[GreenKeys] Far OT: audio inputs
Gerry Block
gblock at sbcglobal.net
Mon Feb 16 14:59:15 EST 2009
Oops..
In my response I said '0' level at 1.23vrms... I meant .775vrms...
Gerry
-----Original Message-----
From: greenkeys-bounces at mailman.qth.net
[mailto:greenkeys-bounces at mailman.qth.net] On Behalf Of Gerry Block
Sent: Monday, February 16, 2009 10:06 AM
To: gil at baudot.net; eugene at hertzmail.com
Cc: greenkeys at mailman.qth.net
Subject: Re: [GreenKeys] Far OT: audio inputs
Gil,
I wanted to add a couple of small facts which might interest the net. As
far as transformers go, in professional use they were always pretty good -
in fact the debate still rages over transformers vs direct. If you want an
idea of how good transformers can be, see as an example
http://www.jensentransformers.com/datashts/11ssp6m.pdf
They have inherently good common mode rejection ratio (can reject hum in
balanced configuration) whereas with opamps you can get it good but only
using very expensive matched components. Its easier to count the turns on
the transformer.
Transformers are often preferred in an RF environment as they are
essentially immune from RF fields where opamp inputs are not necessarily
unless they have good rf filters. My guess is the inputs on the Harris unit
are probably filtered as it is RF equipment.
And, as you indicated, almost required in long distance transmissions that
are not digital.
Regarding line level, I probably can add to the confusion. If you take
out a voltmeter, +4dBm and +4dBu at '0' level read the same - 1.23vRMS.
What happened a bunch of years ago is someone said "zero reference is 1mw
into 600 ohms and we don't use 600 ohms anymore..." (this was the 'm' in
dBm) and so they kept the same reference voltage level and changed the
suffix to dBu (I'll guess: the 'u' means 'unloaded'). There is also a dBv
where the zero reference is 1vRMS. So for all practical purposes dBm and
dBu are the same if you are using a voltmeter to measure the level, which is
what everybody does. A Vu meter is really a voltmeter with special
ballistics and scale.
I believe the reason for '+4' had to do with the use of the VU meters
themselves. These meters had a rather low impedance and also had a
rectifier in them. With an actual 600 Ohm source (i.e. a transformer output
from a vacuum tube amplifier) they would a) load the line slightly and b)
could cause not just measurable but audible distortion. You could easily
hear it listening to a tone from an oscillator. So a 3600 ohm resistor was
added in series with the meter to reduce the load on the 600 ohm line to
something around 6000 ohms (10:1 rule of thumb). This also had the nice
side effect that if you used a small potentiometer instead of the 3600 ohm
resistor, in a large facility you could calibrate each Vu meter to a precise
accuracy. That is why you will see many Vu meter panels from the old days
with little brackets and small AB pots on the back of them.
The standard for 'professional' audio equipment was a maximum (before
onset of clipping) output of +24dBm, which for a signal registering '0' on
the Vu meter gave 20db of headroom before clipping for signal peaks.
Plenty.
In some instances, for instance with old remote equipment, they often had
the ability to dial in a zero reference level of something other than +4..
for instance I have a Collins 12Z 'remote mixer' which lets you set the zero
reference on the Vu meter to +4 - or +6, +8, +10 etc... This was because
these remote systems were used to drive actual telephone leased lines set up
for the remote broadcasts, and they might need to drive a higher signal
level (at the expense of some potential distortion on the louder signals) to
get rid of crosstalk and noise at the receiving end.
Over the years to make the professional systems more perfect (on paper)
transformers were removed from audio systems and Vu meter isolation amps
(using opamps) were built, finally replaced by electronic Vu meters..
Not that this had anything good to do with the sound quality, which has
continued, mostly, to get worse and worse. This has resulted in a robust
'retro movement' back to vacuum tubes and transformers, a marketers dream
but which misses the point. The problem wasn't the equipment, it was lack
of training. That is another story.
Regards
Gerry Block
Ad6MC
-----Original Message-----
From: greenkeys-bounces at mailman.qth.net
[mailto:greenkeys-bounces at mailman.qth.net] On Behalf Of gil at baudot.net
Sent: Sunday, February 15, 2009 10:54 PM
To: eugene at hertzmail.com
Cc: greenkeys at mailman.qth.net
Subject: Re: [GreenKeys] Far OT: audio inputs
Hey Eugene:
I am curious to see posts about this as well. I don't have your
answers, but I can add a couple of things. I always wondered where the
600-ohm standard evolved from -- anybody? It seems too high to be the
characteristic impedance of a cable; I expect that it is a practical
value for transformer design, but undoubtably there will be some other
reason. Bell telephone perhaps?
As for the inputs on your gear: a transformer input allows you to
connect an audio signal and keep both sides electrically isolated from
your chassis -- a balanced signal input with neither side referenced to
ground. This is used to prevent a ground loop between systems, which
can result in a hum superimposed on the audio. I'm not sure what was
generally done with the center tap in those applications, but I presume
you could leave it open, or tie it to ground (via a short, resistor, or
capacitor; whichever helped). I'd like to hear from folks who used this
type of gear, with their experiences. A 600-ohm transformer input is
driven from a 600-ohm source (active, or transformer output), not unlike
50-ohm antenna lines or 75-ohm video signals (except that those coax
lines are unbalanced). Your differential-amplifier inputs sound like
they are also balanced inputs, but I can't suggest why one would be
better than the transformer input on that gear of yours.
I can add a bit of information related to unbalanced and balanced audio
equipment, and used in consumer and pro-audio applications:
A ground-loop hum may be noticed when an unbalanced-audio output from a
consumer audio source (eg: cd player) is connected to
amplifiers/speakers in a distant room. A line-level coaxial cable is
connecting the source and the amps, which are on different circuit
breakers. These ground-loop problems are caused by different ground
potentials (perhaps only
millivolts rms) of the two AC power circuits feeding the two systems
(source/amplifier). When you connect these systems
together with an unbalanced line-level coaxial cable, the outer shield
of the cable connects to each chassis, and hence to the two AC grounds,
resulting in a small AC current flow in the ground shield. In an
unbalanced signal connection, this AC
current flow in the shield becomes an audible signal: the 60-cycle hum
you hear (and harmonics: 120-Hz...) In Europe and other places that use
50-Hz power you are, of course, hearing 50-Hz, 100-Hz...).
You can convert an unbalanced signal to balanced, and this is just what
is done with pro-audio gear, where you see the three-pin XLR connectors
on everything. Transformers can be used in this application, but
generally don't have enough bandwidth or linearity for a high-fidelity
application, so active systems are generally used (transformers are just
fine for voice use). An active system can also use a specific
source/termination impedance (like 600-ohms), or can be a
low-Z-source/hi-Z-termination design (which has less loss).
Balanced audio cables are usually twisted pair, either unshielded (like
100-ohm CAT-5 cable), or shielded twisted pair, in which case the shield
is usually connected only to the ground at the receiving end of the
cable.
Unbalanced and balanced audio systems use different references for their
nominal volume level:
The nominal level for unbalanced signals (eg: consumer cd-player) is -10
dBV. The unit dBV, is a voltage level with respect to 1 Vrms. A nominal
-10 dBV signal is 0.316 Vrms from center pin to ground (-10 =
20log(0.316/1)). The nominal level for balanced (pro-audio) signals is
+4 dBu. The unit dBu, is a voltage level with respect to 0.775 Vrms. A
nominal +4 dBu signal is 1.23 Vrms from + to - terminals (+4 =
20log(1.23/0.775)). Why 0.775V ref? I have no idea.
An older reference is dBm, which is 1 mW into 600-ohms -- this came from
days when balanced signals used 600-ohm transformer coupling, a lossy
conversion (which also limits bass response). Sometimes dBm is used in
specs of balanced systems
that are not 600-ohm impedance -- this is incorrect. Transformer-coupled
systems use 600-ohm output and input impedances, for maximum power
transfer over these balanced links. Most electronic balanced audio
systems today use a low output impedance and a high input impedance, for
minimum voltage loss, and amplify/attenuate as needed for the
conversion, that is, -10 dBV is amplified to +4 dBu, and vice-versa.
This is not a 14-dB difference -- since there are different reference
levels, there is electrically just an 11.8 dB difference.
It is easy to use opamps to build balanced drivers and receivers for
audio; one I did drives well over 1000-feet of CAT-5 with virtually no
distortion or noise.
Buy you may not need to convert signals. You PC output is unbalanced
(the computer ground is the shield of the coax) and typically sees a
load of perhaps 5K or greater -- if you connect this to a 600-ohm
transformer input, you will get the benefit of the balanced input since
the transformer's input side is floating. The PC needs to be able to
drive the 600-ohm impedance, which it likely can, or you could insert
some resistance in the line to lighten the load on the PC's driver.
Whether there will be a compatible signal level is another story, but I
would guess that you would be able to adjust the PC output to a decent
volume to drive your gear. If the PC cannot drive a 600-ohm
transformer, you will need to insert a small preamp, but again, it does
not necessarily need to have a balanced output if you are connecting to
a floating transformer.
Your microphones are likely a different matter, and one that I can't
comment on -- but I presume that some sort of preamp will be needed to
drive a 600-ohm input.
FWIW,
gil
greenkeys moderator
gil at baudot.net
www.baudot.net
Vaux Electronics: 480-354-5556
> -------- Original Message --------
> Subject: [GreenKeys] Far OT: audio inputs
> From: <eugene at hertzmail.com>
> Date: Sun, February 15, 2009 5:41 pm
> To: <greenkeys at mailman.qth.net>
>
> I have an exciter (rf-1310) that has several audio inputs. Any audio
> input can be selected for any type of mode (two different inputs are
> selected for ISB mode).
>
> Two of the inputs are transformer input with 3 connections. 2 outside
> and one center tap connection.
>
> Three of the inputs have differential amplifier inputs (op amps?) with
> only two connections (basically inputs to the inverted and non-inverted
> inputs to the op amp)
>
> All inputs seem identical in terms of expected input levels (0dbm
> nominal into 600 ohms). So why would I choose one type of input over the
> other? Whats the benefit/difference/proper application?
>
> The inputs I am interested in using are
>
> a) PC audio out
>
> b) A Sure 55S dynamic mic with High (100K ohms) Medium (150-250K
> ohms) and Low (30-50) impedence selection
>
> c) A D104 mic non-amplified
>
> Do I need an amplifier beteween the mics and the 600 ohm 0dbm inputs of
> this exciter?
>
> So what do I need to know about audio inputs?
>
> Thanks
>
> Eugene
______________________________________________________________
GreenKeys mailing list
Home: http://mailman.qth.net/mailman/listinfo/greenkeys
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:GreenKeys at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
GreenKeys mailing list
Home: http://mailman.qth.net/mailman/listinfo/greenkeys
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:GreenKeys at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
More information about the GreenKeys
mailing list