[Ian Bell[_2_]]

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Cheers

Ian

[David Light]

On Wed, 09 Sep 2009 10:33:42 +0100, Ian Bell
wrote:

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Very nice. I look forward to seeing photos of the finished mixer and
hearing some mixes.

[Ian Iveson]

Ian Bell wrote:

Following the discussion about a RAT mixer started well
over a year ago by Cipher I am finally building him an all
tube 6 into 2 mixer and building a business based on it.

Initial details he

www.customtubeconsoles.com

I thought of you as English, and wondered what's wrong with
"Bespoke Valve Mixing Desks"?

I can't find the circuit diagram.

You need to work on the feedback argument I think. Why is it
OK in the mu-follower (originated in 7th century Damascus,
apparently, BTW) when it's not acceptable in an equaliser
circuit? It's not that I want to know particularly...just
that the question rather poses itself, so you should answer
it, or rely on stupid buyers who don't know how circuits
work, which turns every rationale into snake oil.

Ian

[John Byrns]

In article ,
Ian Bell wrote:

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Hi Ian,

It's good to see someone take up "Cipher's" challenge and actually build
something. A few questions if I might.

1. Why is the high pass filter switch arranged with the "FLAT" position
in the middle? I find this jarring, but then I am probably just a
traditionalist who would put the flat position full CCW with ascending
cutoff frequencies in the CW direction. I'm guessing it was done to
minimize the number of switch clicks necessary to get from "FLAT" to any
choice of cutoff frequency?

2. "Cipher's" original wish list included pan pots on the input
modules, it appears that requirement has been dropped, I don't even see
switches to assign the individual channel modules either Left, Center,
Right, or even just to the Left or Right mix buses, what happened to
this feature, did "Cipher" drop it?

3. You mention that the low gain of tubes gives too much cross talk in
a "virtual earth" mixer, how does a passive mixer solve this problem?
The crosstalk problem was the biggest single design issue I encountered
in my design, using either a "virtual earth" or passive mixer.
Crosstalk seems like a non issue if the input modules can only only be
connected to either the left or right mixing bus, crosstalk only seems
to be an issue when an input module must drive both mix busses. Since
your mixer design doesn't appear to have either pan pots or switches
that allow assigning a input module to both mix buses simultaneously,
what crosstalk problem are your referring to?

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/

[Phil Allison[_3_]]

"Ian Bell"

Following the discussion about a RAT mixer started well over a year ago by
Cipher I am finally building him an all tube 6 into 2 mixer and building a
business based on it.

Initial details he

www.customtubeconsoles.com


** Congrats for re-inventing the horse and cart !!

I take issue with some of your comments about virtual earth mixing near the
bottom of the "Design Philosophy " page.

Firstly - it is not difficult to make sure there is simply no "ground
noise " on the signal ground path of in a mixer. Just keep any power
transformers well away from the cabling and do not connect de-coupling
electros from noisy supply rails to it.

Secondly, the claim that a 32 input virtual earth mixer, when adjusted for
unity signal gain, will have a noise gain of 33 times is bunk !!!

What do you imagine happens when there are 32 input signals, all peaking
around 0dBv, being summed by mixer that supplies unity signal gain to all
those inputs ????

Obviously, you get one HELL of a lot of signal coming out !!!!

Means the summing stage will be driven into gross overload, most or all the
time.

So it does not happen.

Fact is, the gain setting used for a virtual earth summing stage is
INVERSELY proportional to the number of input channels in use at any time.

The only time it will be set to " unity gain" is when only one or two input
channels are in use - as extra input channels are engaged, the virtual
earth mixer's gain setting must go down to * less than unity * to keep the
summed output level close to nominal recording level.

And hence, the imagined noise build up disappears.


..... Phil

[Ian Bell[_2_]]

David Light wrote:
On Wed, 09 Sep 2009 10:33:42 +0100, Ian Bell
wrote:

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Very nice. I look forward to seeing photos of the finished mixer and
hearing some mixes.

Cipher has a prototype channel and has used it a lot. He has promised me
some samples so when I get them I'll post them on the web site.


Cheers

Ian

[Ian Bell[_2_]]

John Byrns wrote:
In article ,
Ian Bell wrote:

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Hi Ian,

It's good to see someone take up "Cipher's" challenge and actually build
something. A few questions if I might.

1. Why is the high pass filter switch arranged with the "FLAT" position
in the middle? I find this jarring, but then I am probably just a
traditionalist who would put the flat position full CCW with ascending
cutoff frequencies in the CW direction. I'm guessing it was done to
minimize the number of switch clicks necessary to get from "FLAT" to any
choice of cutoff frequency?


Because those to the left of centre are 6dB/octave and those to the
right are 12dB/octave - it's what the client asked for.


2. "Cipher's" original wish list included pan pots on the input
modules, it appears that requirement has been dropped, I don't even see
switches to assign the individual channel modules either Left, Center,
Right, or even just to the Left or Right mix buses, what happened to
this feature, did "Cipher" drop it?


Yes, pan pots have been dropped. This resulted from discussions with
Cipher and with Jan-Eric Persson of Opus 3 Records. You may remember the
pictures Cipher showed of a tube mixer he liked - that was one developed
many years ago for Opus 3 and used on many of their recordings. It has
fixed pan for each channel in order to minimise the signal path. So one
channel is panned left, one right, one centre and the other right of
centre (this was specific to Opus 3 recording needs at the time)
Cipher's recording needs are similar so we agreed on two channels panned
left, two panned right and two panned centre. This allows him to use two
stereo mics and two spot mics.

3. You mention that the low gain of tubes gives too much cross talk in
a "virtual earth" mixer, how does a passive mixer solve this problem?
The crosstalk problem was the biggest single design issue I encountered
in my design, using either a "virtual earth" or passive mixer.
Crosstalk seems like a non issue if the input modules can only only be
connected to either the left or right mixing bus, crosstalk only seems
to be an issue when an input module must drive both mix busses. Since
your mixer design doesn't appear to have either pan pots or switches
that allow assigning a input module to both mix buses simultaneously,
what crosstalk problem are your referring to?


You are correct, crosstalk is only an issue when one channel drives both
mix busses, so this is an issue for the two channels that are hard
panned centre.

The two main factors determining cross talk in a passive mix bus are the
ratio of the channel source impedance to the bus mix resistors and the
bus loss. Crosstalk is the amount of one bus signal that reaches the
other bus. This occurs where more than one bus is fed from an active
output stage of the channel because there is a route from bus to bus vis
the output impedance of the active stage.

The route is first from the bus via the bus resistor to the channel
output. This signal is attenuated by the ratio of the bus resistor to
the channel output impedance. The larger the bus resistor and the
smaller the output impedance, the lower the crosstalk. When a channel is
muted its bus resistors are grounded making the effective source
impedance very low indeed.

The last part of the route is from the channel output back onto the
other bus. The crosstalk signal is thus attenuated further by the the
bus loss which is usually 1/N where N is the number of bus feeds. In
Cipher's mixer, although there are 6 channels, because four are routed
directly to left or right, each bus ends up having only four feeds so N
is 4 and bus attenuation is just 12dB. The more bus feeds you have, the
better the crosstalk. You can also slug the bus to increase its
attenuation and improve crosstalk but of course you need to watch the noise.

In the original design, with pan pots, there was a cathode follower
after each arm of the pan pot so in fact there was no direct path
between busses. In the fixed pan design I have deleted these cathode
followers so the crosstalk depends mainly on the output impedance of the
stage driving the bus resistors and the value of the bus resistors
themselves. Currently the bus resistor value is 220K and the channel
output impedance is 800 ohms giving a crosstalk attenuation of about
48dB. This plus the 12dB loss in the bus means crosstalk will be -60dB
when a centre panned channel is un-muted.

Cheers

Ian

[Ian Bell[_2_]]

Phil Allison wrote:
"Ian Bell"
Following the discussion about a RAT mixer started well over a year ago by
Cipher I am finally building him an all tube 6 into 2 mixer and building a
business based on it.

Initial details he

www.customtubeconsoles.com


** Congrats for re-inventing the horse and cart !!

I take issue with some of your comments about virtual earth mixing near the
bottom of the "Design Philosophy " page.

Firstly - it is not difficult to make sure there is simply no "ground
noise " on the signal ground path of in a mixer. Just keep any power
transformers well away from the cabling and do not connect de-coupling
electros from noisy supply rails to it.

Secondly, the claim that a 32 input virtual earth mixer, when adjusted for
unity signal gain, will have a noise gain of 33 times is bunk !!!

What do you imagine happens when there are 32 input signals, all peaking
around 0dBv, being summed by mixer that supplies unity signal gain to all
those inputs ????

Obviously, you get one HELL of a lot of signal coming out !!!!

Means the summing stage will be driven into gross overload, most or all the
time.

So it does not happen.

Fact is, the gain setting used for a virtual earth summing stage is
INVERSELY proportional to the number of input channels in use at any time.

The only time it will be set to " unity gain" is when only one or two input
channels are in use - as extra input channels are engaged, the virtual
earth mixer's gain setting must go down to * less than unity * to keep the
summed output level close to nominal recording level.

And hence, the imagined noise build up disappears.


.... Phil




Phil, for once your normally wide ranging knowledge fails you. It is
clear from the above you have very little understanding of audio mixers
and noise in particular. I am not about to teach you - instead I will
simply refer you to one who wrote much of this down some time ago.

http://84.255.203.119/Steve-Dove-Console-Design.pdf

Cheers

Ian

[Phil Allison[_3_]]

"Ian Bell"
Phil Allison wrote:
"Ian Bell"

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com


** Congrats for re-inventing the horse and cart !!

I take issue with some of your comments about virtual earth mixing near
the bottom of the "Design Philosophy " page.

Firstly - it is not difficult to make sure there is simply no "ground
noise " on the signal ground path of in a mixer. Just keep any power
transformers well away from the cabling and do not connect de-coupling
electros from noisy supply rails to it.

Secondly, the claim that a 32 input virtual earth mixer, when adjusted
for unity signal gain, will have a noise gain of 33 times is bunk !!!

What do you imagine happens when there are 32 input signals, all peaking
around 0dBv, being summed by mixer that supplies unity signal gain to all
those inputs ????

Obviously, you get one HELL of a lot of signal coming out !!!!

Means the summing stage will be driven into gross overload, most or all
the time.

So it does not happen.

Fact is, the gain setting used for a virtual earth summing stage is
INVERSELY proportional to the number of input channels in use at any
time.

The only time it will be set to " unity gain" is when only one or two
input channels are in use - as extra input channels are engaged, the
virtual earth mixer's gain setting must go down to * less than unity *
to keep the summed output level close to nominal recording level.

And hence, the imagined noise build up disappears.



Phil, for once your normally wide ranging knowledge fails you.


** Oh dear - what a waste of my good time writing tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!~!!

This retarded SMUG POMMY ASS is TOTALLY incapable of

following even the simplest of points.

So he used the CHARLATAN' S favourite tactic and posted the ( now dead)
link to where he STOLE the WRONG info from.

**** YOU Ian Bell,

Pommy **** !!



..... Phil

[Ian Bell[_2_]]

Phil Allison wrote:


snip

So he used the CHARLATAN' S favourite tactic and posted the ( now dead)
link to where he STOLE the WRONG info from.


snip


..... Phil


The link is not dead - it works - I just tried it and it works. Get it,
read it, then reply.

Cheers

Ian

[Phil Allison[_3_]]

"Ian Bell is a LYING POMMY **** "


The link is not dead - it works - I just tried it and it works.


** Did not work an hour ago.



...... Phil

[Phil Allison[_3_]]

"Ian Bell"
Phil Allison wrote:
"Ian Bell"

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com


** Congrats for re-inventing the horse and cart !!

I take issue with some of your comments about virtual earth mixing near
the bottom of the "Design Philosophy " page.

Firstly - it is not difficult to make sure there is simply no "ground
noise " on the signal ground path of in a mixer. Just keep any power
transformers well away from the cabling and do not connect de-coupling
electros from noisy supply rails to it.

Secondly, the claim that a 32 input virtual earth mixer, when adjusted
for unity signal gain, will have a noise gain of 33 times is bunk !!!

What do you imagine happens when there are 32 input signals, all peaking
around 0dBv, being summed by mixer that supplies unity signal gain to all
those inputs ????

Obviously, you get one HELL of a lot of signal coming out !!!!

Means the summing stage will be driven into gross overload, most or all
the time.

So it does not happen.

Fact is, the gain setting used for a virtual earth summing stage is
INVERSELY proportional to the number of input channels in use at any
time.

The only time it will be set to " unity gain" is when only one or two
input channels are in use - as extra input channels are engaged, the
virtual earth mixer's gain setting must go down to * less than unity *
to keep the summed output level close to nominal recording level.

And hence, the imagined noise build up disappears.



Phil, for once your normally wide ranging knowledge fails you.


** Oh dear - what a waste of my good time writing tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!~!!

This retarded SMUG POMMY ASS is TOTALLY incapable of

following even the simplest of points.

So he used the CHARLATAN' S favourite tactic and posted the ( now dead)
link to where he STOLE the WRONG info from.

**** YOU Ian Bell,

Pommy **** !!




..... Phil

[Ian Bell[_2_]]

Phil Allison wrote:


** Did not work an hour ago.




Well it works every time I try it. Don't forget, Phil, it is a 40Mbyte
file which even on my ADSL connection downloads only at about
80Kbytes/sec so it takes a few minutes.

If you still cannot get it just try to root so you can see the list of
files and try a smaller one first; So, just go to:

http://84.255.203.119


and you should see the directory and file structure. The file you want
is five up from the bottom.


To be honest, Phil, I am surprised a man of your intellect had not
already tried this.

Cheers

Ian

[Ian Bell[_2_]]

Ian Iveson wrote:
Ian Bell wrote:

Following the discussion about a RAT mixer started well
over a year ago by Cipher I am finally building him an all
tube 6 into 2 mixer and building a business based on it.

Initial details he

www.customtubeconsoles.com

I thought of you as English, and wondered what's wrong with
"Bespoke Valve Mixing Desks"?

I am English and I have been trying to come up with a catchy name for
some time. CTC is just a placeholder right now. Actually, I do rather
like "Bespoke Valve Mixing Desks" - sounds like a cross between Neve and
Saville Row. May I use it?


I can't find the circuit diagram.


What circuit diagram would that be ;-)

You need to work on the feedback argument I think. Why is it
OK in the mu-follower (originated in 7th century Damascus,
apparently, BTW) when it's not acceptable in an equaliser
circuit? It's not that I want to know particularly...just
that the question rather poses itself, so you should answer
it, or rely on stupid buyers who don't know how circuits
work, which turns every rationale into snake oil.

I tried hard to make it clear that it is "global" negative feedback that
is to be avoided and the reasons why. The only NFB in a mu follower is
local to the top cathode follower and operates down to dc so it avoids
all the problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input

Cheers

Ian


Ian

[Phil Allison[_3_]]

Ian Bell = a congenital LYING moron


** What a waste of my good time writing any tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!!

This RETARDED POMMY ASS is TOTALLY incapable of

following even the simplest of tech points.

So the **** used the CHARLATAN'S favourite CHEAT and just

posted a link to where he STOLE the WRONG info from.

IMBECILE !!!

**** YOU Ian Bell !!

One UGLY pommy **** !!



..... Phil

[Phil Allison[_3_]]

"Ian Bell is a LYING TURD "


I tried hard to make it clear that it is "global" negative feedback that
is to be avoided and the reasons why.

** The scumbag just trotted out one of the usual stale, fake, audiophool
****wit excuses for producing a bad design.


Total waste of time writing any tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!!

This RETARDED POMMY ASS is TOTALLY incapable of

following even the simplest of tech points.

So the **** used the CHARLATAN'S favourite CHEAT and just

posted a link to where he STOLE the WRONG info from.

IMBECILE !!!

**** YOU Ian Bell !!

One FUG UGLY pommy **** !!




..... Phil

[Bob Eld]

"Phil Allison" wrote in message
...
"Ian Bell is a LYING TURD "


I tried hard to make it clear that it is "global" negative feedback that
is to be avoided and the reasons why.

** The scumbag just trotted out one of the usual stale, fake, audiophool
****wit excuses for producing a bad design.


Total waste of time writing any tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!!

This RETARDED POMMY ASS is TOTALLY incapable of

following even the simplest of tech points.

So the **** used the CHARLATAN'S favourite CHEAT and just

posted a link to where he STOLE the WRONG info from.

IMBECILE !!!

**** YOU Ian Bell !!

One FUG UGLY pommy **** !!




..... Phil

I see Phyllis is having a Tourettes Attack again. Sad, but the poor *******
needs help.

[Ian Bell[_2_]]

Bob Eld wrote:
"Phil Allison" wrote in message
...
"Ian Bell is a LYING TURD "


I tried hard to make it clear that it is "global" negative feedback that
is to be avoided and the reasons why.
** The scumbag just trotted out one of the usual stale, fake, audiophool
****wit excuses for producing a bad design.


Total waste of time writing any tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!!

This RETARDED POMMY ASS is TOTALLY incapable of

following even the simplest of tech points.

So the **** used the CHARLATAN'S favourite CHEAT and just

posted a link to where he STOLE the WRONG info from.

IMBECILE !!!

**** YOU Ian Bell !!

One FUG UGLY pommy **** !!




..... Phil

I see Phyllis is having a Tourettes Attack again. Sad, but the poor *******
needs help.




I don't know but I think he is on medication because he does have quite
a few lucid and expletive free periods - although right now he seems to
be mid relapse.

Cheers

Ian

[Phil Allison[_3_]]

"Ian Bell is a LYING TURD "

I tried hard to make it clear that it is "global" negative feedback that
is to be avoided and the reasons why.

** The scumbag just trotted out one of the usual stale, fake, audiophool
****wit excuses for producing a bad design.


Total waste of time writing any tech sense to a

CONGENITAL ****ING MORON like * Ian ****ing Bell ** !!!

This RETARDED POMMY ASS is TOTALLY incapable of

following even the simplest of tech points.

So the **** used the CHARLATAN'S favourite CHEAT and just

posted a link to where he STOLE the WRONG info from.

IMBECILE !!!

**** YOU Ian Bell !!

One FUG UGLY pommy **** !!



..... Phil

[John Byrns]

In article ,
Ian Bell wrote:

John Byrns wrote:
In article ,
Ian Bell wrote:

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Hi Ian,

It's good to see someone take up "Cipher's" challenge and actually build
something. A few questions if I might.

1. Why is the high pass filter switch arranged with the "FLAT" position
in the middle? I find this jarring, but then I am probably just a
traditionalist who would put the flat position full CCW with ascending
cutoff frequencies in the CW direction. I'm guessing it was done to
minimize the number of switch clicks necessary to get from "FLAT" to any
choice of cutoff frequency?


Because those to the left of centre are 6dB/octave and those to the
right are 12dB/octave - it's what the client asked for.

Hi Ian,

Thanks for the clarification on the operation of the high pass filter
switch. The reason for the "FLAT" position being in the center wasn't
at all obvious to me; you have cleared that up nicely.

While we are on the subject of the high pass filter, could you say
something about your design philosophy for the 12 dB/octave filter?
From the photos it appears that the filter circuit is built from only
Resistors and Capacitors, with no inductors or active devices used in
the filter? I'm no filter guru but it has been my impression that it is
not possible to build a 12 dB/octave high pass filter this way without
making some compromises in the sharpness of the knee in the filters
response curve. Could you elaborate on your design considerations for
the 12 dB/octave high pass filter circuit? It occurs to me that the
³soft knee² this type of filter has may be considered to be an advantage
for audiophile oriented audio applications because of less abrupt phase
changes?

On a completely different subject, what is the power supply design for
the Custom Tube Console like? While the power supply is a somewhat
prosaic item, many people consider the power supply to be perhaps the
most sonically important section of an audio product, so I am curious
about what your thoughts are on this section of the design?

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/

[Ian Bell[_2_]]

John Byrns wrote:
In article ,
Ian Bell wrote:

John Byrns wrote:
In article ,
Ian Bell wrote:

Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com
Hi Ian,

It's good to see someone take up "Cipher's" challenge and actually build
something. A few questions if I might.

1. Why is the high pass filter switch arranged with the "FLAT" position
in the middle? I find this jarring, but then I am probably just a
traditionalist who would put the flat position full CCW with ascending
cutoff frequencies in the CW direction. I'm guessing it was done to
minimize the number of switch clicks necessary to get from "FLAT" to any
choice of cutoff frequency?

Because those to the left of centre are 6dB/octave and those to the
right are 12dB/octave - it's what the client asked for.

Hi Ian,

Thanks for the clarification on the operation of the high pass filter
switch. The reason for the "FLAT" position being in the center wasn't
at all obvious to me; you have cleared that up nicely.

While we are on the subject of the high pass filter, could you say
something about your design philosophy for the 12 dB/octave filter?
From the photos it appears that the filter circuit is built from only
Resistors and Capacitors, with no inductors or active devices used in
the filter? I'm no filter guru but it has been my impression that it is
not possible to build a 12 dB/octave high pass filter this way without
making some compromises in the sharpness of the knee in the filters
response curve. Could you elaborate on your design considerations for
the 12 dB/octave high pass filter circuit? It occurs to me that the
³soft knee² this type of filter has may be considered to be an advantage
for audiophile oriented audio applications because of less abrupt phase
changes?


It is a very simple passive design, just two series C, parallel R stages
and yes it does have quite a soft knee. This design is typical of many
professional mixers.


On a completely different subject, what is the power supply design for
the Custom Tube Console like? While the power supply is a somewhat
prosaic item, many people consider the power supply to be perhaps the
most sonically important section of an audio product, so I am curious
about what your thoughts are on this section of the design?


The power supply is one of the most difficult and important parts of the
mixer. First it is mostly physically separate from the mixer itself in a
self contained box so as to avoid problems with mains transformer fields
affecting sensitive circuits. It is being built in an enclosure intended
for a 30W guitar amp so it has a chassis and a ventilated top cage. The
heater side provides about 17V dc roughly smoothed (100,000uF) at 6amps
from a toroidal transformer. Inside the mixer is a 12V regulator per
module each providing 12V dc at 0.6A (there will be 9 of these). In
theory this may not be necessary. So far I have used rectified smoothed
heaters but not regulated. The prototype channel was fist powered up
with ac heaters and a hum bucking pot. I found the 50Hz heater noise was
10dB above the broadband system noise. So I simply rectified the 6.3V ac
and smoothed it with 47,000uF and fed it to the heaters and it measured
6.3V dc when loaded with 1.2A (it was a 1.5A winding). Hum at 50Hz and
100Hz was then completely absent. Theoretical pp 100Hz on the heater
supply was then about 250mV compared with the ~18V pp due to 6.3V ac (a
37dB reduction). So with 100,000uF at 12V and 5.4amps load I reckon the
pp 100Hz hum should be about 0.5V or about 30dB below 6.3V ac. Since the
hum was only 10dB above system noise on ac heaters then this should put
it 20dB below system noise. The two channel demo unit I am building now
has 6.3V dc heaters with ~200,000uF of smoothing (4-off 47,000uF).

The HT side is less certain at present. As everything is class A there
are no peak currents to cause droop so regulation is not a requirement
form that point of view. The main issue is noise on the HT line. The
basic HT requirement is 300V at just under 100mA. The prototype channel
HT power supply used five cascaded stages of 1K and 470uF. In theory
this produces something like 130dB attenuation of the 100Hz ripple which
brings it down to the microvolt region. In practice it is rock steady no
matter what the signal conditions. For the mixer the current draw is 10
times as much so one simple solution is to use five cascaded stages of
100R and 470uF but this considerably worsens the ripple so I plan
initially to try four stages of 100R and 470uF with a per module final
1K and 470uF. This does not take into account the PSRR of the modules
which should be around 20dB.

Cheers

Ian

[Ian Iveson]

I am English and I have been trying to come up with a
catchy name for some time. CTC is just a placeholder right
now. Actually, I do rather like "Bespoke Valve Mixing
Desks" - sounds like a cross between Neve and Saville Row.
May I use it?

I was only pointing out that you've used three American
words to describe a British product. I wondered how it plays
to Americans. In the UK it could possibly pass as a brand
name, but in America it's a simple product description.
Isn't there a problem with registering such a name?

As for "BVMD", If you were a Saville Row tailor, you
wouldn't call yourself
"Bespoke Menswear". Even if you were the only maker of
made-to-measure suits, you'd still be wise to name yourself
something like "Gieves & Hawkes". Gives the impression of
tradition and personal service from grovelling artisans. The
acronym sounds like a mining corporation or heavy
engineering conglomerate.

I'd call my company "Iveson" or "Kahn" but you'd sound like
a telephone exchange or helicopter manufacturer.

I guess "Confluence" is already taken too, but there's
Chinese words...or some more obscure language perhaps, with
a nice symbol. Japanese is handy because they use both Roman
script and symbols, and they are associated with quality
(perhaps their word for "entrance" as in way in, rather than
put into a trance).

Ian Bell wrote:

I can't find the circuit diagram.

What circuit diagram would that be ;-)

We don't generally discuss projects without access to a
circuit diagram. Or at least we didn't until that mixer
thread mentioned money.

You need to work on the feedback argument I think. Why is
it OK in the mu-follower (originated in 7th century
Damascus, apparently, BTW) when it's not acceptable in an
equaliser circuit? It's not that I want to know
particularly...just that the question rather poses
itself, so you should answer it, or rely on stupid buyers
who don't know how circuits work, which turns every
rationale into snake oil.

I tried hard to make it clear that it is "global" negative
feedback that is to be avoided and the reasons why. The
only NFB in a mu follower is local to the top cathode
follower and operates down to dc so it avoids all the
problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input

Without a circuit diagram it's impossible to say whether
your use of the mu-follower suffers from the problem of
feedback that you identify. "...operates down to DC so..."
could be misleading, depending on what it's driving, seems
to me.

Feedback in the equalisation circuit wouldn't be global, but
you appear to reject its use there as a matter of principle.
So where, between global and local, does this principle
cease to apply? If there isn't a simple, sensible and
complete answer
to this question, I suggest you either abandon the argument,
or make a mixer with zero feedback so you can honestly say
so. Maybe anyone who objects to feedback doesn't need you to
tell them why. Just give the impression that you agree with
whatever they already think. Say it strangles the sound,
music needs air to breathe and feedback consumes oxygen,
that kind of stuff.

As it stands, your "philosophy" has quite a few gaps, and
counterposing valves and opamps without mentioning discrete
transistors appears contrived.

Ian

[Ian Bell[_2_]]

Ian Iveson wrote:
I am English and I have been trying to come up with a
catchy name for some time. CTC is just a placeholder right
now. Actually, I do rather like "Bespoke Valve Mixing
Desks" - sounds like a cross between Neve and Saville Row.
May I use it?

I was only pointing out that you've used three American
words to describe a British product. I wondered how it plays
to Americans. In the UK it could possibly pass as a brand
name, but in America it's a simple product description.
Isn't there a problem with registering such a name?


Asd I said, CTC is just a placeholde - I am open to suggestion for
suitable names.

As for "BVMD", If you were a Saville Row tailor, you
wouldn't call yourself
"Bespoke Menswear". Even if you were the only maker of
made-to-measure suits, you'd still be wise to name yourself
something like "Gieves & Hawkes". Gives the impression of
tradition and personal service from grovelling artisans. The
acronym sounds like a mining corporation or heavy
engineering conglomerate.

I'd call my company "Iveson" or "Kahn" but you'd sound like
a telephone exchange or helicopter manufacturer.


My full surname is Thompson-Bell which sounds a bit 'posher' don't you
think.

I guess "Confluence" is already taken too, but there's
Chinese words...or some more obscure language perhaps, with
a nice symbol. Japanese is handy because they use both Roman
script and symbols, and they are associated with quality
(perhaps their word for "entrance" as in way in, rather than
put into a trance).

Ian Bell wrote:
I can't find the circuit diagram.

What circuit diagram would that be ;-)

We don't generally discuss projects without access to a
circuit diagram. Or at least we didn't until that mixer
thread mentioned money.


Indeed.


You need to work on the feedback argument I think. Why is
it OK in the mu-follower (originated in 7th century
Damascus, apparently, BTW) when it's not acceptable in an
equaliser circuit? It's not that I want to know
particularly...just that the question rather poses
itself, so you should answer it, or rely on stupid buyers
who don't know how circuits work, which turns every
rationale into snake oil.
I tried hard to make it clear that it is "global" negative
feedback that is to be avoided and the reasons why. The
only NFB in a mu follower is local to the top cathode
follower and operates down to dc so it avoids all the
problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input

Without a circuit diagram it's impossible to say whether
your use of the mu-follower suffers from the problem of
feedback that you identify. "...operates down to DC so..."
could be misleading, depending on what it's driving, seems
to me.

The local feedback in the top CF operates down to dc; what it is driving
does not alter that.


I am sure I don't need to draw you the circuit of a mu follower and I
am sure you know the only NFb it has is in the top CF.

Feedback in the equalisation circuit wouldn't be global,

That is an interesting assumption. There are indeed many valve tone
control circuits that use feedback around a single CC stage (i.e local NFB

) but they all perform poorly in terms of distortion because at maximum
boost they have very little feedback. To keep distortion at a low level
needs more open loop gain which means more than one stage and hence
global feedback.


but
you appear to reject its use there as a matter of principle.
So where, between global and local, does this principle
cease to apply? If there isn't a simple, sensible and
complete answer
to this question, I suggest you either abandon the argument,
or make a mixer with zero feedback so you can honestly say
so.

The principle is that NFB should extend down to dc, whether NFB is local
or global. You can make multi-stage amplifiers where this is true - the
Pultec mic pre is a good example.




Maybe anyone who objects to feedback doesn't need you to
tell them why. Just give the impression that you agree with
whatever they already think. Say it strangles the sound,
music needs air to breathe and feedback consumes oxygen,
that kind of stuff.

As it stands, your "philosophy" has quite a few gaps, and
counterposing valves and opamps without mentioning discrete
transistors appears contrived.

In what way contrived? Op amps are merely some transistors in a box.

Cheers

Ian

Ian

[Patrick Turner]

On Sep 9, 7:33*pm, Ian Bell wrote:
Following the discussion about a RAT mixer started well over a year ago
by Cipher I am finally building him an all tube 6 into 2 mixer and
building a business based on it.

Initial details he

www.customtubeconsoles.com

Cheers

Ian

Keep up the good work.

Diecast boxes might seem just fine but they don't do much to screen
out magnetic fields. However, they may not be a problem if you have
fairly remote PSUs.
Its usually amazing how quiet and hum free normal unbalanced audio
signal circuits become once you run all the heaters on DC and have a
remote PSU with good filtering.

Good pro gear does not seem to utilize diecast boxes. They will use
maybe a rack mount case specially made for the job so several prams or
mixers can be stacked in a rack. Units were made so its very easy to
replace a tube, and the tubes are set up so there is a natural breeze
past the glass. They often used clip on tube cover cylinders over the
tubes which had a spring inside to make sure the tube could never
wriggle loose or be bent over sideways which happens today with
stupidly designed audiophile gear.

Patrick Turner.

[Patrick Turner]

On Sep 11, 1:29*am, Ian Bell wrote:
Ian Iveson wrote:
Ian Bell wrote:

Following the discussion about a RAT mixer started well
over a year ago by Cipher I am finally building him an all
tube 6 into 2 mixer and building a business based on it.

Initial details he

www.customtubeconsoles.com

I thought of you as English, and wondered what's wrong with
"Bespoke Valve Mixing Desks"?

I am English and I have been trying to come up with a catchy name for
some time. CTC is just a placeholder right now. Actually, I do rather
like "Bespoke Valve Mixing Desks" - sounds like a cross between Neve and
Saville Row. May I use it?



I can't find the circuit diagram.

What circuit diagram would that be ;-)

You need to work on the feedback argument I think. Why is it
OK in the mu-follower (originated in 7th century Damascus,
apparently, BTW) when it's not acceptable in an equaliser
circuit? It's not that I want to know particularly...just
that the question rather poses itself, so you should answer
it, or rely on stupid buyers who don't know how circuits
work, which turns every rationale into snake oil.

I tried hard to make it clear that it is "global" negative feedback that
is to be avoided and the reasons why. The only NFB in a mu follower is
local to the top cathode follower and operates down to dc so it avoids
all the problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input

Cheers

Ian


Not many people who buy something with vacuum tubes within actually
have any idea about any kind of circuit working.

Of every 1,000 ppl buying tube gear this week around the Earth, maybe
2 will have a slight idea about how NFB works and how it's
effectiveness is affected by the open loop N&D before NFB is
connected, the OL bandwidth, the spectra involved and the amount of
NFB used.

So the safest thing to say about one's pet product when trying to sell
it is a polite load of utter bull****, and most listening will hear a
few key words and conclude that the speaker is a learned man and then
they buy it.

When I have to speak to people about amps I do make and sell, I do try
to tell the truth all the time. I back it up with a website which has
22MB of infomation which hardly anyone I know understands at all. For
me its easier to just befuddle and bambozzle with the truth than to
think up bull****, because that takes time, and its difficult to be
consistent with bull****, and you never know when some brighter than
average punter will point out to all and sundry that what you said was
utter bull****, and contradictory to what was said half an hour
before.

As time rolls by, people find it doesn't matter that they cannot
undestand a single sylable about "impedance" or what a dB is, or why I
I make tube amplifiers. At least they begin to realise things are not
so bad because they ain't listeing to a politician or a Wall Street
financial ex ****in spert.

Most buyers of audio gear and probably all those who insist on
recording studio tube gear will be buying because of the sound
quality, and their gut feeling about what seems to be a good deal.

Mu-followers do have a shirt and trouser load of local NFB around the
top tube which works like a normal cathode follower, or like a non-
inverting opamp with a unity gain where ALL the output is in series
with the input. (99% of people operating studios would be totally
bamboozled with just this much info)

Where triodes are used, the bottom gain producing triode of the mu-
foll operates with a very high RL of many times the Ra due to the
bootstrapping from the top output follower triode. Therefore the
voltage gain approaches µ, and the signal current change becomes
extremely low and the distortion is minimal, with 0.1% THD at 10Vrms
being achievable, instead of the usual 0.1% at 1Vrms with a
conventional resistance loading. One reason why the THD becomes lowest
when the load becomes many times Ra is because the internal NFB within
the triode applies itself maximally when the internal gain without
that internal FB would otherwise be extremely high.

Unfortunately, the last paragraph is meaningless to 99.9999999999999%
of the population.

Nearly all amplifiers with series voltage NFB display increasing THD/
IMD with load reduction and increasing output current for a given
output voltage.

It is because the amount of applied NFB reduces as the open loop gain
reduces with reducing load.

It applies with a simple cathode follower or to a 3 stage amplifier.

Why not call your proposed amplifiers "Bell Tube Amps", or just "Bell
Mixers"?

If you were worried what people thought about your schematics, publish
them on the website with a full explanation and with references to
what each of many necessary enginering terms means.

Maybe just doing that takes you untill 2013.......

Patrick Turner






Ian- Hide quoted text -

- Show quoted text -

[Ian Iveson]

Ian Bell wrote:

My full surname is Thompson-Bell which sounds a bit
'posher' don't you think.

Posh isn't what you want. Posh people are stupid consumers,
whereas you need to appear to be a clever provider; an
accomplished artisan; a grovelling servant to a master who
believes he decides what you make. These conflicting
requirements, requiring some pretences, are part of the
feeling of special relationship, like an in joke.

I guess "Confluence" is already taken too, but there's
Chinese words...or some more obscure language perhaps,
with a nice symbol. Japanese is handy because they use
both Roman script and symbols, and they are associated
with quality (perhaps their word for "entrance" as in way
in, rather than put into a trance).

We don't generally discuss projects without access to a
circuit diagram. Or at least we didn't until that mixer
thread mentioned money.

Indeed.

Feeding parasitic profiteers isn't very recreational, in my
view.

You need to work on the feedback argument I think. Why
is
it OK in the mu-follower (originated in 7th century
Damascus, apparently, BTW) when it's not acceptable in
an
equaliser circuit? It's not that I want to know
particularly...just that the question rather poses
itself, so you should answer it, or rely on stupid
buyers
who don't know how circuits work, which turns every
rationale into snake oil.

I tried hard to make it clear that it is "global"
negative
feedback that is to be avoided and the reasons why. The
only NFB in a mu follower is local to the top cathode
follower and operates down to dc so it avoids all the
problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input

Without a circuit diagram it's impossible to say whether
your use of the mu-follower suffers from the problem of
feedback that you identify. "...operates down to DC
so..." could be misleading, depending on what it's
driving, seems to me.

The local feedback in the top CF operates down to dc; what
it is driving does not alter that.

I said "misleading". Yes the feedback is direct, and in some
academic sense you could say it therefore operates at DC.
But in reality the mu-follower itself doesn't operate at DC.
That's only one reason why your argument is misleading. When
you constructed or borrowed your agument, the ability to
operate at DC was not a primary requirement was it? Neither
was it the only derived requirement for acceptable feedback.

What it is driving *does* alter whether the mu-follower
meets your primary requirement for a feedback circuit to be
acceptable. I guess about half of your mu-follows suffer
from one of the pitfalls of misused feedback that you
identify. If you daren't post your circuit, I guess I can
assume the worst.

I am sure I don't need to draw you the circuit of a mu
follower

It's *your* mu-follower I want to see, not some idealised
notion. In the context of the rest of *your* circuit.

and I am sure you know the only NFb it has is in the top
CF.

Yes. Try telling Patrick. Actually, don't. No point in
provoking another rash of drivel from a fellow trader
looking for a sales pitch.

Feedback in the equalisation circuit wouldn't be global,

That is an interesting assumption. There are indeed many
valve tone control circuits that use feedback around a
single CC stage (i.e local NFB

) but they all perform poorly in terms of distortion
because at maximum boost they have very little feedback.
To keep distortion at a low level needs more open loop
gain which means more than one stage and hence global
feedback.

From which I should take it that you mean local feedback is
round a single stage, whereas feedback round more than one
stage is global? That's not what is usually meant by global.
Global usually means round the entire circuit, from output
to input. Global feedback round an equalisation circuit
surely defeats the object?

Anyway, you haven't answered the question. When is feedback
OK and when is it not? So far I might infer that you believe
that only DC-coupled feedback round a single stage is
acceptable. Is that correct? If so, your arguments fall a
long way short of legitimising your opinion. As I've said,
either get them right or leave them out and think of some
more attractive bull**** instead.

but
you appear to reject its use there as a matter of
principle.
So where, between global and local, does this principle
cease to apply? If there isn't a simple, sensible and
complete answer
to this question, I suggest you either abandon the
argument,
or make a mixer with zero feedback so you can honestly
say
so.

The principle is that NFB should extend down to dc,
whether NFB is local or global. You can make multi-stage
amplifiers where this is true - the Pultec mic pre is a
good example.

So why have you limited yourself to the most local feedback
possible? That's another problem with what you call your
"philosophy": you imply that you have avoided more extensive
feedback because it is difficult to achieve. It appears
therefore that your "philosophy" is to do whatever is
easiest. Not befitting, as it were, of a Saville Row tailor.

Maybe anyone who objects to feedback doesn't need you to
tell them why. Just give the impression that you agree
with
whatever they already think. Say it strangles the sound,
music needs air to breathe and feedback consumes oxygen,
that kind of stuff.

As it stands, your "philosophy" has quite a few gaps, and
counterposing valves and opamps without mentioning
discrete transistors appears contrived.

In what way contrived? Op amps are merely some transistors
in a box.

Well that's another example of one of several species of
contrivance employed in your argument...or whoever's
argument it is. Presumably this is deliberate chicanery?

Perhaps I should offer a clue in case some dimwits are
reading. Had you counterposed valves and transistors, and
had I then objected to the absence of any mention of opamps,
then your response would have been legitimate.

Try drawing yourself a little Venn diagram and your error
should jump out at you.

http://en.wikipedia.org/wiki/Venn_diagram

Acquaintance with some rudiments of classical logic might
help, eg:

http://en.wikipedia.org/wiki/Syllogism

Ian

[Ian Bell[_2_]]

Ian Iveson wrote:
Ian Bell wrote:

My full surname is Thompson-Bell which sounds a bit
'posher' don't you think.

Posh isn't what you want. Posh people are stupid consumers,
whereas you need to appear to be a clever provider; an
accomplished artisan; a grovelling servant to a master who
believes he decides what you make. These conflicting
requirements, requiring some pretences, are part of the
feeling of special relationship, like an in joke.


Yes, you are right. Again, CTC is only a placeholder and if you have
any good ideas for a name I would be glad to hear them.

I guess "Confluence" is already taken too, but there's
Chinese words...or some more obscure language perhaps,
with a nice symbol. Japanese is handy because they use
both Roman script and symbols, and they are associated
with quality (perhaps their word for "entrance" as in way
in, rather than put into a trance).

We don't generally discuss projects without access to a
circuit diagram. Or at least we didn't until that mixer
thread mentioned money.

Indeed.

Feeding parasitic profiteers isn't very recreational, in my
view.


Well, it is unusual for someone to ask about commisioning a design on
this newsgroup so I can understand people's caution.

You need to work on the feedback argument I think. Why
is
it OK in the mu-follower (originated in 7th century
Damascus, apparently, BTW) when it's not acceptable in
an
equaliser circuit? It's not that I want to know
particularly...just that the question rather poses
itself, so you should answer it, or rely on stupid
buyers
who don't know how circuits work, which turns every
rationale into snake oil.

I tried hard to make it clear that it is "global"
negative
feedback that is to be avoided and the reasons why. The
only NFB in a mu follower is local to the top cathode
follower and operates down to dc so it avoids all the
problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input
Without a circuit diagram it's impossible to say whether
your use of the mu-follower suffers from the problem of
feedback that you identify. "...operates down to DC
so..." could be misleading, depending on what it's
driving, seems to me.
The local feedback in the top CF operates down to dc; what
it is driving does not alter that.

I said "misleading". Yes the feedback is direct, and in some
academic sense you could say it therefore operates at DC.

There's nothing academic about it.

But in reality the mu-follower itself doesn't operate at DC.

I never said it did. I said the NFB in the CF at the top of the
mu-follower operates down to dc. That's the only NFB in a mu-follower.

That's only one reason why your argument is misleading. When
you constructed or borrowed your agument, the ability to
operate at DC was not a primary requirement was it?

As I said before, its the NFB that needs to operate down to dc not the
circuit as a whole.

Neither
was it the only derived requirement for acceptable feedback.

What it is driving *does* alter whether the mu-follower
meets your primary requirement for a feedback circuit to be
acceptable.

What do you think is my primary requirement for NFB to be acceptable?

I guess about half of your mu-follows suffer
from one of the pitfalls of misused feedback that you
identify. If you daren't post your circuit, I guess I can
assume the worst.


Well, you have already seen one of my mu-follower circuits that I
published in the article I wrote about distortion in mu-followers. You
can find it he

http://www.ianbell.ukfsn.org/data/mu...distortion.pdf

I am sure I don't need to draw you the circuit of a mu
follower

It's *your* mu-follower I want to see, not some idealised
notion. In the context of the rest of *your* circuit.

and I am sure you know the only NFb it has is in the top
CF.

Yes. Try telling Patrick. Actually, don't. No point in
provoking another rash of drivel from a fellow trader
looking for a sales pitch.

Feedback in the equalisation circuit wouldn't be global,
That is an interesting assumption. There are indeed many
valve tone control circuits that use feedback around a
single CC stage (i.e local NFB

) but they all perform poorly in terms of distortion
because at maximum boost they have very little feedback.
To keep distortion at a low level needs more open loop
gain which means more than one stage and hence global
feedback.

From which I should take it that you mean local feedback is
round a single stage, whereas feedback round more than one
stage is global? That's not what is usually meant by global.

Actually, that is exactly what is generally meant by global feedback.

Global usually means round the entire circuit, from output
to input.

Not, it CAN mean that but not exclusively.

Global feedback round an equalisation circuit
surely defeats the object?


No, equalisation is achieved by changing the feedback.

Anyway, you haven't answered the question. When is feedback
OK and when is it not?

OK when local, not when global unless effective down to dc.

So far I might infer that you believe
that only DC-coupled feedback round a single stage is
acceptable. Is that correct?

No, but it is reasonably safe.


If so, your arguments fall a
long way short of legitimising your opinion. As I've said,
either get them right or leave them out and think of some
more attractive bull**** instead.

See above.


but
you appear to reject its use there as a matter of
principle.
So where, between global and local, does this principle
cease to apply? If there isn't a simple, sensible and
complete answer
to this question, I suggest you either abandon the
argument,
or make a mixer with zero feedback so you can honestly
say
so.
The principle is that NFB should extend down to dc,
whether NFB is local or global. You can make multi-stage
amplifiers where this is true - the Pultec mic pre is a
good example.

So why have you limited yourself to the most local feedback
possible?
That's another problem with what you call your
"philosophy": you imply that you have avoided more extensive
feedback because it is difficult to achieve. It appears
therefore that your "philosophy" is to do whatever is
easiest. Not befitting, as it were, of a Saville Row tailor.


The aim is to provide excellent performance. Traditionally, NFB has been
used to apparently provide the means to do this with relatively poor
active elements - a bit like making a purse out of a sow's ear.

By using a mu-follwer with an inherently low distortion tube I can avoid
all the major pitfalls of NFB and achieve the same or better
performance. That's not doing whatever is easiest, its good engineering.

Maybe anyone who objects to feedback doesn't need you to
tell them why. Just give the impression that you agree
with
whatever they already think. Say it strangles the sound,
music needs air to breathe and feedback consumes oxygen,
that kind of stuff.

As it stands, your "philosophy" has quite a few gaps, and
counterposing valves and opamps without mentioning
discrete transistors appears contrived.
In what way contrived? Op amps are merely some transistors
in a box.

Well that's another example of one of several species of
contrivance employed in your argument...or whoever's
argument it is. Presumably this is deliberate chicanery?

Perhaps I should offer a clue in case some dimwits are
reading. Had you counterposed valves and transistors, and
had I then objected to the absence of any mention of opamps,
then your response would have been legitimate.

Try drawing yourself a little Venn diagram and your error
should jump out at you.

http://en.wikipedia.org/wiki/Venn_diagram

Acquaintance with some rudiments of classical logic might
help, eg:

http://en.wikipedia.org/wiki/Syllogism


Sorry, that makes no sense to me. You need to explain it in layman's terms.

Cheers

Ian
Ian

[bigwig]

On Sep 18, 10:40*am, Ian Bell wrote:
Ian Iveson wrote:
Ian Bell wrote:

My full surname is Thompson-Bell which sounds a bit
'posher' don't you think.

Posh isn't what you want. Posh people are stupid consumers,
whereas you need to appear to be a clever provider; an
accomplished artisan; a grovelling servant to a master who
believes he decides what you make. These conflicting
requirements, requiring some pretences, are part of the
feeling of special relationship, like an in joke.

Yes, you are right. *Again, CTC is only a placeholder and if you have
any good ideas for a name I would be glad to hear them.





I guess "Confluence" is already taken too, but there's
Chinese words...or some more obscure language perhaps,
with a nice symbol. Japanese is handy because they use
both Roman script and symbols, and they are associated
with quality (perhaps their word for "entrance" as in way
in, rather than put into a trance).

We don't generally discuss projects without access to a
circuit diagram. Or at least we didn't until that mixer
thread mentioned money.

Indeed.

Feeding parasitic profiteers isn't very recreational, in my
view.

Well, it is unusual for someone to ask about commisioning a design on
this newsgroup so I can understand people's caution.





You need to work on the feedback argument I think. Why
is
it OK in the mu-follower (originated in 7th century
Damascus, apparently, BTW) when it's not acceptable in
an
equaliser circuit? It's not that I want to know
particularly...just that the question rather poses
itself, so you should answer it, or rely on stupid
buyers
who don't know how circuits work, which turns every
rationale into snake oil.

I tried hard to make it clear that it is "global"
negative
feedback that is to be avoided and the reasons why. The
only NFB in a mu follower is local to the top cathode
follower and operates down to dc so it avoids all the
problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input
Without a circuit diagram it's impossible to say whether
your use of the mu-follower suffers from the problem of
feedback that you identify. "...operates down to DC
so..." could be misleading, depending on what it's
driving, seems to me.
The local feedback in the top CF operates down to dc; what
it is driving does not alter that.

I said "misleading". Yes the feedback is direct, and in some
academic sense you could say it therefore operates at DC.

There's nothing academic about it.

But in reality the mu-follower itself doesn't operate at DC.

I never said it did. I said the NFB in the CF at the top of the
mu-follower operates down to dc. That's the only NFB in a mu-follower.

That's only one reason why your argument is misleading. When
you constructed or borrowed your agument, the ability to
operate at DC was not a primary requirement was it?

As I said before, its the NFB that needs to operate down to dc not the
circuit as a whole.

Neither

was it the only derived requirement for acceptable feedback.

What it is driving *does* alter whether the mu-follower
meets your primary requirement for a feedback circuit to be
acceptable.

What do you think is my primary requirement for NFB to be acceptable?

* I guess about half of your mu-follows suffer

from one of the pitfalls of misused feedback that you
identify. If you daren't post your circuit, I guess I can
assume the worst.

Well, you have already seen one of my mu-follower circuits that I
published in the article I wrote about distortion in mu-followers. You
can find it he

http://www.ianbell.ukfsn.org/data/mu...distortion.pdf





I am sure I don't need to draw you the circuit of a mu
follower

It's *your* mu-follower I want to see, not some idealised
notion. In the context of the rest of *your* circuit.

and I am sure you know the only NFb it has is in the top
CF.

Yes. Try telling Patrick. Actually, don't. No point in
provoking another rash of drivel from a fellow trader
looking for a sales pitch.

Feedback in the equalisation circuit wouldn't be global,
That is an interesting assumption. There are indeed many
valve tone control circuits that use feedback around a
single CC stage (i.e local NFB

) but they all perform poorly in terms of distortion
because at maximum boost they have very little feedback.
To keep distortion at a low level needs more open loop
gain which means more than one stage and hence global
feedback.

From which I should take it that you mean local feedback is
round a single stage, whereas feedback round more than one
stage is global? That's not what is usually meant by global.

Actually, that is exactly what is generally meant by global feedback.

Global usually means round the entire circuit, from output
to input.

Not, it CAN mean that but not exclusively.

* Global feedback round an equalisation circuit

surely defeats the object?

No, equalisation is achieved by changing the feedback.

Anyway, you haven't answered the question. When is feedback
OK and when is it not?

OK when local, not when global unless effective down to dc.

* So far I might infer that you believe

that only DC-coupled feedback round a single stage is
acceptable. Is that correct?

No, but it is reasonably safe.

* If so, your arguments fall a

long way short of legitimising your opinion. As I've said,
either get them right or leave them out and think of some
more attractive bull**** instead.

See above.







but
you appear to reject its use there as a matter of
principle.
So where, between global and local, does this principle
cease to apply? If there isn't a simple, sensible and
complete answer
to this question, I suggest you either abandon the
argument,
or make a mixer with zero feedback so you can honestly
say
so.
The principle is that NFB should extend down to dc,
whether NFB is local or global. You can make multi-stage
amplifiers where this is true - the Pultec mic pre is a
good example.

So why have you limited yourself to the most local feedback
possible?

*That's another problem with what you call your

"philosophy": you imply that you have avoided more extensive
feedback because it is difficult to achieve. It appears
therefore that your "philosophy" is to do whatever is
easiest. Not befitting, as it were, of a Saville Row tailor.

The aim is to provide excellent performance. Traditionally, NFB has been
used to apparently provide the means to do this with relatively poor
active elements - a bit like making a purse out of a sow's ear.

By using a mu-follwer with an inherently low distortion tube I can avoid
all the major pitfalls of NFB and achieve the same or better
performance. That's not doing whatever is easiest, its good engineering.





Maybe anyone who objects to feedback doesn't need you to
tell them why. Just give the impression that you agree
with
whatever they already think. Say it strangles the sound,
music needs air to breathe and feedback consumes oxygen,
that kind of stuff.

As it stands, your "philosophy" has quite a few gaps, and
counterposing valves and opamps without mentioning
discrete transistors appears contrived.
In what way contrived? Op amps are merely some transistors
in a box.

Well that's another example of one of several species of
contrivance employed in your argument...or whoever's
argument it is. Presumably this is deliberate chicanery?

Perhaps I should offer a clue in case some dimwits are
reading. Had you counterposed valves and transistors, and
had I then objected to the absence of any mention of opamps,
then your response would have been legitimate.

Try drawing yourself a little Venn diagram and your error
should jump out at you.

http://en.wikipedia.org/wiki/Venn_diagram

Acquaintance with some rudiments of classical logic might
help, eg:

http://en.wikipedia.org/wiki/Syllogism

Sorry, that makes no sense to me. You need to explain it in layman's terms.

Cheers

Ian



Ian- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -

Hello Ian,
Sorry I have missed all of this. I was giving RAT a break after
getting the usual abuse from the other(possibly slightly retarded)
Ian. Anyway I would like to congratulate you on your fortitude in
coming forth and and actually doing something. I have been spending
some time designing and building a switch mode supply that should be
suitable in different forms for use in almost every valve application.
On another point I have found a load of Schroff modular type
enclosures cheap as. If your interested.
Cheers Matt.

[bigwig]

On Sep 21, 12:27*am, bigwig wrote:
On Sep 18, 10:40*am, Ian Bell wrote:





Ian Iveson wrote:
Ian Bell wrote:

My full surname is Thompson-Bell which sounds a bit
'posher' don't you think.

Posh isn't what you want. Posh people are stupid consumers,
whereas you need to appear to be a clever provider; an
accomplished artisan; a grovelling servant to a master who
believes he decides what you make. These conflicting
requirements, requiring some pretences, are part of the
feeling of special relationship, like an in joke.

Yes, you are right. *Again, CTC is only a placeholder and if you have
any good ideas for a name I would be glad to hear them.

I guess "Confluence" is already taken too, but there's
Chinese words...or some more obscure language perhaps,
with a nice symbol. Japanese is handy because they use
both Roman script and symbols, and they are associated
with quality (perhaps their word for "entrance" as in way
in, rather than put into a trance).

We don't generally discuss projects without access to a
circuit diagram. Or at least we didn't until that mixer
thread mentioned money.

Indeed.

Feeding parasitic profiteers isn't very recreational, in my
view.

Well, it is unusual for someone to ask about commisioning a design on
this newsgroup so I can understand people's caution.

You need to work on the feedback argument I think. Why
is
it OK in the mu-follower (originated in 7th century
Damascus, apparently, BTW) when it's not acceptable in
an
equaliser circuit? It's not that I want to know
particularly...just that the question rather poses
itself, so you should answer it, or rely on stupid
buyers
who don't know how circuits work, which turns every
rationale into snake oil.

I tried hard to make it clear that it is "global"
negative
feedback that is to be avoided and the reasons why. The
only NFB in a mu follower is local to the top cathode
follower and operates down to dc so it avoids all the
problems cited.

Do you think I need to emphasise the distinction more?

Thanks for the input
Without a circuit diagram it's impossible to say whether
your use of the mu-follower suffers from the problem of
feedback that you identify. "...operates down to DC
so..." could be misleading, depending on what it's
driving, seems to me.
The local feedback in the top CF operates down to dc; what
it is driving does not alter that.

I said "misleading". Yes the feedback is direct, and in some
academic sense you could say it therefore operates at DC.

There's nothing academic about it.

But in reality the mu-follower itself doesn't operate at DC.

I never said it did. I said the NFB in the CF at the top of the
mu-follower operates down to dc. That's the only NFB in a mu-follower.

That's only one reason why your argument is misleading. When
you constructed or borrowed your agument, the ability to
operate at DC was not a primary requirement was it?

As I said before, its the NFB that needs to operate down to dc not the
circuit as a whole.

Neither

was it the only derived requirement for acceptable feedback.

What it is driving *does* alter whether the mu-follower
meets your primary requirement for a feedback circuit to be
acceptable.

What do you think is my primary requirement for NFB to be acceptable?

* I guess about half of your mu-follows suffer

from one of the pitfalls of misused feedback that you
identify. If you daren't post your circuit, I guess I can
assume the worst.

Well, you have already seen one of my mu-follower circuits that I
published in the article I wrote about distortion in mu-followers. You
can find it he

http://www.ianbell.ukfsn.org/data/mu...distortion.pdf

I am sure I don't need to draw you the circuit of a mu
follower

It's *your* mu-follower I want to see, not some idealised
notion. In the context of the rest of *your* circuit.

and I am sure you know the only NFb it has is in the top
CF.

Yes. Try telling Patrick. Actually, don't. No point in
provoking another rash of drivel from a fellow trader
looking for a sales pitch.

Feedback in the equalisation circuit wouldn't be global,
That is an interesting assumption. There are indeed many
valve tone control circuits that use feedback around a
single CC stage (i.e local NFB

) but they all perform poorly in terms of distortion
because at maximum boost they have very little feedback.
To keep distortion at a low level needs more open loop
gain which means more than one stage and hence global
feedback.

From which I should take it that you mean local feedback is
round a single stage, whereas feedback round more than one
stage is global? That's not what is usually meant by global.

Actually, that is exactly what is generally meant by global feedback.

Global usually means round the entire circuit, from output
to input.

Not, it CAN mean that but not exclusively.

* Global feedback round an equalisation circuit

surely defeats the object?

No, equalisation is achieved by changing the feedback.

Anyway, you haven't answered the question. When is feedback
OK and when is it not?

OK when local, not when global unless effective down to dc.

* So far I might infer that you believe

that only DC-coupled feedback round a single stage is
acceptable. Is that correct?

No, but it is reasonably safe.

* If so, your arguments fall a

long way short of legitimising your opinion. As I've said,
either get them right or leave them out and think of some
more attractive bull**** instead.

See above.

but
you appear to reject its use there as a matter of
principle.
So where, between global and local, does this principle
cease to apply? If there isn't a simple, sensible and
complete answer
to this question, I suggest you either abandon the
argument,
or make a mixer with zero feedback so you can honestly
say
so.
The principle is that NFB should extend down to dc,
whether NFB is local or global. You can make multi-stage
amplifiers where this is true - the Pultec mic pre is a
good example.

So why have you limited yourself to the most local feedback
possible?

*That's another problem with what you call your

"philosophy": you imply that you have avoided more extensive
feedback because it is difficult to achieve. It appears
therefore that your "philosophy" is to do whatever is
easiest. Not befitting, as it were, of a Saville Row tailor.

The aim is to provide excellent performance. Traditionally, NFB has been
used to apparently provide the means to do this with relatively poor
active elements - a bit like making a purse out of a sow's ear.

By using a mu-follwer with an inherently low distortion tube I can avoid
all the major pitfalls of NFB and achieve the same or better
performance. That's not doing whatever is easiest, its good engineering..

Maybe anyone who objects to feedback doesn't need you to
tell them why. Just give the impression that you agree
with
whatever they already think. Say it strangles the sound,
music needs air to breathe and feedback consumes oxygen,
that kind of stuff.

As it stands, your "philosophy" has quite a few gaps, and
counterposing valves and opamps without mentioning
discrete transistors appears contrived.
In what way contrived? Op amps are merely some transistors
in a box.

Well that's another example of one of several species of
contrivance employed in your argument...or whoever's
argument it is. Presumably this is deliberate chicanery?

Perhaps I should offer a clue in case some dimwits are
reading. Had you counterposed valves and transistors, and
had I then objected to the absence of any mention of opamps,
then your response would have been legitimate.

Try drawing yourself a little Venn diagram and your error
should jump out at you.

http://en.wikipedia.org/wiki/Venn_diagram

Acquaintance with some rudiments of classical logic might
help, eg:

http://en.wikipedia.org/wiki/Syllogism

Sorry, that makes no sense to me. You need to explain it in layman's terms.

Cheers

Ian

Ian- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -- Hide quoted text -

- Show quoted text -

Hello Ian,
* Sorry I have missed all of this. I was giving RAT a break after
getting the usual abuse from the other(possibly slightly retarded)
Ian. Anyway I would like to congratulate you on your fortitude in
coming forth and and actually doing something. I have been spending
some time designing and building a switch mode supply that should be
suitable in different forms for use in almost every valve application.
* On another point I have found a load of Schroff modular type
enclosures cheap as. If your interested.
Cheers Matt.- Hide quoted text -

- Show quoted text -

Sorry not "Ian" I meant Phillip the prick. Sorry to high jack your
post Ian, but I will no longer stand for this sort of MORON.

[Ian Iveson]

bigwig wrote:

Hello Ian,
Sorry I have missed all of this. I was giving RAT a
break after
getting the usual abuse from the other(possibly slightly
retarded)
Ian.

If that's aimed at me, it's a simple, backhanded, nasty lie,
and a sick little insult to boot. Feels like abuse but
perhaps you have an explanation for your outburst?

Ian

[Ian Bell[_2_]]

bigwig wrote:
Hello Ian,
Sorry I have missed all of this. I was giving RAT a break after
getting the usual abuse from the other(possibly slightly retarded)
Ian. Anyway I would like to congratulate you on your fortitude in
coming forth and and actually doing something. I have been spending
some time designing and building a switch mode supply that should be
suitable in different forms for use in almost every valve application.
On another point I have found a load of Schroff modular type
enclosures cheap as. If your interested.
Cheers Matt.

Thanks for the kind words Matt. A SMPSU for valves looks very
interesting. I considered using one for the heater supply of the mixer
as I need 12V at nearly 6amps.

I am REALY INTERESTED in the cheap Schroff modular enclosures. Please
sned more details.
Cheers

Ian

[mick]

On Thu, 10 Sep 2009 16:29:48 +0100, Ian Bell wrote:

snip

I am English and I have been trying to come up with a catchy name for
some time. CTC is just a placeholder right now. Actually, I do rather
like "Bespoke Valve Mixing Desks" - sounds like a cross between Neve and
Saville Row. May I use it?




Just a thought - Claribell Audio
:-)


--
Mick (Working in a M$-free zone!)
Web: http://www.nascom.info
Filtering everything posted from googlegroups to kill spam.

[Ian Bell[_2_]]

mick wrote:
On Thu, 10 Sep 2009 16:29:48 +0100, Ian Bell wrote:

snip
I am English and I have been trying to come up with a catchy name for
some time. CTC is just a placeholder right now. Actually, I do rather
like "Bespoke Valve Mixing Desks" - sounds like a cross between Neve and
Saville Row. May I use it?




Just a thought - Claribell Audio
:-)



Nice one Mick. A definite possibility.

Cheers

Ian