in telephony (Martin, W.H., "DeciBel --
the new name for the transmission unit. Bell System Tech. J. January,
1929), where signal loss is a logarithmic function of the cable length.

I have actually read that original paper by Martin. I have also read
another white paper with comments similar to your own, and have tried
(in my own circle of influence) to educate people about the proper use
of the dB.

The relevant 'education' is now adopted by the AES ( Audio Engineering Society
) who write the standards on this matter. Their standards are adopted by
bodies such as the IEC, ANSI and ISO

In it's purest (and according to Martin) use, the Bel or deciBel is to
be used ONLY for power comparisons at one point in a circuit. That is,
signal 1's power was measured at point-X in a system, and signal 2's
power was measured at the SAME POINT in the system. The base-10-log of
the ratio of the 2 powers were compared and that number was given the
unit "Bel".

But he failed to mention in that paper that Bell Labs knew that 0dBm - i.e
1mW - was not a sufficiently large signal to ensure adequate volume for
clarity of conversation over the phone.

Put a telephone headphone up to your ear and listen to a one-mW signal. You
may change your mind.

Your point being ?

A modern telephone is very different to those in use in the 1920s that had carbon
granule mics and armature 'receivers'.

Therefore Bell Labs 'invented' the VU ( volume unit ) meter.

Horsepucky - the Volume Unit meter was designed to be used *after* a
facility was properly aligned with a "dB meter." to observe voice and music.
As I recall, it was slightly dampened.

So..... where did the dB meter come from ? Did it predate the dB ? LOL ! In which
case it should have been calibrated in 'transmission units' shouldn't it ?

I reckon most ppl in the developed world have seen a VU meter even if only a
miniature version on their cassette recorder.

The VU meter was calibrated such that '0' on the VU scale - normally referred
to as '0 VU' was a *voltage* equivalent to +4dBm ( 600 ohms ) in the load.

It appears you may be, incorrectly, implying the "V" means Voltage, not
"Volume."

No : VU means 'Volume Unit' and always has done. The VU meter is a voltage measuring
device though.

Note *voltage*. The VU meter was / is simply a moving coil meter with a
rectifier ( copper oxide in early examples - germanium in more recent ) and
therefore a *voltage* driven instrument.

The VU meter had / has an internal impedance of 3900 ohms and was used with an
external fixed series resistor of 3600 ohms to provide a load of 7500 ohms.
The '0 VU' mark was calibrated to be equivalent to +4dBm ( 600ohms ) i.e.
1.228 V.

So........ All the time that Bell Labs were supposedly saying that the dB was
only 'pure' if considered as power - they were *ACTUALLY* measuring it as a
*VOLTAGE* !

More horsepucky. Only a complete moron would assume that.

And only a complete moron ( aka Macon ) would disagree with the AES and the IEC too
!

Graham

 

[Pooh Bear]

Guy Macon wrote:

Adrian Tuddenham wrote:

Kevin Aylward <see_website@anasoft.co.uk> wrote:

Look mate. The definition of the dB, today, is 20 log(Q1/Q2). End of
story. Get over it.

If only we had realised this years ago - then audio designers could have
dispensed with valves, transistors and all the problems that come with
amplifiers. All you need is a step-up transformer between the
microphone and the loudspeaker.

I worship your wit and wisdom, O Great One.

Why don't you slip your tongue up his arse whilst you're at it ?

Graham

 

[Bob]

You may want to re-think that.

Here are some specs for non-loaded 26G pic cable; a common Exchange Cable
type:

Resistance per mile from 1 Hz to 15kHz = 441 Ohms. Beyond that freq the
skin effect starts increasing the resistance. At 1MHz it's 463 Ohms, and
at
5 MHz it's 2044 Ohms.


Approximate Z per mile at 1 Hz is 20,562 Ohms, and is up to 2057 Ohms at
100
Hz.

In the band from 300 Hz to 3 kHz the Z runs from 1189 Ohms to 383 Ohms.
At
5Mhz the Z is 96 Ohms/mile.

In rural areas that are still served by copper, rather than a nearby mux,
the cables are H88 loaded, and can be assumed to have a 600 Ohm Z.

Don

An ex-Toll bastard.

These load coils were a great invention (for their time). Whoever invented
them (I think it was Heaviside) was pretty short-sighted, though. He should
have known it would really fugg-up our DSL service!

Graham, you've really caused quite a stir, with this thread. I'll bet you
didn't expect a sort of Spanish Inquisition, did you?

Bob

 

[Ban]

Adrian Tuddenham wrote:

tlbs <tlbs101@excite.com> wrote:

Here is an example of BAD usage of the dB:
If you ever find an OpAmp datasheet that charts its' gain vs.
frequency, and the Y-axis is expressed in dB, that's bad. Why?
because the input impedance of an Op Amp is nearly infinite, while
the output impedance is nearly zero. The impedances are NOT the
same, therefore you cannot express the gain ratio in terms of dB.

I used to see this all the time in the 70's and occasionally in the
80's. The CORRECT y-axis units is Volts/Volt, and if they want to
express it as a logarithm, 20*log(Vout/Vin) that's perfectly fine,
but it is NOT a dB. I repeat that is NOT a dB.

Here is an example of GOOD usage of the dB:
The examples of RF components from Mini-Circuits given above and
their use of dB, or dBm IS PERFECTLY OK. Why? because their
components are all 50 Ohm and 75 Ohm systems and are explicitly
stated as such. 50 Ohm inputs, 50 Ohm outputs. 75 Ohm inputs, 75
Ohm outputs. The gains or attenuations listed assume (properly)
that the source impedance and load impedances connected to the
devices sold are also of the same impedance as the devices sold. So
if it states that a device has a gain of 13 dB, it has a gain of 13
dB in an implicit 50 Ohm (or 75 Ohm for that part number) system.
If the output is stated at +33 dBm, then it will produce 10 V across
50 Ohms and 2W into 50 Ohms.

I have designed quite a few RF circuits in my time -- this is how
it's done.

Units like dBc generally fall within the original definition of the
Bel and dBel, because the power levels of the carrier and the
modulating signal can be measured at the same point in the circuit.

Thank you for that interesting information and brilliantly clear
exposition. I only wish you had written some of the modern textbooks,
then we wouldn't have the present misunderstandings.

(Sadly, though, we'd probably still have to put up with the occasional
objectionable and obstreperous individuals who disgrace the profession
and its institutions)

This discussion has been going on for 50+yrs. The "purists" want to restrict
the use of dB strictly to a power ratio, and the engineers use it for
voltage or even current ratio. Of course everyone knows what is meant exept
a few college kids, who are supposed to learn it.
Now to settle the argument forever the engineer crowd came up with dBu, with
"u" meaning *unspecified* impedance. When the impedance is unspecified, you
cannot express a power ratio. So again the meaning is clear and widely used
by professional audio companies to specify their products, whatever Pooh may
say. It is also clear that in data sheets any ratio (voltage, current,
distortion, noise etc.) is expressed in dB, but never any power ratio. I
doubt that even in RF systems the power ratio is of any importance.
So despite the historical background today we use dBu, dBm, dBV... *only*
for voltage levels. Power levels are hardly expressed at all and if, they
are labeled in Watts.
When somebody expresses the opposite opinion, he does it just for the
arguments sake, knowing perfectly well what is meant.
--
ciao Ban
Bordighera, Italy

 

[Kevin Aylward]

Don Bowey wrote:

On 6/28/05 1:32 PM, in article
9tiwe.35546$Vo6.4966@fe3.news.blueyonder.co.uk, "Kevin Aylward"
see_website@anasoft.co.uk> wrote:


But I strongly suspect Einstein stole his best ideas from his wife.

ROTHLMAO

By the way 10log P1/P2 is still much in use, your authoritative
declarations not withstanding.

Sure, people can use 20.log(Q1/Q2) with the Qs set equal to power. The
math is completely general and can be applied to anything. What's your
point?

Kevin Aylward
informationEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Kevin Aylward]

Adrian Tuddenham wrote:

Kevin Aylward <see_website@anasoft.co.uk> wrote:


[...]

Look mate. The definition of the dB, today, is 20 log(Q1/Q2). End of
story. Get over it.

If only we had realised this years ago - then audio designers could
have dispensed with valves, transistors and all the problems that
come with amplifiers. All you need is a step-up transformer between
the microphone and the loudspeaker.

Ho hummm...complete nonsense again.

What part of voltage gain does not mandate power gain do you not
understand? There is *nothing* in the definition above that states that
*if* the Q's are voltage, that there is power gain. Power is simply not
part of the equation, so the equation says nothing about power.

Now get this, its a mathematical definition, not a physical definition.

Sure, those that are ignorant of electrical engineering might well be
confused as to what gain means.

Kevin Aylward
informationEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Adrian Tuddenham]

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

tlbs wrote:

Pooh Bear wrote:
1. A measuring system first used in telephony (Martin, W.H., "DeciBel --
the new name for the transmission unit. Bell System Tech. J. January,
1929), where signal loss is a logarithmic function of the cable length.

I have actually read that original paper by Martin. I have also read
another white paper with comments similar to your own, and have tried
(in my own circle of influence) to educate people about the proper use
of the dB.

The relevant 'education' is now adopted by the AES ( Audio Engineering Society
) who write the standards on this matter. Their standards are adopted by
bodies such as the IEC, ANSI and ISO

From the reference you provided in the "3dB Bandwidth" thread:

<http://www.aes.org/publications/standards/courtesy.cfm?ID=9>

....it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

 

[Adrian Tuddenham]

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

Adrian Tuddenham wrote:

Kevin Aylward <see_website@anasoft.co.uk> wrote:

[...]

Look mate. The definition of the dB, today, is 20 log(Q1/Q2). End of
story. Get over it.

If only we had realised this years ago - then audio designers could have
dispensed with valves, transistors and all the problems that come with
amplifiers. All you need is a step-up transformer between the
microphone and the loudspeaker.

Look here plonker.... It was actually the transistor that caused all this
fuss. Not to mention the damn IC !

If, like you, we still lived back in the 1930s there wouldn't be this
problem. And we would be listening to 'chamber music' on the BBC third
service for our entertainment !

In the 1930s the BBC broadcast the National and the Regional programmes.
The Third Programme wasn't started until 1946.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

 

[Adrian Tuddenham]

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

Don Bowey wrote:

On 6/28/05 1:52 PM, in article 42C1B8A8.42C55756@hotmail.com, "Pooh Bear"
rabbitsfriendsandrelations@hotmail.com> wrote:


But he failed to mention in that paper that Bell Labs knew that 0dBm
- i.e1mW - was not a sufficiently large signal to ensure adequate
volume for clarity of conversation over the phone.

Put a telephone headphone up to your ear and listen to a one-mW signal. You
may change your mind.

Your point being ?

A modern telephone is very different to those in use in the 1920s that had
carbon granule mics and armature 'receivers'.

The type of mic doesn't alter the sensitivity of the receiver to line
levels.

The receivers in 1920s telephones had magnetic diaphragms, not rocking
armatures. I have just tested a '105 CB' "candlestick" telephone from
the 1920s with 0dBm line input level at 1 Kc/s (from a 600 ohm source),
it gives 104dBA SPL at the earpiece.

I hope your next theory is better than that one.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

 

[Guy Macon]

Adrian Tuddenham wrote:

...it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.

You are incorrect.
http://www.aes.org/standards/
http://www.aes.org/publications/standards/list.cfm
http://www.aes.org/standards/b_comments/index.cfm

What is inappropriate about the plonkee's attempt to use the AES
as an authority is that the AES has standards for practices *within
the audio industry*. They are the primary standards bearers for
such things a XLR microphone connector polarity and gender but are
secondary standards for such things as the definition of dB or V
which are used inside of and outside of the audio industry.

 

[martin griffith]

On Wed, 29 Jun 2005 03:21:01 GMT, in sci.electronics.design "Bob"
<nimby1_notspamm_@earthlink.net> wrote:

snip

Graham, you've really caused quite a stir, with this thread. I'll bet you
didn't expect a sort of Spanish Inquisition, did you?

Bob

wait until his next post

"The Truth about Ohms law"

you aint seen nuffing yet!


martin

 

[Guy Macon]

martin griffith wrote:

"Bob" <nimby1_notspamm_@earthlink.net> wrote:

Graham, you've really caused quite a stir, with this thread. I'll bet you
didn't expect a sort of Spanish Inquisition, did you?

wait until his next post

"The Truth about Ohms law"

you aint seen nuffing yet!

Then comes "The Truth About VA vs. W"

After that comes "The Truth About Electron Flow and Hole Flow"

Then comes "The Truth: I Flame Because I am an Assh*le"

 

[Pooh Bear]

Winfield Hill wrote:

martin griffith wrote...

Nope, I just play the 15ips 1/4" tape, much more satifying.

I've forgotten, how long does a 7" 15ips reel of tape play,
say compared to a CD? Or do you keep and play larger reels?

Off the top of my head I seem to remember 30 mins for pro stuff . But
depended on the tape thickness.

But then pros used 10 1/2 inch spools ! Typically good for 30 ~ 45 mins
@ 15 ips.

You could get SP ( standard play ) , LP ( long play ) and DP ( double
play ) tapes.

Obviously the polyester ( don't even mention acetate ! ) film base had
to be made thinner ( and therefore less mechanically robust ) as 'play
time' was increased.

The tape maufacturers tried to maintain oxide thickness ( for s/n
reasons ) but there's no doubt that DP never performed as well as SP.

Most ppl compromised on LP tape.

You saw a similar thing with compact cassette tape. The play time was
inversely proportional to the tape width. The thinner tapes didn't last
long in an automotive player environment !

C90 cassettes were normally considered optimal. So you had C60 ( SP ) ,
C90 ( LP ), C120 ( DP ) and a few clowns used C180s !

A few cassettes were loaded with less than the maximum possible amount
of tape for limited usage for the like of transcription / dictation
purposes such as C30s. I think even a few C15s existed.

The C number referred to the total number of play minutes.


Graham

 

[Adrian Tuddenham]

Guy Macon <_see.web.page_@_www.guymacon.com_> wrote:

Adrian Tuddenham wrote:

...it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.
[...]
They are the primary standards bearers for
such things a XLR microphone connector polarity and gender but are
secondary standards for such things as the definition of dB or V
which are used inside of and outside of the audio industry.

The definitions of dB and V were the standards we were talking about.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

 

[Don Bowey]

On 6/28/05 11:44 PM, in article
Grrwe.36256$Vo6.863@fe3.news.blueyonder.co.uk, "Kevin Aylward"
<see_website@anasoft.co.uk> wrote:

Don Bowey wrote:
On 6/28/05 1:32 PM, in article
9tiwe.35546$Vo6.4966@fe3.news.blueyonder.co.uk, "Kevin Aylward"
see_website@anasoft.co.uk> wrote:


But I strongly suspect Einstein stole his best ideas from his wife.

ROTHLMAO


By the way 10log P1/P2 is still much in use, your authoritative
declarations not withstanding.

Sure, people can use 20.log(Q1/Q2) with the Qs set equal to power. The
math is completely general and can be applied to anything. What's your
point?

My point is, 10x and 20x the log of a number, obviously provides two
different solutions.

Why do you continue to argue against the obvious?

BTW she was the brighter of the two.

Don

 

[Don Bowey]

On 6/29/05 4:48 AM, in article
1gyx4r5.i35grr17n8c5qN%poppy.uk@ukonline.invalid.invalid, "Adrian Tuddenham"
<poppy.uk@ukonline.invalid.invalid> wrote:

Guy Macon <_see.web.page_@_www.guymacon.com_> wrote:

Adrian Tuddenham wrote:

...it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.
[...]
They are the primary standards bearers for
such things a XLR microphone connector polarity and gender but are
secondary standards for such things as the definition of dB or V
which are used inside of and outside of the audio industry.

The definitions of dB and V were the standards we were talking about.

They don't write standards redefining dB nor V either.

They are preparing application *guidelines* for their members, which cannot
be published as "Standards." Although the AEC is not affiliated with ANSI,
I see they are using much of the same organization as other ANSI groups such
as ANSI TIA, etc. As such they should issue the final guidelines an an
Appendix to a standard, or as a Technical Report, but NOT within the body of
a standard.

Don

 

[Pooh Bear]

Adrian Tuddenham wrote:

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

tlbs wrote:

Pooh Bear wrote:
1. A measuring system first used in telephony (Martin, W.H., "DeciBel --
the new name for the transmission unit. Bell System Tech. J. January,
1929), where signal loss is a logarithmic function of the cable length.

I have actually read that original paper by Martin. I have also read
another white paper with comments similar to your own, and have tried
(in my own circle of influence) to educate people about the proper use
of the dB.

The relevant 'education' is now adopted by the AES ( Audio Engineering Society
) who write the standards on this matter. Their standards are adopted by
bodies such as the IEC, ANSI and ISO

From the reference you provided in the "3dB Bandwidth" thread:

http://www.aes.org/publications/standards/courtesy.cfm?ID=9

...it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.

The AES is represented on an official basis on the Working Groups.

The are also *audio specific* standards that the AES develops that are then formally
adopted by the well known standards bodies. Not the other way round.

Semantics.

Graham

 

[Don Bowey]

On 6/29/05 1:29 PM, in article 42C304B5.89046EC4@hotmail.com, "Pooh Bear"
<rabbitsfriendsandrelations@hotmail.com> wrote:

Adrian Tuddenham wrote:

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

tlbs wrote:

Pooh Bear wrote:
1. A measuring system first used in telephony (Martin, W.H., "DeciBel --
the new name for the transmission unit. Bell System Tech. J. January,
1929), where signal loss is a logarithmic function of the cable length.

I have actually read that original paper by Martin. I have also read
another white paper with comments similar to your own, and have tried
(in my own circle of influence) to educate people about the proper use
of the dB.

The relevant 'education' is now adopted by the AES ( Audio Engineering
Society
) who write the standards on this matter. Their standards are adopted by
bodies such as the IEC, ANSI and ISO

From the reference you provided in the "3dB Bandwidth" thread:

http://www.aes.org/publications/standards/courtesy.cfm?ID=9

...it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.

The AES is represented on an official basis on the Working Groups.

Are you *certain* of that? I bet the governing body is the AES, and the
Working Groups work under the AES.

The are also *audio specific* standards that the AES develops that are then
formally
adopted by the well known standards bodies. Not the other way round.

The AES "standards" may or may not be adopted by other standards bodies.

Semantics.

Important, meaningful semantics.


The AES is NOT issuing a "standard" having new definitions. What you
have been saying is NOT CORRECT.

Weasel-word around it all you want, but it won't change the facts.

Don

 

[John Fields]

On Wed, 29 Jun 2005 17:09:09 -0700, Don Bowey <dbowey@comcast.net>
wrote:

On 6/29/05 1:29 PM, in article 42C304B5.89046EC4@hotmail.com, "Pooh Bear"
rabbitsfriendsandrelations@hotmail.com> wrote:

Adrian Tuddenham wrote:

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

tlbs wrote:

Pooh Bear wrote:
1. A measuring system first used in telephony (Martin, W.H., "DeciBel --
the new name for the transmission unit. Bell System Tech. J. January,
1929), where signal loss is a logarithmic function of the cable length.

I have actually read that original paper by Martin. I have also read
another white paper with comments similar to your own, and have tried
(in my own circle of influence) to educate people about the proper use
of the dB.

The relevant 'education' is now adopted by the AES ( Audio Engineering
Society
) who write the standards on this matter. Their standards are adopted by
bodies such as the IEC, ANSI and ISO

From the reference you provided in the "3dB Bandwidth" thread:

http://www.aes.org/publications/standards/courtesy.cfm?ID=9

...it is abundantly clear that the AES does *not* write the standards.
They adopt the standards of the IEC etc. and then publish documents
explaining them to their membership.

The AES is represented on an official basis on the Working Groups.

Are you *certain* of that? I bet the governing body is the AES, and the
Working Groups work under the AES.


The are also *audio specific* standards that the AES develops that are then
formally
adopted by the well known standards bodies. Not the other way round.

The AES "standards" may or may not be adopted by other standards bodies.


Semantics.

Important, meaningful semantics.


The AES is NOT issuing a "standard" having new definitions. What you
have been saying is NOT CORRECT.

Weasel-word around it all you want, but it won't change the facts.

Don

---
Bottom line?

The AES sucks unless I care about how my XLS connector pins out
against yours.

ISTM It's basically all just wiring.

Or maybe I'm missing something...

--
John Fields
Professional Circuit Designer