Stupid audio transformer question

[Andre Majorel]

Been thinking that, instead of getting transformer versions of
everything, I could make myself a transformer box and get the
transformer sound when I want it without having to pay for
transformers on every eq and/or desk channel.

I'm not sure how you go about that, though. Is it as simple as a
1:1 transformer ? Typical use case is unbalanced line output
into unbalanced line input.

You may think I'm confused. You may be right.

--
André Majorel http://www.teaser.fr/~amajorel/
J'ai des droits. Les autres ont des devoirs.

[Joe Stavitsky]

There are many prebuilt transformer boxes/adapters. Some of them have
circuit diagrams on the housing/in the documentation. Comparing it
with publically available wiring diagrams via google images should
yield useful results. Also I believe that National Instruments
Electronics Workbench has a public diagram database.

I am not entirely sure what you mean by "transformer sound". AFAIK the
purpose of a transformer is to suppress some noise that occurs in the
cable and/or connector, but I could be wrong.

IMO the prevasive availability of cheap transformer adapters makes
making one yourself useful only as a learning experience. Prices
should not exceed $25.

Best

Joe

[jwvm]

On Aug 27, 7:00*pm, Andre Majorel > wrote:
> Been thinking that, instead of getting transformer versions of
> everything, I could make myself a transformer box and get the
> transformer sound when I want it without having to pay for
> transformers on every eq and/or desk channel.

What do you mean by transformer sound? Transformers can be much more
nonlinear than other electronic components but that is not likely to
be useful. If you do search in this newsgroup, you can find posts by
Scott Dorsey about eliminating transformers in some microphones for
better sound.

>
> I'm not sure how you go about that, though. Is it as simple as a
> 1:1 transformer ? Typical use case is unbalanced line output
> into unbalanced line input.

You might want to do that to keep grounds separated but generally this
approach would not provide a significant benefit and could be
detrimental. A better case would be a transformer to convert a
balanced output to an unbalanced input. However, even really
inexpensive sound equipment has balanced and unbalanced inputs on each
channel.

>
> You may think I'm confused. You may be right.

Perhaps one could say that there are better ways to spend money for
higher sound quality than using additional transformers.

>
> --
> André Majorelhttp://www.teaser.fr/~amajorel/
> J'ai des droits. Les autres ont des devoirs.

[joe h]

This is way beyond my pay grade, but I do know that Lundahl
transformers are prized in the audio circuit community.

If you're going for a transformer sound, it probably makes sense to
get a good one with high-quality specs.

[Andre Majorel]

On 2011-08-28, joe h > wrote:

> This is way beyond my pay grade, but I do know that Lundahl
> transformers are prized in the audio circuit community.

> If you're going for a transformer sound, it probably makes
> sense to get a good one with high-quality specs.

Yes, Lundahl, Sowter, Jensen... Or maybe some cheap crap.
Anything and everything as long as I like the effect. But before
I worry about make and model, I need to know a little more about
the theory.

My understanding is that, for a 1:1 turns ratio, each
source/load sees the impedance on the other winding. So this
ought to work :
_ ___ ___ _
From unbal output | | V )||( V | | To unbal input
Z_out = 0.1 ~ 1 k |_| )||( |_| Z_in = 10 ~ 100 k
|______)||(______|

Am I on the right track ?

--
André Majorel http://www.teaser.fr/~amajorel/
J'ai des droits. Les autres ont des devoirs.

[Mike Rivers]

On 8/28/2011 11:46 AM, joe h wrote:

> If you're going for a transformer sound, it probably makes sense to
> get a good one with high-quality specs.

I would think that if you're going for a "transformer sound"
you should get a crummy transformer. A high quality
transformer doesn't really have a sound.

My definition of "transformer sound" is the distortion that
people think a transformer imparts. Same for "tube sound"
only substitute "tube" for "transformer." And you can
probably extrapolate what I think "analog tape sound" is.


--
"Today's production equipment is IT based and cannot be
operated without a passing knowledge of computing, although
it seems that it can be operated without a passing knowledge
of audio." - John Watkinson

http://mikeriversaudio.wordpress.com - useful and
interesting audio stuff

[PStamler]

On Aug 28, 3:10*pm, Andre Majorel > wrote:
> On 2011-08-28, joe h > wrote:
>
> > This is way beyond my pay grade, but I do know that Lundahl
> > transformers are prized in the audio circuit community.
> > If you're going for a transformer sound, it probably makes
> > sense to get a good one with high-quality specs.
>
> Yes, Lundahl, Sowter, Jensen... Or maybe some cheap crap.
> Anything and everything as long as I like the effect. But before
> I worry about make and model, I need to know a little more about
> the theory.
>
> My understanding is that, for a 1:1 turns ratio, each
> source/load sees the impedance on the other winding. So this
> ought to work :
> * * * * * * * * * * * _ * ___ * *___ * _
> * From unbal output *| | V * )||( * V | | *To unbal input
> * Z_out = 0.1 ~ 1 k *|_| * * )||( * * |_| *Z_in = 10 ~ 100 k
> * * * * * * * * * * * |______)||(______|
>
> Am I on the right track ?

Yes. The thing is, what you're probably after with "transformer sound"
is the sound imparted to signals by the high-quality professional-
grade transformers of the 1950s and 1960s, as heard in the recordings
of those days. So you probably *don't* really want cheap crap (which
will sound awful), but decent-quality older-design transformers, not
the modern transformers which impart almost no sound to signals which
through them. Look for UTC or Triad transformers, or buy some old,
good quality gear and dismantle it. Expect to pay some money, though;
people know now what good transformers are worth.

Oh, you can still get new Triad and UTC transformers, but plan to pay
a lot.

Peace,
Paul

[Trevor]

"joe h" > wrote in message
...
> If you're going for a transformer sound, it probably makes sense to
> get a good one with high-quality specs.

Surely if you actually want a transformer to *change* the "sound", you DON'T
want a really good one that won't! :-)

Trevor.

[William Sommerwerck]

"Trevor" > wrote in message
...
> "joe h" > wrote in message
> ...

>> If you're going for a transformer sound, it probably makes sense to
>> get a good one with high-quality specs.

> Surely if you actually want a transformer to *change* the "sound", you
DON'T
> want a really good one that won't! :-)

I think he's suggesting that older, "good" transformers somewhat rounded-off
the sound, in a euphonic way.

[jersey123]

In the New York Giants, there is a severe competition in the wide receiver position, between Victor Cruz and Domenik Hixon. Wearing Cheap NFL Jerseys (http://www.ucnoqta.com/), Cruz has earned a reputation as Mr. August after making big plays in the past two preseason games. Victor Cruz was originally a backup wide-out. However, Steve Smith has left to wear Steve Smith Colts jersey, so Big Blue is hoping for a little more consistency from the second-year wide-out. Therefore, we see Victor Cruz is competing with Domenik Hixon for the vacant slot receiver job. Of course, wearing Wholesale Jerseys (http://www.ucnoqta.com/), Hixon has eben the most productive of the Giants receivers, catching six passes for 62 yards less-than-crisp route running. Nevertheless, Victor Cruz has frustrated coaches with his less-than-crisp route running with his habit of acrobatic catches. Of course, it’s clear that Victor Cruz has much room to improve to compete with the outstanding receiver Domenik Hixon. However, no matter what they are competing, their primary job is to receive the ball from Eli Manning, who wears Eli Manning Giants jersey. And what’s more important, whichever one can gain a rapport with Eli to have a good chance to play in the regular season.

[Trevor]

"William Sommerwerck" > wrote in message
...
>>> If you're going for a transformer sound, it probably makes sense to
>>> get a good one with high-quality specs.
>
>> Surely if you actually want a transformer to *change* the "sound", you
> DON'T
>> want a really good one that won't! :-)
>
> I think he's suggesting that older, "good" transformers somewhat
> rounded-off
> the sound, in a euphonic way.


Well by my definition a transformer that *does* change the sound is NOT a
particularly "good" one. A good transformer does the job it's designed for
totally transparently.
Far easier to use software these days to provide almost any style of audible
change that YOU might personally deem "euphonic", since like beauty, that's
in eye, or ear, of the beholder.

Trevor.

[Arny Krueger[_4_]]

"William Sommerwerck" > wrote in message
...
> "Trevor" > wrote in message
> ...
>> "joe h" > wrote in message
>> ...
>
>>> If you're going for a transformer sound, it probably makes sense to
>>> get a good one with high-quality specs.
>
>> Surely if you actually want a transformer to *change* the "sound", you
> DON'T
>> want a really good one that won't! :-)
>
> I think he's suggesting that older, "good" transformers somewhat
> rounded-off
> the sound, in a euphonic way.

Audio transformers have a number of audible and measurable effects that are
in some sense failures to perform ideally. However, in their correct
situation they can be very helpful.

Their degree varies from in-your-face to probably inaudible.

(1) Low and high frequency cut-offs, for sure.
(2) Low and high frequency rises or bumps right before the roll-offs, maybe.
These can often be tuned with resistive loading.
(3) Nonlinear distortion, starting at the lowest frequencies and moving up
in frequency as you push the transformer harder with larger signals.
(4) Random noise, since the windings have resistance.
(5) Ambient noise, since transformers can pick up external magnetic fields
if not shielded with great care. Pull the "I" core off of an EI core
transformer and you have a pretty good pickup coil.

IME most comparable other form of allegedly euphonic distortion would be
magnetic tape which makes some sense since they are both based on
magnetizing and demagnetizing ferrous materials. However, the best
transformers seem to vastly outperform the best analog tape if your ideal is
no distortion of any kind.

The big plus to transformers is that they can blithely ignore massive common
mode voltages, and of course their potential for efficiency by matching
impedances. Their built-in bandpass filtering, especially at high
frequencies can also be a plus and is often intentionally exploited in
power transformers, input transformers and output transformers. For example
SP/DIF output transformers can make or break a piece of digital equipment's
ability to pass FCC Part 15 tests.

[mcp6453[_2_]]

On 8/29/2011 12:57 AM, PStamler wrote:
>
> Yes. The thing is, what you're probably after with "transformer sound"
> is the sound imparted to signals by the high-quality professional-
> grade transformers of the 1950s and 1960s, as heard in the recordings
> of those days. So you probably *don't* really want cheap crap (which
> will sound awful), but decent-quality older-design transformers, not
> the modern transformers which impart almost no sound to signals which
> through them. Look for UTC or Triad transformers, or buy some old,
> good quality gear and dismantle it. Expect to pay some money, though;
> people know now what good transformers are worth.
>
> Oh, you can still get new Triad and UTC transformers, but plan to pay
> a lot.

Like a lot of you, I've probably thrown a retirement's fund worth of
transformers into the dumpster. My question is, why are we not able to buy the
same transformers that were made in the fifties and sixties? It's not like we
don't have the technology. I understand that sales are dramatically lower, but
again, with regard to modern manufacturing capabilities, how hard is it to
select the right core, use the right wire, and wind the transformer? Is the
issue just one of cost?

[Scott Dorsey]

In article >,
Andre Majorel > wrote:
>Been thinking that, instead of getting transformer versions of
>everything, I could make myself a transformer box and get the
>transformer sound when I want it without having to pay for
>transformers on every eq and/or desk channel.

This may or may not be a good idea. Try it.

>I'm not sure how you go about that, though. Is it as simple as a
>1:1 transformer ? Typical use case is unbalanced line output
>into unbalanced line input.

http://www.edcorusa.com/products/9-m2m600-600.aspx

You may need to add a termination resistor depending on your load.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

joe h > wrote:
>This is way beyond my pay grade, but I do know that Lundahl
>transformers are prized in the audio circuit community.
>
>If you're going for a transformer sound, it probably makes sense to
>get a good one with high-quality specs.

Most of the Lundahl transformers are designed to be as transparent
as possible and not to have any sound. However, this isn't the case
for all of them....
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[hank alrich]

Trevor > wrote:

> "William Sommerwerck" > wrote in message
> ...
> >>> If you're going for a transformer sound, it probably makes sense to
> >>> get a good one with high-quality specs.
> >
> >> Surely if you actually want a transformer to *change* the "sound", you
> > DON'T
> >> want a really good one that won't! :-)
> >
> > I think he's suggesting that older, "good" transformers somewhat
> > rounded-off
> > the sound, in a euphonic way.
>
>
> Well by my definition a transformer that *does* change the sound is NOT a
> particularly "good" one. A good transformer does the job it's designed for
> totally transparently.
> Far easier to use software these days to provide almost any style of audible
> change that YOU might personally deem "euphonic", since like beauty, that's
> in eye, or ear, of the beholder.
>
> Trevor.

A desire to optimize signal levels where noise floors are an issue leads
to driving even the best transformers out of the range of their lowest
distortion characteristics. The alleged "sweetness" of the sound when so
treated is due to the euphonic characteristics of the distortion
products of the better transformers. This often does not apply to the
distortion products of the cheap stuff.

--
shut up and play your guitar * http://hankalrich.com/
http://www.youtube.com/walkinaymusic
http://www.sonicbids.com/HankandShaidri

[PStamler]

On Aug 29, 7:21*am, mcp6453 > wrote:
> On 8/29/2011 12:57 AM, PStamler wrote:
>
>
>
> > Yes. The thing is, what you're probably after with "transformer sound"
> > is the sound imparted to signals by the high-quality professional-
> > grade transformers of the 1950s and 1960s, as heard in the recordings
> > of those days. So you probably *don't* really want cheap crap (which
> > will sound awful), but decent-quality older-design transformers, not
> > the modern transformers which impart almost no sound to signals which
> > through them. Look for UTC or Triad transformers, or buy some old,
> > good quality gear and dismantle it. Expect to pay some money, though;
> > people know now what good transformers are worth.
>
> > Oh, you can still get new Triad and UTC transformers, but plan to pay
> > a lot.
>
> Like a lot of you, I've probably thrown a retirement's fund worth of
> transformers into the dumpster. My question is, why are we not able to buy the
> same transformers that were made in the fifties and sixties? It's not like we
> don't have the technology. I understand that sales are dramatically lower, but
> again, with regard to modern manufacturing capabilities, how hard is it to
> select the right core, use the right wire, and wind the transformer? Is the
> issue just one of cost?

There are two issues. First, transformer winding isn't that simple;
you have several windings carefully interleaved between and on top of
each other. It's a skilled act, and there aren't a lot of folks still
living who have the skills (and can teach them to a new generation).

Second, you used to find audio transformers in everything, from high-
quality professional equipment to crappy table radios. With a big
market, and millions of transformers being produced by the factories,
overhead was amortized over a wide base and transformers were cheap.
Now a company specializing in audio transformers sells maybe a few
hundred or thousand a year.

For both of those reasons, prices go up.

Peace,
Paul

[Mike Rivers]

On 8/29/2011 8:21 AM, mcp6453 wrote:

> My question is, why are we not able to buy the
> same transformers that were made in the fifties and sixties? It's not like we
> don't have the technology.

We sort of don't have the technology. Winding transformers
is a mechanical process that can't really be improved by
computers other than perhaps to count the turns and turn the
winding machine off at the right time. When I visited Manley
Labs several years back, EveAnna showed me their transformer
winding machine and it looked like a real Rube Goldberg
arrangement, requiring a fair amount of skill to set up and
wind one transformer. They use few enough transformers so
that they can wind their own and know that they're getting
what they need without having to place an order for 10 years
worth of transformers.

Since just about any company who builds audio gear employing
transformers, unless it's just a transformer as a gimmick or
in some non-critical path has their own custom design that
was arrived at through experimentation. Bill Putnam will
tell you that the transformers are an important part of the
sound of the UA compressors, and it was only because of a
lab notebook detailing winding experiments that they were
able to accurately reproduce the sound of the original
compressors in their 21st centruy re-creations.

> with regard to modern manufacturing capabilities, how hard is it to
> select the right core, use the right wire, and wind the transformer? Is the
> issue just one of cost?

I suppose if you copy someone else's design you can probably
reproduce the transformer, but which transformer do you
choose? Wanna start a "transformer sound" company and pick
one out?


--
"Today's production equipment is IT based and cannot be
operated without a passing knowledge of computing, although
it seems that it can be operated without a passing knowledge
of audio." - John Watkinson

http://mikeriversaudio.wordpress.com - useful and
interesting audio stuff

[Trevor]

"hank alrich" > wrote in message
...
> A desire to optimize signal levels where noise floors are an issue leads
> to driving even the best transformers out of the range of their lowest
> distortion characteristics. The alleged "sweetness" of the sound when so
> treated is due to the euphonic characteristics of the distortion
> products of the better transformers. This often does not apply to the
> distortion products of the cheap stuff.

Right, but trying to obtain a certain change in "sound" by trial and error
substitution of expensive transformers seems silly when similar experimental
changes can be made FAR more quickly at FAR less expense these days by
digital software processsing. Of course those who simply yearn for the "good
old days" can never be convinced any new technology is an improvement.

Trevor.

[John Hardy]

On 8/29/2011 8:05 PM, Mike Rivers wrote:
> We sort of don't have the technology. Winding transformers is a
> mechanical process that can't really be improved by computers other than
> perhaps to count the turns and turn the winding machine off at the right
> time. When I visited Manley Labs several years back, EveAnna showed me
> their transformer winding machine and it looked like a real Rube
> Goldberg arrangement, requiring a fair amount of skill to set up and
> wind one transformer.

I have not seen the winding equipment at Manley, but, generally speaking:

Some winding equipment is much more accurate and controllable (and
probably less Rube Goldbergish) than others. Audio transformers
typically use very fine wire for the windings. The more consistent and
tighter the tension is maintained on the fine wires (without breaking,
stretching or in any way damaging the wires), the more precise the
placement of each turn of wire will be and the more precise the
placement of each layer of wire and additional layers of wire will be.
This can make it possible to fit more turns of wire into a particular
winding space, allowing higher signal levels to be handled. Of course,
with more wire in a winding, the design needs to be tweaked because of
the added resistance of the longer wire and other factors. The higher
the consistency of the windings, the higher the consistency of the
performance of the transformers from one batch to the next. Some winding
equipment doesn't even accurately control the number of turns.

If the design side of things is state-of-the-art, and the winding
machinery is state-of-the-art, and the people doing the termination of
the wires and all the other detail work along the way know what they are
doing, you get a superior transformer. Whether or not that is the kind
of transformer a person wants is a separate issue.

John Hardy
The John Hardy Co.
www.johnhardyco.com

[Arny Krueger[_4_]]

"Trevor" > wrote in message
...
>
> "hank alrich" > wrote in message
> ...
>> A desire to optimize signal levels where noise floors are an issue leads
>> to driving even the best transformers out of the range of their lowest
>> distortion characteristics. The alleged "sweetness" of the sound when so
>> treated is due to the euphonic characteristics of the distortion
>> products of the better transformers. This often does not apply to the
>> distortion products of the cheap stuff.

> Right, but trying to obtain a certain change in "sound" by trial and error
> substitution of expensive transformers seems silly when similar
> experimental changes can be made FAR more quickly at FAR less expense
> these days by digital software processsing. Of course those who simply
> yearn for the "good old days" can never be convinced any new technology is
> an improvement.


Duplicating the FR of a classic transformer is very doable. Duplicating the
nonlinear distortion, not so much.

I know how to do frequency dependent nonlinear distortion of any order with
CEP and have done so, but actually doing it takes a lot of careful steps and
mouth holding,

The good news is that the FR changes are probably far and away the most
audible effects.


The hard-to-guess parts are the advantages of transformers when it comes to
common mode, and out-of-band rejection. YMMV, depending on the application.
While you can approximate the common mode rejection with active circuits,
the common mode dynamic range characteristics of transformers are in a
league of their own. My experience is that you rarely need either, but when
you do...

[Mike Rivers]

On 8/30/2011 12:03 AM, Trevor wrote:

> Right, but trying to obtain a certain change in "sound" by trial and error
> substitution of expensive transformers seems silly when similar experimental
> changes can be made FAR more quickly at FAR less expense these days by
> digital software processsing. Of course those who simply yearn for the "good
> old days" can never be convinced any new technology is an improvement.

No argument that some technology is an improvement, but we
have a hard time being convinced that it's a direct
replacement. As far as I know, nobody has accurately modeled
the audio qualities of a "famous" transformer. And I have to
chuckle about the new round of analog tape simulators.

The fact that we have more ways to modify sound does not
mean that we have ways of achieving the identical effect as
by using real components. We gain some, we lose some, we go
on. I see no reason to mimic the sound of an analog
recorder. If a plug-in that claims to do this gives the user
what he's looking for, that's fine. He probably doesn't know
what an analog recorder sounds like anyway.



--
"Today's production equipment is IT based and cannot be
operated without a passing knowledge of computing, although
it seems that it can be operated without a passing knowledge
of audio." - John Watkinson

http://mikeriversaudio.wordpress.com - useful and
interesting audio stuff

[Scott Dorsey]

In article >,
Mike Rivers > wrote:
>On 8/29/2011 8:21 AM, mcp6453 wrote:
>
>> My question is, why are we not able to buy the
>> same transformers that were made in the fifties and sixties? It's not like we
>> don't have the technology.
>
>We sort of don't have the technology. Winding transformers
>is a mechanical process that can't really be improved by
>computers other than perhaps to count the turns and turn the
>winding machine off at the right time. When I visited Manley
>Labs several years back, EveAnna showed me their transformer
>winding machine and it looked like a real Rube Goldberg
>arrangement, requiring a fair amount of skill to set up and
>wind one transformer. They use few enough transformers so
>that they can wind their own and know that they're getting
>what they need without having to place an order for 10 years
>worth of transformers.

But, we CAN buy the same transformers that were made in the fifties and
sixties. There are folks making them.

The only problem is that those crappy transformers that were cheap mass
production items with junk core alloys back in the fifties and sixties
are now hand-made with what have become exotic core materials.

Consequently, we get stuff like the original Pultec units, which used
inductors that were chosen because they were the cheapest possible coils
available.... it's possible to recreate them accurately, but making copies
of those coils runs you around $2k just in magnetics. For items that were
probably around $5 originally.

>I suppose if you copy someone else's design you can probably
>reproduce the transformer, but which transformer do you
>choose? Wanna start a "transformer sound" company and pick
>one out?

That's what people do. Plenty of folks making duplicates of classic
transformer designs out there. They're all very, very expensive, because
stuff that was made by the tens of thousands are now made by the dozens.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

Trevor > wrote:
>"hank alrich" > wrote in message
>> A desire to optimize signal levels where noise floors are an issue leads
>> to driving even the best transformers out of the range of their lowest
>> distortion characteristics. The alleged "sweetness" of the sound when so
>> treated is due to the euphonic characteristics of the distortion
>> products of the better transformers. This often does not apply to the
>> distortion products of the cheap stuff.
>
>Right, but trying to obtain a certain change in "sound" by trial and error
>substitution of expensive transformers seems silly when similar experimental
>changes can be made FAR more quickly at FAR less expense these days by
>digital software processsing. Of course those who simply yearn for the "good
>old days" can never be convinced any new technology is an improvement.

I buy equipment that sounds good, then I keep using it as long as it sounds
good. After a while, it either becomes valuable or not, but either way it
still sounds good.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Frank Stearns]

Mike Rivers > writes:

>on. I see no reason to mimic the sound of an analog

x1000 on that.

Working mostly in the classical and acoustic realm, the Holy Grail in my world was
removal of any sound changers, starting with the 12-20 transformers that were in the
complete signal path of a typical 1960s-1980s+ multitrack studio, and working
outward from there. (Removing/improving tape, with all its failings, was next.)

Having said that, there was something about the final sound of a typical
higher-end path of the day that was compelling.

But here's a modern trade secret: when you boil it all down, a big chunk of the
"analog with xformers, tape, et all 'sounds better'" tonality can be easily
accomplished in the digital world. You do this simply by putting a
medium-to-very-steep high-pass as the *first* item for each input in your digital
mix system. (Not at the end on the mix bus mind you, but at the beginning, on each
input.) Set it somewhere around 25-30 Hz.

Go a bit higher, say 40 Hz, if you don't need the LF -- but careful you don't ruin
attacks; and lower or not at all if you're doing pipe organs, or some such.

This highpass is exactly what the band-limiting transformers and tape machines of
the day where doing. And as you stacked transformers by going through the complete
signal path, the net roll-off got steeper.

(Remember, xformers weren't just at the mic pres, you found them at console
interstage points, ouptuts, tape machine ins and outs, line ins back at the console,
mix machine i/o, ad nauseam.)

Another chunk of "the sound" are the stocastic resonances imparted by noise -- tape
hiss. You can get some of that same affectation in digital land by adding a
low-level, dense, long decay reverb; or, a re-tweaking of your reverb fields in
general.

A remaining piece to consider is the hysteresis, phase shift, etc, etc, from all
those magnetics in the path. Do you really want that? (Remember, the better
magnetics of the day tried to minimize that.)


One final thing that I am so very happy to see go away is scrape flutter. Some tape
machines had less, but it was always there. Now that it's gone with digital I
couldn't be more delighted. (Scrape flutter was sort of replaced with jitter in some
early digital settings, but that can be much less an issue these days.)

I point to the work of those who've used bias signatures to digitally remove scrape
flutter from various famous recordings, and the resulting rave reviews. Have not
heard those myself, but I have no doubt as to the improvements.

I don't mean to start a holy war, but it's important, IMO, to understand why
something might be perceived sonically "better" in one domain, then see if you can
"fix" it in the other domain. YMMV.

Frank
Mobile Audio
--

[Trevor]

"Mike Rivers" > wrote in message
...
> No argument that some technology is an improvement, but we have a hard
> time being convinced that it's a direct replacement.

The Question is why you would want a direct replacement for inferior sound
technology? IF you don't want accurate sound, then the real quest is to
find the next hot NEW sound IMO.


>As far as I know, nobody has accurately modeled the audio qualities of a
>"famous" transformer. And I have to chuckle about the new round of analog
>tape simulators.

Of course, since everyone has different ideas of what the distortions were
that made some people prefer them.


> The fact that we have more ways to modify sound does not mean that we have
> ways of achieving the identical effect as by using real components.


No argument, for the nostalgia brigade, nothing will replace the original.


> We gain some, we lose some, we go on. I see no reason to mimic the sound
> of an analog recorder.


Right, me either.


>If a plug-in that claims to do this gives the user what he's looking for,
>that's fine. He probably doesn't know what an analog recorder sounds like
>anyway.

Or care. As long as the result is what he is after, why does it matter?

Trevor.

[Trevor]

"Arny Krueger" > wrote in message
...
> Duplicating the FR of a classic transformer is very doable. Duplicating
> the nonlinear distortion, not so much.

Actually it's not that difficult, just time consuming. If there is a market
then the time to develop proper software is OK. If not....


> I know how to do frequency dependent nonlinear distortion of any order
> with CEP and have done so, but actually doing it takes a lot of careful
> steps and mouth holding,

Right, using generic tools like CEP to do it is difficult.


> The hard-to-guess parts are the advantages of transformers when it comes
> to common mode, and out-of-band rejection.


I'm not arguing transformers have no use at all, just that when they do what
they are designed for properly, they do not change the "sound" compared to
digital filtering etc.
The argument is all about changing the "sound" by using transformers.

Trevor.

[Trevor]

"Scott Dorsey" > wrote in message
...
> I buy equipment that sounds good, then I keep using it as long as it
> sounds
> good. After a while, it either becomes valuable or not, but either way it
> still sounds good.

Yep, me too for many things. Other things that sounded good at the time,
still sound good, *but* NOT as good as what is available now (often for less
money) so I simply upgrade. If others nostalgia makes my old inferior gear
worth money, BONUS! :-)

Trevor.

[garyvee]

On Aug 30, 7:38*am, Frank Stearns >
wrote:
> Mike Rivers > writes:
> >on. I see no reason to mimic the sound of an analog
>
> x1000 on that.
>
> Working mostly in the classical and acoustic realm, the Holy Grail in my world was
> removal of any sound changers, starting with the 12-20 transformers that were in the
> complete signal path of a typical 1960s-1980s+ multitrack studio, and working
> outward from there. (Removing/improving tape, with all its failings, was next.)
>
> Having said that, there was something about the final sound of a typical
> higher-end path of the day that was compelling.
>
> But here's a modern trade secret: when you boil it all down, a big chunk of the
> "analog with xformers, tape, et all 'sounds better'" tonality can be easily
> accomplished in the digital world. You do this simply by putting a
> medium-to-very-steep high-pass as the *first* item for each input in your digital
> mix system. (Not at the end on the mix bus mind you, but at the beginning, on each
> input.) Set it somewhere around 25-30 Hz.
>
> Go a bit higher, say 40 Hz, if you don't need the LF -- but careful you don't ruin
> attacks; and lower or not at all if you're doing pipe organs, or some such.
>
> This highpass is exactly what the band-limiting transformers and tape machines of
> the day where doing. *And as you stacked transformers by going through the complete
> signal path, the net roll-off got steeper.
>
> (Remember, xformers weren't just at the mic pres, you found them at console
> interstage points, ouptuts, tape machine ins and outs, line ins back at the console,
> mix machine i/o, ad nauseam.)
>
> Another chunk of "the sound" are the stocastic resonances imparted by noise -- tape
> hiss. You can get some of that same affectation in digital land by adding a
> low-level, dense, long decay reverb; or, a re-tweaking of your reverb fields in
> general.
>
> A remaining piece to consider is the hysteresis, phase shift, etc, etc, from all
> those magnetics in the path. Do you really want that? (Remember, the better
> magnetics of the day tried to minimize that.)
>
> One final thing that I am so very happy to see go away is scrape flutter. Some tape
> machines had less, but it was always there. Now that it's gone with digital I
> couldn't be more delighted. (Scrape flutter was sort of replaced with jitter in some
> early digital settings, but that can be much less an issue these days.)
>
> I point to the work of those who've used bias signatures to digitally remove scrape
> flutter from various famous recordings, and the resulting rave reviews. Have not
> heard those myself, but I have no doubt as to the improvements.
>
> I don't mean to start a holy war, but it's important, IMO, to understand why
> something might be perceived sonically "better" in one domain, then see if you can
> "fix" it in the other domain. YMMV.
>
> Frank
> Mobile Audio
> --
> *.

Frank, thanks for the great post. I couldn't agree more. I use very
good pre-amps, converters and outboard analog gear including a _very_
nice line input/line output console and frankly I couldn't be happier
with the sound. It has never occurred to me that I should impart an
effect on the sound to get "that" sound.

Gary V

[Mike Rivers]

On 8/30/2011 7:30 PM, Trevor wrote:
> "Arny > wrote in message
> ...
>> Duplicating the FR of a classic transformer is very doable. Duplicating
>> the nonlinear distortion, not so much.
>
> Actually it's not that difficult, just time consuming. If there is a market
> then the time to develop proper software is OK. If not....

It's more than time consuming, it's processor intensive,
too. Focusrite has explored modeling further than most
hardware manufacturers. Several years back they partnered
with Sintefex to develop the original Liquid Channel using a
process called "Dynamic Convolution." This involves taking
lots of impulse response samples over both the full
frequency and amplitude range. Depending on the amplitude
and frequency of the input, the proper impulse response was
chosen to use in a convolution processor to provide an
output that closely resembled the real thing. While they
used this process to model electronics (preamps, equalizers,
and dynamics processors) they switched in real transformers,
inductors, and capacitors rather than modeling the physical
input of the preamp because that was easier (or maybe more
accurate).

> I'm not arguing transformers have no use at all, just that when they do what
> they are designed for properly, they do not change the "sound" compared to
> digital filtering etc.

Right. And all the modeling in the world won't provide the
common mode rejection of a good transformer (though some
analog electronics can get mighty close).

--
"Today's production equipment is IT based and cannot be
operated without a passing knowledge of computing, although
it seems that it can be operated without a passing knowledge
of audio." - John Watkinson

http://mikeriversaudio.wordpress.com - useful and
interesting audio stuff

[Andre Majorel]

On 2011-08-29, PStamler > wrote:

> The thing is, what you're probably after with "transformer sound"
> is the sound imparted to signals by the high-quality professional-
> grade transformers of the 1950s and 1960s, as heard in the recordings
> of those days.

Actually, it's whatever makes some say that the transformer
version of the Speck ASC sounds better than the electronically
balanced version. That sort of thing. But thank you for the
heads up.

--
André Majorel http://www.teaser.fr/~amajorel/
J'ai des droits. Les autres ont des devoirs.

[Scott Dorsey]

Andre Majorel > wrote:
>On 2011-08-29, PStamler > wrote:
>
>> The thing is, what you're probably after with "transformer sound"
>> is the sound imparted to signals by the high-quality professional-
>> grade transformers of the 1950s and 1960s, as heard in the recordings
>> of those days.
>
>Actually, it's whatever makes some say that the transformer
>version of the Speck ASC sounds better than the electronically
>balanced version. That sort of thing. But thank you for the
>heads up.

Well, in some situations the transformer-coupled version of a piece of
gear is more transparent because of the RF noise in the non-transformer-coupled
one. You pays your money and you takes your chance.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."