The truth about decibels

[Mac]

On Mon, 27 Jun 2005 20:50:49 +0100, Pooh Bear wrote:

decibel Abbr. dB Equal to one-tenth of a bel. [After Alexander Graham
Bell.]
[snip]

The dBm ( 600 ohms ) is all but extinct in real usage.

Perhaps in audio. In the RF world, it is still alive and kicking, although
it would usually be into 50 Ohms.

Graham

--Mac

 

[Pooh Bear]

Mac wrote:

On Mon, 27 Jun 2005 20:50:49 +0100, Pooh Bear wrote:

decibel Abbr. dB Equal to one-tenth of a bel. [After Alexander Graham
Bell.]
[snip]

The dBm ( 600 ohms ) is all but extinct in real usage.

Perhaps in audio. In the RF world, it is still alive and kicking, although
it would usually be into 50 Ohms.

Which is why I precisely mentioned the dBm ( 600ohms ).

The dBm ( 600 ohms ) is all but dead.

I'm perfectly aware that the dBm is used at other reference impedances for RF.

Indeed this is *exactly* the problem with quoting the dBm without reference to
impedance ! Sadly many do this.

Graham

 

[Pooh Bear]

"James T. White" wrote:

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:WZYve.421$0V3.386@newssvr13.news.prodigy.com...
Hello Graham,

dBm all over the place...

Plus dBi and dBd for antennas!
--
James T. White

Did I *not*say that the RF guys adopted the dB ? ;-)

Since you mention them - us LF Luddites ;-) would be quite interested as
to how you RF guys use the dB !

The measures you mention are new to me.

I'd heard of dBc - third order intercept ( whatever that is ! ) but it
never related to my area of interest so I didn't pay enough attention.

As for attennas, I suspect you're talking of directional gain ?

That was important in CH and CHL.

Graham

 

[Ban]

Pooh Bear wrote:

I could give you a typical example of how audio working dropped the
600 ohm fiasco and bettered its performance by leaps and bounds
overnight but it's a long story !

Graham

Well 30yrs. ago everything in audio was expressed in dBm and people started
to argue if that was a power or voltage ratio etc etc. Now we use dBu, which
explicitly means *unspecified* impedance. So there is no question of power-
but only voltage ratio. But the reference level remained the same.
Another bad thing that happened at the same time was, that on the commercial
HiFi gear the dBV had been introduced by some japonese companies with a
different reference level (1V).
This contributed more to the confusion than the beforementioned definition
problem, because now we have two different reference levels for HiFi and Pro
audio. Studio reference is +4dBu, HiFi is -10dBV, but their difference is
not 14 but only 11.79dB.
--
ciao Ban
Bordighera, Italy

 

[Pooh Bear]

Ban wrote:

Pooh Bear wrote:

I could give you a typical example of how audio working dropped the
600 ohm fiasco and bettered its performance by leaps and bounds
overnight but it's a long story !

Graham

Well 30yrs. ago everything in audio was expressed in dBm and people started
to argue if that was a power or voltage ratio etc etc. Now we use dBu, which
explicitly means *unspecified* impedance.

But specified *voltage* : 0dBu = 0.775v just like 0 dBm ( 600 ohm )

So there is no question of power-
but only voltage ratio. But the reference level remained the same.

Yes, for historical reasons that wisely weren't discarded.

Another bad thing that happened at the same time was, that on the commercial
HiFi gear the dBV had been introduced by some japonese companies with a
different reference level (1V).

I'm sure you mean -10dBV ?

This contributed more to the confusion than the beforementioned definition
problem, because now we have two different reference levels for HiFi and Pro
audio. Studio reference is +4dBu, HiFi is -10dBV, but their difference is
not 14 but only 11.79dB.

I have to deal with this daily ! Good news is that -10dBV working is now
effectively extinct ! It was only ever a ploy by Japanese mfrs to attempt a new
standard that used cheaper build ! The pro-audio industry gave them some space -
but it fell into disrepute of its own accord on account of its built-in
limitations such as poor noise performance.

Graham

 

[martin griffith]

On 27 Jun 2005 19:09:51 -0700, in sci.electronics.design Winfield Hill
<Winfield_member@newsguy.com> 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?

I dont use 7", all 10.5" reels, nominally 30mins.. A lot of stuff, ie
a "song " was distributed on 5" reels in the good old days





martin

 

[martin griffith]

On Tue, 28 Jun 2005 06:49:04 +0100, in sci.electronics.design Pooh
Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

Ban wrote:

Pooh Bear wrote:
snip
This contributed more to the confusion than the beforementioned definition
problem, because now we have two different reference levels for HiFi and Pro
audio. Studio reference is +4dBu, HiFi is -10dBV, but their difference is
not 14 but only 11.79dB.

I have to deal with this daily ! Good news is that -10dBV working is now
effectively extinct ! It was only ever a ploy by Japanese mfrs to attempt a new
standard that used cheaper build ! The pro-audio industry gave them some space -
but it fell into disrepute of its own accord on account of its built-in
limitations such as poor noise performance.

Graham
This is where I get slightly irritated, broadcast vs studio line up

levels. The beeb/broadcasters use 0dBu as ref level, ie PPM4 or -4Vu.
While studio's use +4dBu, on those nasty Vu meters. In the analogue
days it was quite useful to write it on the box, or record report.

But it's all changed with digits, People still use Vu meters when
driving digital devices and calibrate OVu to -20dBfs. quite probably
under driving the ADCs considerably, as digital clipping is not nice.
I think the EBU still say that with PPMs you should calibrate PPM4 to
-18dBfs, I think that in normal practice this gives peaks of about
-6dBfs. but I'm a bit out of touch, so practices may have changed.

There are a couple of good threads in Rec.Arts.Movie.Production.Sound
on this. They seem to be much more clued up on levels than
rec.audio.pro


martin

 

[Kevin Aylward]

Pooh Bear wrote:

decibel Abbr. dB Equal to one-tenth of a bel. [After Alexander Graham
Bell.]


0 dBr An arbitrary reference level (r = re; or reference) that must be
specified. For example, a signal-to-noise graph may be calibrated in
dBr, where 0 dBr is specified to be equal to 1.23 Vrms (+4 dBu);
commonly stated as "dB re +4," that is, "0 dBr is defined to be equal
to +4 dBu."

This is a bad example. A S/N graph doesn't care what the reference is.
It is a pure ratio plot. Its unitless. Do you mean a *noise* graph in
say dBr referenced to nV?

Then 0 dB-SPL = 2.9 nano PSI (rms) -- an unbelievably small value.

This means that since 1 atm = 14.7 PSI, it is equivalent to a loudness
level of 194 dB-SPL!

Ahmm...This is a DC value. It has no loudness level.


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.

 

[Pooh Bear]

Kevin Aylward wrote:

Pooh Bear wrote:
decibel Abbr. dB Equal to one-tenth of a bel. [After Alexander Graham
Bell.]

0 dBr An arbitrary reference level (r = re; or reference) that must be
specified. For example, a signal-to-noise graph may be calibrated in
dBr, where 0 dBr is specified to be equal to 1.23 Vrms (+4 dBu);
commonly stated as "dB re +4," that is, "0 dBr is defined to be equal
to +4 dBu."

This is a bad example. A S/N graph doesn't care what the reference is.
It is a pure ratio plot. Its unitless. Do you mean a *noise* graph in
say dBr referenced to nV?

I agree ( it's a bad example ).

I understand what the AES was trying to say but they explained it badly.

0 dBr can be referenced to *any* arbitrary operating level ! My Audio
Precision test set allows me to do just this actually !

You find a reference point ( of your choice ) - press 'set reference' and
all subsequent measurements will be referenced to the level that you set. -
until you cancel the dBr .

Choosing the 0 VU reference of + 4dBu was a very poor example of how best
to use dBr.

Graham

 

[Rene Tschaggelar]

Pooh Bear wrote:

[snip]

Forgotten :

dBc - decibels relative to the carrier

Rene

 

[Pooh Bear]

Rene Tschaggelar wrote:

Pooh Bear wrote:

[snip]

Forgotten :

dBc - decibels relative to the carrier

Rene

Ahhhh - I was hoping someone would explain dBc. Please continue !

I've seen stuff like " third intercept point = -51dBc ". Way out of my
frame of reference !

Would you care to elaborate for us ?

Graham

 

[Adrian Tuddenham]

Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:

decibel Abbr. dB Equal to one-tenth of a bel. [After Alexander Graham
Bell.]

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

That looks like a fairly up-to-date reference, I'm pleased to see that
it reiterates the 1929 definition of the decibel and there is no
indication that it has been re-defined as others have claimed.

2. The preferred method and term for representing the ratio of different
audio levels. It is a mathematical shorthand that uses logarithms (a
shortcut using the powers of 10 to represent the actual number) to
reduce the size of the number. For example, instead of saying the
dynamic range is 32,000 to 1, we say it is 90 dB [the answer in dB
equals 20 log x/y, where x and y are the different signal levels].

[... useful compendium of derived units snipped ...]

Note that *voltage* ( or pressure in the case of acoustic dBs )
reference related decibels are now the accepted norm.

They are the easiest things to measure; but if you use the above
formula to compare them, you have to assume equal impedances - even if
you don't have to worry about what they might actually be.

If you don't take account of the impedance levels, you can finish up
with nonsense, such as 'proving' that a transformer has gain or that the
output stage of a valve audio amplifier has no gain at all.

This is why I have been so pedantic about the subject. The derivation
of the present-day voltage useage of decibels from the fundamental power
definition (which is unchanged) seems to have been forgotten. This does
not matter for a lot of the simple op-amp designs which make up a huge
percentage of analogue engineering but there are times when the designer
must have this basic knowledge or he/she will get into a tangle.

The dBm ( 600 ohms ) is all but extinct in real usage.

It was always neglected in consumer items, but I doubt if
broadcasting, telecomms, AF and RF distribution and other line engineers
would agree.


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

 

[Rene Tschaggelar]

Pooh Bear wrote:

dBc - decibels relative to the carrier



Ahhhh - I was hoping someone would explain dBc. Please continue !

I've seen stuff like " third intercept point = -51dBc ". Way out of my
frame of reference !

Would you care to elaborate for us ?

Well,
dBc means dB in respect to the carrier. The "third intercept
point" specifies some nonlinearity in mixers.

google for "third intercept point" bring up :
http://rfdesign.com/mag/radio_efficiently_simulating_thirdorder/
http://www.maxim-ic.com/appnotes.cfm/appnote_number/749
http://tele-tech-rf.com/mixspec.htm

All of them better sum up what it means.

Rene
--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net

 

[James T. White]

"Pooh Bear" <rabbitsfriendsandrelations@hotmail.com> wrote in message
news:42C0DEBC.BF071D14@hotmail.com...

As for attennas, I suspect you're talking of directional gain ?

You are correct. They are used to describe the gain of an antenna. dBi is gain
relative to an isentropic radiator and dBd is gain relative to a simple dipole
radiator.

--
James T. White

 

[John Larkin]

On Tue, 28 Jun 2005 06:23:08 +0100, Pooh Bear
<rabbitsfriendsandrelations@hotmail.com> wrote:

"James T. White" wrote:

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:WZYve.421$0V3.386@newssvr13.news.prodigy.com...
Hello Graham,

dBm all over the place...

Plus dBi and dBd for antennas!
--
James T. White

Did I *not*say that the RF guys adopted the dB ? ;-)

Since you mention them - us LF Luddites ;-) would be quite interested as
to how you RF guys use the dB !

The measures you mention are new to me.

I'd heard of dBc - third order intercept ( whatever that is ! ) but it
never related to my area of interest so I didn't pay enough attention.

dBc is "power level relative to the carrier", commonly used to measure
spurs and such.

Third-order intercept is usually measured in dBm. That's the
extrapolated intersection point of the fundamental power with the 3rd
harmonic as the operating level varies around some nominal operating
point. It means something about linearity somehow.

As for attennas, I suspect you're talking of directional gain ?

There's always dBi, field strength relative to an isotropic radiator.

John

 

[Guy Macon]

James T. White wrote:

(nobody important) wrote:

As for attennas, I suspect you're talking of directional gain ?

You are correct. They are used to describe the gain of an antenna.
dBi is gain relative to an isentropic radiator and dBd is gain
relative to a simple dipole radiator.

That brings back memories. Junior high school basic
electronics class. In the same week, I was taught;

[1] Because some passive devices can appear to have gain if
all you do is measure the voltage or if you measure VA instead
of W (missing the effect of phase), the fellows who invented
the vacuum tube and the transistor made oscillator cirvcuits
to prove that there really was gain.

[2] Passive antennas have gain.

I had a good teacher, though; he explained all.

 

[John Perry]

Pooh Bear (and others) wrote:

600 ohms working was only ever needed for long 'land lines' - and how many
of
them still exist ? It's all concentrated at the local exchange and
distributed digitally via optic fibre these days !

Actually, in some sparsely populated rural US areas there are still the
ancient 600-ohm lines in use. You can recognize them by the bare wires
held up on glass insulators and twisted every few hundred feet by
dropping one to a lower crosstie so it can swap places on the next
crosstie with its return wire. They're now used mainly for frequency-
division multiplexing (FDM), I think, but there are still a few out there.

I *know* ! It's actually difficult to make a practical twisted pair that *isn't*
around 100 ohms at HF !

In fact, twisted pair is around 100 ohms from DC to microwave, and
probably beyond (I'm not an RF person, so I don't know whether losses
influence characteristic impedance at EHF and light frequencies). Its
characteristic impedance is unimportant at audio and VLF frequencies
because the reflection time is usually so short that reflections are an
insignificant part of the total signal. Slew rate of the signal is too low.

The true nature of twisted-pair's characteristic impedance becomes
obvious when you apply a TDR (time-domain reflectometer). These devices
use variations in the line's characteristic impedance to find crimps,
breaks, shorts, etc. in a long transmission line.

They are also important to FDM on both 600- and 100-ohm lines, because
improperly terminated long lines can have enough reflection to corrupt
the higher-frequency multiplexed channels.

And they were important to teletype (is this even used any more?)
because improperly terminated 100-ohm twisted pair limited the baud rate
possible over long subscriber lines.

Part of my job 35 years ago as a toll office tech was troubleshooting
these things . So some details may be distorted by faulty memory and
subsequent education. But the gist is correct.

John Perry

 

[Joerg]

Hello Adrian,

The dBm ( 600 ohms ) is all but extinct in real usage.

It was always neglected in consumer items, but I doubt if
broadcasting, telecomms, AF and RF distribution and other line engineers
would agree.

They don't. But Graham has a point in that impedance must always be
mentioned here. Us RF guys assume the world is, by default, living in a
50 ohm range. Phone guys assume 600 ohms. (Graham: POTS ain't dead, not
at all).

I take all these blunt predictions with a grain of salt. I remember when
lots of EE magazine articles predicted the death of the CD4000 series
more than 20 years ago. Heck, even a sales engineer from a major semi
mfg tried to convince me of that recently. That was and is baloney. When
I look at my designs from a parts cost times production volume point of
view I wouldn't be surprised if CD4000 chips would be the big ticket line.

Regards, Joerg

http://www.analogconsultants.com

 

[Guy Macon]

John Perry wrote:

In fact, twisted pair is around 100 ohms from DC to microwave,

Are you sure you don't want to change "DC" to some low AC frequency?
At DC the insulation resistance is a bit higher than 100 Ohms...

 

[Pooh Bear]

John Larkin wrote:

On Tue, 28 Jun 2005 06:23:08 +0100, Pooh Bear
rabbitsfriendsandrelations@hotmail.com> wrote:

I'd heard of dBc - third order intercept ( whatever that is ! ) but it
never related to my area of interest so I didn't pay enough attention.


dBc is "power level relative to the carrier", commonly used to measure
spurs and such.

Ahhh - Ok !

Third-order intercept is usually measured in dBm. That's the
extrapolated intersection point of the fundamental power with the 3rd
harmonic as the operating level varies around some nominal operating
point. It means something about linearity somehow.

I kinda guessed something to that effect. Not my field, so not important to
me at this time but thanks for the explanation.

As for attennas, I suspect you're talking of directional gain ?

There's always dBi, field strength relative to an isotropic radiator.

I wonder what the dBi of CH or CHL was ? Presumably it's all pattern
dependent ?


Graham