[sci.electronics] FM audio recording?

rees@apollo.uucp (Jim Rees) (09/14/87)

(Maybe this belongs in rec.audio, but I don't want to trigger a
long discussion.  I just want the facts, ma'am.)

How come FM was never widely used for audio recording?  It seems like
FM would be the ideal way to overcome the inherent non-linearities
and noise of magnetic tape.  You could do away with the bias circuitry
and the sensitive adjustments that depend on the type of tape.  It
would even be relatively insensitive to saturation and variations
in tape output.  And erasing would be incredibly easy.  You wouldn't
even need a separate erase head, I claim!  It seems so obvious, there
must be something wrong with it.  FM has long been used for
instrumentation and video recording for these reasons.

adam@cunixc.columbia.edu (Adam J. Kucznetsov) (09/15/87)

In article <3746d52b.b8ab@apollo.uucp> rees@apollo.uucp (Jim Rees) writes:
..
>How come FM was never widely used for audio recording?

Bandwidth, bandwidth, y'know. Audio tape recorders have been around for
a long time, and they're pretty standardized (since there isn't that
much of a 'format' to standardize!) but the more complex circuitry and
the mechanisms to deal with the higher-bandwidth information transfer
from and to magnetic tape probably never caught on in the audio field.
Now we have video recorders and this funny DAT system too... maybe
there's something yet to be done? :-) adam
-- 
---------------------------------------------------------------------------
Cat (Adam) J. Kucznetsov	adam@cunixc.columbia.edu and cunixc.UUCP
Columbia University, NYC	UI.ADAM@CU20B.BITNET  AJUUS@CUVMA.BITNET
     When in danger, when in doubt, run in circles, scream and shout.

john@hpcvlo.HP.COM (John Eaton) (09/15/87)

<<<<
< How come FM was never widely used for audio recording?  It seems like
< FM would be the ideal way to overcome the inherent non-linearities
< and noise of magnetic tape.
<
----------
Because you would wind up with a freq response of DC-100hz with consumer
grade equipment.



John Eaton
!hplabs!hp-pcd!john

ron@topaz.rutgers.edu (Ron Natalie) (09/15/87)

What do you think the HiFi Videotape sound is?  The major problem is
that by the time recording gear capable of doing FM speed recording
on a resonable amount of tape and for resonable cost, Digital recording
already had a jump on the industry.

-Ron

sjb@piglet.UUCP (Seth J. Bradley) (09/16/87)

> How come FM was never widely used for audio recording?


I believe HiFi VCRs record the audio portion in Frequency Modulation.
The problem with recording audio in FM on a fixed head deck is the
ungodly bandwidth FM requires.  15KHz stereo broadcasts use about
160 KHz of bandwidth, and it would be difficult, if not impossible,
to get that kind of frequency response on a fixed head deck.  And
as long as you have a rotating head, why not go to DAT?

jgh@root.co.uk (Jeremy G Harris) (09/16/87)

In article <3746d52b.b8ab@apollo.uucp> rees@apollo.uucp (Jim Rees) writes:
>How come FM was never widely used for audio recording?

You lose out so much on the bandwidth that the tape costs would be
exorbitant. FM trades performance across a noisy channel for bandwidth
of the encoded signal. Audio-quality FM is wideband. Narrow-band FM
is occasionally used in comms, but is just as susceptible as AM to
noise (TANSTAAFL). The encoded bandwidth is something like
2B + D where B= encoding signal bandwidth  and  D= deviation

Jeremy
-- 
Jeremy Harris			jgh@root.co.uk

king@dciem.UUCP (Stephen King) (09/16/87)

In article <3746d52b.b8ab@apollo.uucp> rees@apollo.uucp (Jim Rees) writes:
>How come FM was never widely used for audio recording?  It seems like ...

Beta Hi-Fi uses audio frequencey modulation for the hifi sound tracks...sjk




-- 
 * Defence & Civil Institute *		...!utzoo!dciem!king 
 * of Environmental Medicine *		Stephen J King
- Simulation & Training Group -		(416) 635-2149

plate@dicome.UUCP (Douglas B. Plate) (09/17/87)

In article <231@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
)) How come FM was never widely used for audio recording?
)
)
)I believe HiFi VCRs record the audio portion in Frequency Modulation.
)The problem with recording audio in FM on a fixed head deck is the
)ungodly bandwidth FM requires.  15KHz stereo broadcasts use about
)160 KHz of bandwidth, and it would be difficult, if not impossible,
)to get that kind of frequency response on a fixed head deck.  And
)as long as you have a rotating head, why not go to DAT?

Just to clarify:
"An FM broadcast channel is 200kHz wide."  
-Electronic Communication,  Robert L. Shrader  (my tech school text book).

		Doug Plate

sjb@piglet.UUCP (Seth J. Bradley) (09/19/87)

In article <1574@dicome.UUCP>, plate@dicome.UUCP (Douglas B. Plate) writes:
> In article <231@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
> )ungodly bandwidth FM requires.  15KHz stereo broadcasts use about
> )160 KHz of bandwidth, and it would be difficult, if not impossible,
> Just to clarify:
> "An FM broadcast channel is 200kHz wide."  
> -Electronic Communication,  Robert L. Shrader  (my tech school text book).
> 
> 		Doug Plate

I cannot find my reference just now, and perhaps the radio buffs out
there can clarify.  Although FM broadcast channels are spaced 200KHz
apart(which is obvious just looking at the dial)  there exist in that
channel guard bands, pilot signals, and subbands.  I believe(and if anyone
out there knows differently they are welcome to correct me) that the
actual audio portion of the main FM band takes up around 160 KHz.

ken@cs.rochester.edu (Ken Yap) (09/21/87)

|I cannot find my reference just now, and perhaps the radio buffs out
|there can clarify.  Although FM broadcast channels are spaced 200KHz
|apart(which is obvious just looking at the dial)  there exist in that
|channel guard bands, pilot signals, and subbands.  I believe(and if anyone
|out there knows differently they are welcome to correct me) that the
|actual audio portion of the main FM band takes up around 160 KHz.

Actually, if I remember all those Bessel functions they taught us, FM
results in infinite sidebands but attenuates rapidly away from the
carrier so for practical purposes the bandwidth is decreed to be the
width at which the sideband drops to X dB below the carrier, where
X is something like 75.

	Ken

ron@topaz.rutgers.edu (Ron Natalie) (09/21/87)

Just to clarify:
  Yes, a FM broadcast channel is 200KHz wide, That is convenience (keeps
the frequencies from getting into wierd fractions).  FM bandwidth is
dependent on the amplitude of the modulated signal.  It's nice to keep
your signal amplitude limitted so that you stay within your own channel.
160KHz is not an unreasonable number.

-Ron

plate@dicome.UUCP (Douglas B. Plate) (09/22/87)

In article <235@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
)In article <1574@dicome.UUCP), plate@dicome.UUCP (Douglas B. Plate) writes:
)) In article <231@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
)) )ungodly bandwidth FM requires.  15KHz stereo broadcasts use about
)) )160 KHz of bandwidth, and it would be difficult, if not impossible,
)) Just to clarify:
)) "An FM broadcast channel is 200kHz wide."  
)) -Electronic Communication,  Robert L. Shrader  (my tech school text book).
)) 
)) 		Doug Plate
)
)I cannot find my reference just now, and perhaps the radio buffs out
)there can clarify.  Although FM broadcast channels are spaced 200KHz
)apart(which is obvious just looking at the dial)  there exist in that
)channel guard bands, pilot signals, and subbands.  I believe(and if anyone
)out there knows differently they are welcome to correct me) that the
)actual audio portion of the main FM band takes up around 160 KHz.

OK, I should of included the entire reference from the book.  Your
very close.  The sound itself deviates 75kHz either side of center
frequency at 100% modulation (150kHZ) and there is a 25kHz guard band
on each side (another 50kHz=200kHz total).  I guess I zoomed in on
the words "stereo broadcasts use" and forgot that that we were actually
talking about FM being put on to tape.  Sorry.  
		Doug Plate
(info given here was taken from the same book sited above)

plate@dicome.UUCP (Douglas B. Plate) (09/22/87)

Newsgroups: sci.electronics
Subject: Re: FM audio recording?
Summary: 
Expires: 
References: <3746d52b.b8ab@apollo.uucp> <231@piglet.UUCP> <1574@dicome.UUCP> <235@piglet.UUCP>
Sender: 
Reply-To: plate@dicome.UUCP (Douglas B. Plate)
Followup-To: 
Distribution: 
Organization: DICOMED Corp., Minneapolis
Keywords: 

In article <235@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
)In article <1574@dicome.UUCP), plate@dicome.UUCP (Douglas B. Plate) writes:
)) In article <231@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
)) )ungodly bandwidth FM requires.  15KHz stereo broadcasts use about
)) )160 KHz of bandwidth, and it would be difficult, if not impossible,
)) Just to clarify:
)) "An FM broadcast channel is 200kHz wide."  
)) -Electronic Communication,  Robert L. Shrader  (my tech school text book).
)) 
)) 		Doug Plate
)
)I cannot find my reference just now, and perhaps the radio buffs out
)there can clarify.  Although FM broadcast channels are spaced 200KHz
)apart(which is obvious just looking at the dial)  there exist in that
)channel guard bands, pilot signals, and subbands.  I believe(and if anyone
)out there knows differently they are welcome to correct me) that the
)actual audio portion of the main FM band takes up around 160 KHz.

OK, I should have included the entire reference from the book.  You're
very close.  The sound itself deviates 75kHz either side of center
frequency at 100% modulation (150kHZ) and there is a 25kHz guard band
on each side (another 50kHz=200kHz total).  I guess I zoomed in on
the words "stereo broadcasts use" and forgot that that we were actually
talking about FM being put on to tape.  Sorry.  
		Doug Plate
(info given here was taken from the same book sited above)

john@tower.UUCP (John Moore) (09/25/87)

In article <1588@dicome.UUCP> plate@dicome.UUCP (Douglas B. Plate) writes:
>Newsgroups: sci.electronics
>Subject: Re: FM audio recording?
>Summary: 
>Expires: 
>References: <3746d52b.b8ab@apollo.uucp> <231@piglet.UUCP> <1574@dicome.UUCP> <235@piglet.UUCP>
>Sender: 
>Reply-To: plate@dicome.UUCP (Douglas B. Plate)
>Followup-To: 
>Distribution: 
>Organization: DICOMED Corp., Minneapolis
>Keywords: 
>
>In article <235@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
>)In article <1574@dicome.UUCP), plate@dicome.UUCP (Douglas B. Plate) writes:
>)) In article <231@piglet.UUCP> sjb@piglet.UUCP (Seth J. Bradley) writes:
>)) )ungodly bandwidth FM requires.  15KHz stereo broadcasts use about

Since the discussion is on bandwidth, let me point out that FM bandwidth
(especially wideband FM  such as commercial broadcasting) is not simply
the sum of the audio frequencies present, or even just the highest
frequency present. Rather, it is a complex set of bessel functions.
For a single tone, one gets an infinite series of sideband terms, the
amplitude of the Nth sideband equal to the Nth bessel function of
a relationship between the peak frequency deviation and the frequency
of the tone (sorry I'm being vague, but in a hurry and don't have time
to look up the exact formula).  For more than one tone, one gets multiple
series based on cross products of all the components, yielding a large
number of sidebands (more than you would expect from any linear analysis).
Thus, the "bandwidth" is infinite, but for practical reasons the
bandwidth is defined by when the sidebands drop below a certain level
relative to the carrier.
	Now, you might ask, why the 200Khz deviation? Well, it turns out
that in a noisy environment, the recovered signal is less noisy for
wider bandwidths. FM could have used a smaller bandwidth, and transmit
the data with the same fidelity, with the exception of signal-to-noise
ration.
	For further reference on this, look at the Radio Engineers
Handbook.

	John Moore
	Anasazi, Inc
	Somewhere in Phoenix, AZ

elg@usl (Eric Lee Green) (09/26/87)

in article <430@root44.co.uk>, jgh@root.co.uk (Jeremy G Harris) says:
> In article <3746d52b.b8ab@apollo.uucp> rees@apollo.uucp (Jim Rees) writes:
>>How come FM was never widely used for audio recording?
> 
> You lose out so much on the bandwidth that the tape costs would be
> exorbitant. FM trades performance across a noisy channel for bandwidth
> of the encoded signal. Audio-quality FM is wideband. 

Now, I'm no electronics genius, but even I know that television
signals have a huge bandwidth requirement... if we can store
television signals, why not FM audio signals, which have a much
narrower bandwidth requirement?

Someone else mentioned that the problem was that digital, a superior
technology, came along at the same time as FM audio became technically
feasible. I must agree somewhat, when it comes to consumer equipment,
as VCRs have only recently come down to a decent price, and tapes for
VCRs are still quite expensive compared to audio tape. Still, in
studio settings, I fail to see why they didn't move to FM equipment
long ago... after all, television studios have long used videotape,
long before consumers could afford them... if the studios had gone to
that superior technology, we wouldn't have to be putting up with the
tape hiss that is so evident when you listen to vintage recordings of
the '60s and '70s off of CD.

But then again, the music industry, in the U.S. at least, is totally
uninterested in progress and innovation... witness their struggle to
squash digital audio tape in the cradle. It's remarkable that they
even put out stuff on CDs. One reason they may have even consented to
put out stuff in CD format, was because they hoped to lure customers
who'd otherwise go to home-taping of albums... after hearing the CD of
an albums, you just aren't satisfied with just having a taped copy of
it.

--
Eric Green  elg@usl.CSNET       from BEYOND nowhere:
{ihnp4,cbosgd}!killer!elg,      P.O. Box 92191, Lafayette, LA 70509
{akgua,killer}!usl!elg        "there's someone in my head, but it's not me..."

jnp@calmasd.GE.COM (John Pantone) (09/28/87)

(Eric Lee Green) writes:
...discussion of FM audio recording...
> One reason they may have even consented to
> put out stuff in CD format, was because they hoped to lure customers
> who'd otherwise go to home-taping of albums... after hearing the CD of
> an albums, you just aren't satisfied with just having a taped copy of
> it.

	Just a note: I just bought a "hi-fi stereo" VCR.  I have recorded
from a CD and the quality is startlingly good.  Far better than conventional
tape recording - to my ear.
-- 
These opinions are solely mine and in no way reflect those of my employer.  
John M. Pantone @ GE/Calma R&D, Data Management Group, San Diego
...{ucbvax|decvax}!sdcsvax!calmasd!jnp          jnp@calmasd.GE.COM

floyd@humu.UUCP (Robert W. Floyd) (10/01/87)

FM recording was originally developed for recording very low frequency
signals on instrumentation tape recorders.  A side advantage was its relatively high dynamic range
(approximately 46 dB back in the 60's).  This was much better than the
26 dB dynamic range of the analog recordings.  The record heads were
optimized to achieve the greatest possible bandwidth at a given speed, and this
was done at the expense of dynamic range.  An additional problem was that
this performance was achieved partially by making the recording heads very
thin, so that they had to be replaced every thousand hours or so.
	The net result was that by 1970 it cost approzimately $10,000 to 
purchase an FM recorder that could achieve an audio bandwidth of 20kHz at
60 inches per second, and achieve a dynamic range about 3 dB greater than
consumer equipment.  The Navy might pay that much for 3 dB, but the average
consumer wouldn't.  In the meantime, many engineers, particularly in the
telephone business, had realized that the information content of a signal
meant for human ears could be compressed since humans were sensitive to
relative, not absolute changes in amplitude.  The best known implementation
of this idea is Dolby, which can be done quite inexpensively.
	In summary, the advantage of FM has been dynamic range.  The cost
	has been bandwidth, which is relatively expensive, in terms of
	heads, tape speed, mechanical complexity, etc.  Since effective methods
	such as Dolby and digital recording have been developed to
	improve dynamic range, it is unlikely that FM audio recording will
	every be commercially viable.