Sound Practices Mailing List Files - Volume 1
=========================================================================
From: JIM_FLOWERS@HP-USA-om24.om.hp.com
Subject: Re: Zen Amp
Date: Tue, 14 Oct 97 17:20:18 -0400
Source: Sound Digest Archive v01.n069
Item Subject: cc:Mail Text
Tom,
I built the Zen Preamp and I like it just fine - I might actually like
it more than I am willing to admit as I have convinced myself that I
prefer tubes over solid state. Incidentally, I built the Zen Preamp
into the chassis of what used to be a Threshold FET One Preamp.
I have only built a breadboard version of the "improved" Zen Power
Amp. I plan to try two other different circuits before committing to
a more "finished" version. One of the topologies will mimic the
2-stage approach used by Nelson Pass in his Aleph amplifier series.
Are you using a single MOS FET to carry the 3 amps or are you using
multiple (parallel) devices to share the current in the output stage?
According to the Pass Laboratories website, his amplifiers use
multiple devices. I don't know if this is for sonic reasons, device
longevity, or thermal distribution considerations.
JF
jim_flowers@hp.com
=========================================================================
From: "Hugh R. Dean" <aspen@alphalink.com.au>
Subject: Re: Zen Amp
Date: Wed, 15 Oct 1997 10:53:19
Source: Sound Digest Archive v01.n069
Hi Jim, at 05:20 PM 10/14/97 -0400, you wrote:
> Tom,
>
> I built the Zen Preamp and I like it just fine - I might actually like
> it more than I am willing to admit as I have convinced myself that I
> prefer tubes over solid state. Incidentally, I built the Zen Preamp
> into the chassis of what used to be a Threshold FET One Preamp.
I built a zen amp, then modified it, then cried over its low gain, then
added a 12AX7 SRPP for voltage amplification. Then I solved most of the
problems, converted the configuration from common source to source
follower, and then I found that the top end lacked the magic and no matter
what I did it could never match a good tube amp (yes, even a Cary 35W PP
with 6550s!!!). Very depressing.
> I have only built a breadboard version of the "improved" Zen Power
> Amp. I plan to try two other different circuits before committing to
> a more "finished" version. One of the topologies will mimic the
> 2-stage approach used by Nelson Pass in his Aleph amplifier series.
Yes, my approach as well, though I went for tube front end. If I did it
again, I would choose a cascaded 6SN7 for voltage amplification. I prefer
a simple source follower, not the common source config used by Pass, and
I'd be very surprised if he used a common source in the Aleph series. Zout
is very high.
> Are you using a single MOS FET to carry the 3 amps or are you using
> multiple (parallel) devices to share the current in the output stage?
> According to the Pass Laboratories website, his amplifiers use
> multiple devices. I don't know if this is for sonic reasons, device
> longevity, or thermal distribution considerations.
I think it's longevity, Jim. The 3A device needs about half Vcc, ie 25
volts, to satisfy the voltage and current demands of 30W into 8R. This is,
of course, 75 watts of heat, and on a CCR this is too much for a 150W
device if you use free air convection on even a sizeable heatsink. I chose
300mm x 150mm; they run about 30 degrees Celsius above ambient and at this
temperature the ion migration within the device will wear it out in about
five years at 20 hours per week (this is loosely based on MTBF of 100,000
hours at 25 degrees and the Arrhenius relation which halves semiconductor
life for each 10 degree rise in temperature. I make the derating factor
about 16 times as the device runs a heap hotter than the sink, giving about
6000 hours total.)
Incidentally, after you buy the sinks, machine and finish them to a decent
standard you might as well have bought a Chinese tube and a Hammond OPT.
Ho hum...
The device life is probably close to doubled if you therefore use two
devices to carry 3A. On a long heatsink, there are obvious thermal
advantages as well. I jury rigged a comparator and it is surprisingly easy
to match devices to within a couple of percent.
After all this stuffing around, I have come reluctantly to the conclusion
that a decent SE tube amp can be built for a comparable cost in stereo (my
Zens were monoblocks) and should knock the socks off it sonically. There,
I have said it - now I feel better.
Cheers,
Hugh R. Dean
Melbourne, Australia
=========================================================================
From: "Richard Jones" <RJ0101@Sprynet.Com>
Subject: Zen Amplifier
Date: Wed, 6 Jan 1999 20:23:45 -0600
Source: Sound Digest Archive v01.n560
This is a multi-part message in MIME format.
- ------=_NextPart_000_003C_01BE39B2.75A71A40
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hey folks... I realize this isn't tube based.... oh the horrors!!!! =
but...
I was wondering if any of you out there know if it is possible to build =
a SS SE class A amp with one MOSFET in the signal path with a constant =
current load, but without output coupling caps..
like the Zen amplifier without those 2 2200uf output caps..
see www.passlabs.com
any ideas???
txs
Richard.
- ------=_NextPart_000_003C_01BE39B2.75A71A40
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.72.2106.6"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>Hey folks... I realize this isn't =
tube based....=20
oh the horrors!!!! but...</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>I was wondering if any of you out =
there know if=20
it is possible to build a SS SE class A amp with one MOSFET in the =
signal path=20
with a constant current load, but without output coupling =
caps..</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>like the Zen amplifier without those =
2 2200uf=20
output caps..</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT><FONT size=3D2>see=20
www.passlabs.com</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>any ideas???</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>txs</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>Richard.</FONT></DIV></BODY></HTML>
- ------=_NextPart_000_003C_01BE39B2.75A71A40--
=========================================================================
From: "Norman Tracy" <ntracy@galstar.com>
Subject: Re: Zen Amplifier
Date: Wed, 6 Jan 1999 23:52:07 -0600
Source: Sound Digest Archive v01.n560
Richard,
As usual Nelson Pass can tell us "been there, done almost that". Look at
his 1977 Audio magazine article for guidance:
http://www.passlabs.com/audio77.htm
If we look at the schematic of this amp its output stage looks typical.
However look at the basic circuit in figure 1. Nelson describes it thusly:
"The basic circuit configuration is shown in Fig. 1, where an input
differential transistor pair drives a current-sourced transistor, forming
the two voltage-gain stages of the amplifier. The output of the second
voltage-gain transistor drives a triple emitter-follower output stage,
which provides a current gain of somewhat less than a million. The four
current sources in the circuit are used to simultaneously increase the
bandwidth and linearity, accomplishing this by idling semiconductors at
currents much larger than the currents required to drive the amplifier.
With the exception of the output stage, the gain transistors operate with
only small variations about their operating points."
Now in 1977 no one cared a flip about single ended as a means to our end or
marketing buzz word, so Nelson never bothers to call it "SE". It does look
to me like a start on what you describe below. Converting it to use MOSFETS
is all that remains. :-) Given the hefty power FETs now available it seems
doable to reduce the output devices to a single part sitting atop a CCS.
If by "one MOSFET in the signal path" you mean one lonely FET in the whole
amp, i.e. no input or driver stage. Well all you need is one big hefty
pre-amp! :-/ I recall CODA did something like this with one of their power
amps. A 19" wide 1U sized box held the input and driver stages while the
high current power supply and output stage lived in their own much larger
box.
Now the discussion about is the CCS in the signal path in such a simple amp
can begin. Too late to go there tonight. Will note in closing I tend to
agree with the argument that both series and parallel elements are really
in the circuit once we think about where the current loops go. On paper
schematics parts to ground end at the little arrow ground symbol. In the
real world current flows through that ground along with ultimately
everything else including the output.
happy listening
Norman Tracy
Audio Crafters Guild
ntracy@galstar.com
http://galstar.com/~ntracy/ACG/
voice 918-627-5878
fax 918-622-6398
- ----------
From: Richard Jones <RJ0101@Sprynet.Com>
To: Audio <sound@deliverator.io.com>
Subject: Zen Amplifier
Date: Wednesday, January 06, 1999 8:23 PM
Hey folks... I realize this isn't tube based.... oh the horrors!!!! but...
I was wondering if any of you out there know if it is possible to build a
SS SE class A amp with one MOSFET in the signal path with a constant
current load, but without output coupling caps..
like the Zen amplifier without those 2 2200uf output caps..
see www.passlabs.com
any ideas???
txs
Richard.
=========================================================================
From: Christian Rintelen <rintelen@datacomm.ch>
Subject: Re: Zen Amplifier
Date: Thu, 07 Jan 1999 09:29:27 +0100
Source: Sound Digest Archive v01.n560
Norman Tracy wrote:
> (snip) ...once we think about where the current loops go. On paper
> schematics parts to ground end at the little arrow ground symbol. In thereal
> world current flows through that ground along with ultimately everything
> else including the output.
Thank you very much for pointing that out. It had to be said!
Christian
=========================================================================
From: "Hugh R. Dean" <aspen@alphalink.com.au>
Subject: Re: Zen Amplifier
Date: Thu, 7 Jan 1999 20:16:14 +1100
Source: Sound Digest Archive v01.n560
This is a multi-part message in MIME format.
- ------=_NextPart_000_006E_01BE3A7A.93520640
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hey folks... I realize this isn't tube based.... oh the horrors!!!! =
but...
=20
I was wondering if any of you out there know if it is possible to =
build a SS SE class A amp with one MOSFET in the signal path with a =
constant current load, but without output coupling caps..
=20
like the Zen amplifier without those 2 2200uf output caps..
see www.passlabs.com
=20
any ideas???
=20
=20
Hi Richard,
=20
While in Oz we catch the late mail and two people (Norman Tracy and =
Christian Rintelen) have already pointed out the complete current loop =
of an audio signal from an output stage (of any persuasion), the point =
should be made that in the case of OCL circuits the forward signal =
passes direct to the load and the return path passes via the power =
supply caps whereas in Pass' Zen amp the forward signal passes via a cap =
to the load and one half of the signal cycle (the positive) returns via =
the power supply cap.
=20
Ergo, it probably does not matter whether you use a coupling cap or =
not. In truth, if you choose to use a coupling cap you will probably =
take a lot of care in selection (I use 8 x 220uF LXF Nichicon, 4 x 27uF =
Nichicon, and 10uF Arcotronics MKP). This will ensure it is almost =
transparent.
=20
The Zen amp uses a current source, as I recall. This current source =
supplies the positive half cycle by dint of the amplifying mosfet =
beneath partly switching off. Return current is through the Vcc power =
cap, which should be appropriately bypassed so that its self-inductance =
does not degrade HF sonics.
=20
If we get rid of the coupling cap, quite possible of course with a =
bipolar supply, then some sort of accurate servo is required to keep the =
output floating at zero volts wrt earth. There are many who feel servos =
are audible, so you are back where you started, robbing Peter to pay =
Paul.
=20
FWIW, I have built a Zen some years back, and eventually came to the =
conclusion that far better results could be obtained by turning the =
topology on its head. That is, source follower off the rail, with =
current sink below sitting at earth. This amp achieves 3dB points at =
25Hz and 85KHz using a tube front end. The amp is commercial, and sold =
as the Glass Harmony. Zout is 38 milliohms, OK for zero global =
feedback. There is absolutely no intermodulation distortion.
=20
Cheers,
=20
Hugh R. Dean
Aspen Amplifiers
Melbourne, Australia
- ------=_NextPart_000_006E_01BE3A7A.93520640
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type><!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 =
HTML//EN">
<META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<BLOCKQUOTE=20
style=3D"BORDER-LEFT: #000000 solid 2px; MARGIN-LEFT: 5px; PADDING-LEFT: =
5px">
<DIV><FONT color=3D#000000 size=3D2>Hey folks... I realize this =
isn't tube=20
based.... oh the horrors!!!! but...</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>I was wondering if any of you =
out there know=20
if it is possible to build a SS SE class A amp with one MOSFET in =
the signal=20
path with a constant current load, but without output coupling=20
caps..</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>like the Zen amplifier without =
those 2=20
2200uf output caps..</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT><FONT size=3D2>see=20
www.passlabs.com</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>any ideas???</FONT></DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV><FONT color=3D#000000>Hi Richard,</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV><FONT color=3D#000000>While in Oz we catch the late mail and =
two people=20
(Norman Tracy and Christian Rintelen) have already pointed out the =
complete=20
current loop of an audio signal from an output stage (of any =
persuasion),=20
the point should be made that in the case of OCL circuits the =
forward signal=20
passes direct to the load and the return path passes via the power =
supply=20
caps whereas in Pass' Zen amp the forward signal passes via a cap to =
the=20
load and one half of the signal cycle (the positive) returns via the =
power=20
supply cap.</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV><FONT color=3D#000000>Ergo, it probably does not matter whether =
you use a=20
coupling cap or not. In truth, if you choose to use a coupling =
cap you=20
will probably take a lot of care in selection (I use 8 x 220uF LXF =
Nichicon,=20
4 x 27uF Nichicon, and 10uF Arcotronics MKP). This will ensure =
it is=20
almost transparent.</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV><FONT color=3D#000000>The Zen amp uses a current source, as I=20
recall. This current source supplies the positive half cycle =
by dint=20
of the amplifying mosfet beneath partly switching off. Return =
current=20
is through the Vcc power cap, which should be appropriately bypassed =
so that=20
its self-inductance does not degrade HF sonics.</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV><FONT color=3D#000000>If we get rid of the coupling cap, quite =
possible=20
of course with a bipolar supply, then some sort of accurate servo is =
required to keep the output floating at zero volts wrt earth. =
There=20
are many who feel servos are audible, so you are back where you =
started,=20
robbing Peter to pay Paul.</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV><FONT color=3D#000000>FWIW, I have built a Zen some years back, =
and=20
eventually came to the conclusion that far better results could be =
obtained=20
by turning the topology on its head. That is, source follower =
off the=20
rail, with current sink below sitting at earth. This amp =
achieves 3dB=20
points at 25Hz and 85KHz using a tube front end. The amp is=20
commercial, and sold as the Glass Harmony. Zout is 38 =
milliohms, OK=20
for zero global feedback. There is absolutely no =
intermodulation=20
distortion.</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV><FONT color=3D#000000>Cheers,</FONT></DIV>
<DIV><FONT color=3D#000000></FONT> </DIV>
<DIV>Hugh R. Dean</DIV>
<DIV>Aspen Amplifiers</DIV>
<DIV>Melbourne, Australia</DIV>
<DIV><FONT =
color=3D#000000></FONT> </DIV></BLOCKQUOTE></BODY></HTML>
- ------=_NextPart_000_006E_01BE3A7A.93520640--
=========================================================================
From: "PEARL Cust Serv" <custserv@pearl-hifi.com>
Subject: Re: Zen Amplifier
Date: Thu, 07 Jan 1999 10:06:29 -0700
Source: Sound Digest Archive v01.n561
Richard wrote:
>Hey folks... I realize this isn't tube based.... oh the horrors!!!! but...
>
>I was wondering if any of you out there know if it is possible to build a
>SS SE class A amp with one MOSFET in the signal path with a constant
>current load, but without output coupling caps..
>
>like the Zen amplifier without those 2 2200uf output caps..
>see www.passlabs.com
>
>any ideas???
>
================
Dear Richard:
Remember that in almost all cases, signal current flows through the PS
caps.
Building a negative supply in order to loose the ouput cap will just
move the cap ta a different place in the circuit. Additionally, you'll be
faced the requirement that you keep the output @ 0Vdc and likely you'll have
to servo to accomplish this, imo and experience servos are grief looking for
a home...
Draw a schematic of your proposed idea and trace the current path,
you'll see... The same applies btw to directly coupled line stages...
Happy Trials
Bill Perkins, PEARL Inc.
2528 19 St. S.W., Calgary, Ab.,
Canada T2T 4X3
Ph. 1 403 244 4434 Fx. 1 403 244 9026
=========================================================================
From: "Bill Ceruzzi" <wceruzzi@ntplx.net>
Subject: Re: ZEN amp report
Date: Sun, 12 Oct 1997 16:04:01 -0400
Source: Sound Digest Archive v01.n065
Hi Tom,
I built the original and revised and liked the revised more. The
feedback improved thing for me.
Bill
- -----Original Message-----
From: Tom van der Laan <a.b.e.vanderlaan@student.utwente.nl>
To: sound@deliverator.io.com <sound@deliverator.io.com>
Date: Sunday, October 12, 1997 2:48 PM
Subject: ZEN amp report
>Hello all,
>
>Just thought I'd share an amp building experience. Last friday I finished
>my ZEN amps (two mono's).
>
>I built a somewhat mixed version of the ZEN. I used the orignal schematic,
>but since I have sufficient cooling I decided to run them at 3 ampere.
>Warmup takes about 1 hour and then the amp is about 45 degrees celcius. For
>this first version I used cheap components, metalfilm resistors,
>electrolytic coupling caps and standard copperwire. The power supply is a
>choke coupled pi filter (21mH 30.000uF 1mH 30.000uF) and dead silent (< 5mV
>hum).The schematics of the ZEN amps can be found on the Pass Labs Homepage
>
>And now to the important part. How does it sound? Well, my first impression
>is that they sound very good compared to my QUAD II's. The high's are very
>clean and natural. Through the midrange the zen's are neutral, slightly
>"cool" where the QUAD's are "warm". And the low end sounds warm and
>somewhat woolly, probably due to the high outpt impedance. The soundstage
>is also pretty OK.
>
>I am very interested to hear what other people think of the ZEN amps
>
>Tom
>
>
>
>
=========================================================================
From: Tom van der Laan <a.b.e.vanderlaan@student.utwente.nl>
Subject: ZEN amp report
Date: Sun, 12 Oct 1997 20:02:21 +0200
Source: Sound Digest Archive v01.n065
Hello all,
Just thought I'd share an amp building experience. Last friday I finished
my ZEN amps (two mono's).
I built a somewhat mixed version of the ZEN. I used the orignal schematic,
but since I have sufficient cooling I decided to run them at 3 ampere.
Warmup takes about 1 hour and then the amp is about 45 degrees celcius. For
this first version I used cheap components, metalfilm resistors,
electrolytic coupling caps and standard copperwire. The power supply is a
choke coupled pi filter (21mH 30.000uF 1mH 30.000uF) and dead silent (< 5mV
hum).The schematics of the ZEN amps can be found on the Pass Labs Homepage
And now to the important part. How does it sound? Well, my first impression
is that they sound very good compared to my QUAD II's. The high's are very
clean and natural. Through the midrange the zen's are neutral, slightly
"cool" where the QUAD's are "warm". And the low end sounds warm and
somewhat woolly, probably due to the high outpt impedance. The soundstage
is also pretty OK.
I am very interested to hear what other people think of the ZEN amps
Tom
=========================================================================
From: "Smith, Richard Todd" <SmithRT@lci.com>
Subject: Zen amps
Date: Mon, 2 Feb 1998 09:33:00 -0500
Source: Sound Digest Archive v01.n179
What's the going price for zen amps? How good are they?
=========================================================================
From: "Reid Welch" <rewelch@earthlink.net>
Subject: Zener-regging an AC heater rail?
Date: Fri, 23 Oct 1998 09:58:29 -0400
Source: Sound Digest Archive v01.n465
Another head-scratcher... i'd like to regulate a 6.3VAC heater rail from
line voltage variations. The impetus comes from a 1960 book by Toomer
entitled "how to get the most out of tubes". Toomer writes that a source
resistor in one leg of the secondary, followed by a zener across the whole
heater winding gives excellent regulation. And his accompanying graph does
indicate vast improvement. Beyond this there is no further data or
description, such as the correct zener breakdown voltage. FWIW, the Toomer's
simple line drawing doesn't indicate a center tapped of any sort.
So by what formula do I choose a zener?
Is it any different to the zener, if the heater winding is referenced to
ground by center tapped resistors instead of a grounded center tap?
TIA,
Reid
=========================================================================
From: Grego Sanguinetti <grego@lscpdx.latticesemi.com>
Subject: Re: Zener-regging an AC heater rail?
Date: Fri, 23 Oct 1998 15:02:54 -0700
Source: Sound Digest Archive v01.n465
On Fri, Oct 23, 1998 at 09:58:29AM -0400, Reid Welch wrote:
>
>
> Another head-scratcher... i'd like to regulate a 6.3VAC heater rail from
> line voltage variations. The impetus comes from a 1960 book by Toomer
> entitled "how to get the most out of tubes". Toomer writes that a source
> resistor in one leg of the secondary, followed by a zener across the whole
> heater winding gives excellent regulation. And his accompanying graph does
> indicate vast improvement. Beyond this there is no further data or
> description, such as the correct zener breakdown voltage. FWIW, the Toomer's
> simple line drawing doesn't indicate a center tapped of any sort.
>
> So by what formula do I choose a zener?
>
> Is it any different to the zener, if the heater winding is referenced to
> ground by center tapped resistors instead of a grounded center tap?
If I follow you correctly (questionable after today's lunch meeting B^)
this is a standard zener voltage divider regulator except used across
a floating AC signal. I am assuming by "source resistor" you mean a
series resistor?
YUK!
Ok now that I recovered from almost heaving my lunch. B^)
I sure hope I am missing something or missunderstanding you, which
I have done a few times lately... but hell, I have no pride...
Anyway. So we are forward biasing a zener. Not a good plan. Let's say
the rail with the resistor and the zener anode is the "positive" waveform
and the cathode is connected to the "negative" phase. I would presume
that if we looked at the waveform sampled across the zener we would
see a clipped sine wave whose positive peak is clipped across at
the Vz point, the reverse biased zener voltage, and the negative
half wave would be clipped at the forward voltage point, whatever
that happens to be. Probably pretty small. I suppose they specify this
in zener datasheets but it is not a primary specification, at least as
far as most applications are concerned. At any rate this is going to
be a very assymetric waveform. Kind of almost AC but well,...
It might be interesting to try for giggles but I would expect a
particularly ugly output. The essence of zener noise.
I guess it all depends on one's definition of "excellent regulation".
It does not pass mine. Kind of like powering your filaments with
a cheap 1970's "fuzz tone".
- -grego
=========================================================================
From: "Reid Welch" <rewelch@earthlink.net>
Subject: RE: Zener-regging an AC heater rail?
Date: Fri, 23 Oct 1998 22:33:25 -0400
Source: Sound Digest Archive v01.n465
>On Fri, Oct 23, 1998 at 09:58:29AM -0400, Reid Welch wrote:
>>
>>
>> Another head-scratcher... i'd like to regulate a 6.3VAC heater rail from
>> line voltage variations. The impetus comes from a 1960 book by Tomer
>> entitled "how to get the most out of tubes". Tomer stated that a source
>> resistor in one leg of the secondary, followed by a zener across the
whole
>> heater winding gives excellent regulation. And his accompanying graph
does
>> indicate vast improvement. There is no further data, such as the correct
>> zener breakdown voltage. FWIW,
>> Tomer's simple line drawing doesn't indicate a center tap.
>>
>> So by what formula do I choose a zener?
>>
>> Is it any different to the zener if the heater winding is referenced to
>> ground by center tapped resistors instead of a grounded center tap -or no
>> center tap at all as shown in Tomer's diagram)?
Grego replied:
>If I follow you correctly (questionable after today's lunch meeting B^)
>this is a standard zener voltage divider regulator except used across
>a floating AC signal. I am assuming by "source resistor" you mean a
>series resistor?
>
>YUK!
>
Indeed.
>Ok now that I recovered from almost heaving my lunch. B^)
>I sure hope I am missing something or misunderstanding you, which
>I have done a few times lately... but hell, I have no pride...
Evidently a misunderstanding here: I want to regulate a 6.3VAC heater
circuit, and don't care what the AC waveform looks like. This supply is for
a set of theremin tubes running at very high AC levels, not a MC preamp. And
besides that, all cathodes are IDH and grounded.
Proud to help heave your lunch, <g> I successfully employ conventional zener
shunt regulation on the _DC heater supply_ of the Acme phono preamp which
has 5842 cathodes all grounded . There is absolutely no audible zener noise
or sonic artifacts, although ugly sawtooth ripple is obvious on the scope.
This preamp can handle a 100 microvolt MC cartridge so if zener noise were a
problem I would hear a difference when unplugging the shunt unit. I don't
hear a difference.
>Anyway. So we are forward biasing a zener. Not a good plan. Let's say
>the rail with the resistor and the zener anode is the "positive" waveform
>and the cathode is connected to the "negative" phase. I would presume
>that if we looked at the waveform sampled across the zener we would
>see a clipped sine wave whose positive peak is clipped across at
>the Vz point, the reverse biased zener voltage, and the negative
>half wave would be clipped at the forward voltage point, whatever
>that happens to be. Probably pretty small. I suppose they specify this
>in zener datasheets but it is not a primary specification, at least as
>far as most applications are concerned. At any rate this is going to
>be a very asymmetric waveform. Kind of almost AC but well,...
>
>It might be interesting to try for giggles but I would expect a
>particularly ugly output. The essence of zener noise.
OK. I'm mailing you a GIF -the page with diagram-. Perhaps the author was
an opinionated incompetent and his circuit impossible.
Anyone who wants to see Tomer's zener page can have it for the asking.
>I guess it all depends on one's definition of "excellent regulation".
>It does not pass mine. Kind of like powering your filaments with
>a cheap 1970's "fuzz tone".
Well, I hope the AC zener rail works as the book says. Steady voltage is the
goal not a beautiful wave. Otherwise I can go for DC heating with zener
shunt a' la Acme- a more complicated second choice, since I do not need a
quiet supply.
Reid
=========================================================================
From: dslagle@earthlink.net (Dave Slagle)
Subject: Re: Zener-regging an AC heater rail?
Date: Fri, 23 Oct 1998 20:19:33 -0700 (PDT)
Source: Sound Digest Archive v01.n466
please note... 4 lines of quoted text...the rest is thought...
>> Another head-scratcher... i'd like to regulate a 6.3VAC heater rail from
>> line voltage variations.
a good idea... specially for those of us who use non deutschmann-dollar tubes...
it also strikes me as funny how people will talk about how running fils at
- -10% improves the sound... but as reid states tube life is a concern, and
our line voltages can vary 10% easily... and 10% on the line means 10%
across the board (regulation aside) it may all be BS, but if you state a
speciffic voltage for best tube sound... how do you control it....
>> The impetus comes from a 1960 book by Toomer
>> entitled "how to get the most out of tubes".
wait, this was written in the 60's??? did he know something about tubes
that the transistor people didn't... like maybe for the next 30 years no
quality tubes will be made so I better savr the ones I got???
I am assuming given your staunch defending of AC on DHT's that you are
referring to IHT's... at which point this may be a great way to save those
pricey 7788's 7119's 262A's, 437A's, and that special pair of mesh 27's...
given that none of these tubes will ever be available again... maybe
toomer's approach is valid
at what cost is the question.... I didn't want to try 83's cuz they were a
pain in the ass to use... hindsight being 20/20 I'm glad I did... now the
pain is a matter of fact, and I am glad to endure it... one of the costs of
better sound for me is using 83's... case closed.
I often wonder how much mis-information flows around the internet... coming
from authoratative..urp...jute...urp...mouths.. and how may valid ideas are
discounted for the wrong reasons... I hear zener and filaments in the same
sentence and think yuck but until I try it... maybe it is an easy way to
live with your favorite pair of tubes for a long time...
I mention 83, and people line up to tell me how noisy they are... I have
yet to hear that noise... the music is what I hear... I just trust my ears
as opposed to trusting something someone mis-read somewhere.
dave
=========================================================================
From: "Reid Welch" <rewelch@earthlink.net>
Subject: RE: Zener-regging an AC heater rail?
Date: Sun, 25 Oct 1998 01:45:42 -0400
Source: Sound Digest Archive v01.n466
Davey said:
<good stuff snipped>
>,,I am assuming given your staunch defending of AC on DHT's that you are
>referring to IHT's... at which point this may be a great way to save those
>pricey 7788's 7119's 262A's, 437A's, and that special pair of mesh 27's...
>given that none of these tubes will ever be available again... maybe
>toomer's approach is valid
Why yes! The post was about an interesting, simple method to heat signal
tubes and such with steady AC voltage.
Many think about heaters and filaments as if they were the same thing. They
are not the same thing as far as tubes go.
As a suggestion, all of us should follow the established DH/IDH terminology
in order to avoid confusion and bad postings.
- -Indirectly heated tubes use insulated resistance wire bundled or coiled in
proximity to a coated sleeve. The sleeve is heated to emitting temperature
by the hot tungsten wire. Electrical isolation between cathode and heater
sets these tubes apart from "Directly heated" tubes.
Therefore, "heater" and "heater regulation", as in my last post, indicate
indirectly heated tubes.
- -Directly heated audio tubes employ either nickel alloy resistance wire
coated with emitting oxide, or tungsten wire coated with a mono-atomic layer
of carburized thorium. Here again, the filament serves only as a heater by
which to activate the emitting substance. The heater and the cathode are
electrically connected. While this is a disadvantage in itself, for whatever
current heats the cathode _also influences the tube's operating
characteristics by causing varying biases along the long and sinuous
cathode_. there are compensating advantages to DH construction.
OTOH, compensating benefits of DH operation include better heating
efficiency, faster warm-up and a generally better distribution of emitter
surface. This helps to give power triodes lower distortion.
- -Summary: With few exceptions, hot-cathode tubes are all indirectly heated.
- -In distinction, tubes are sorted into one of the two classes.
- -Each class has a different name for the thermal element: "Heater" for IDH,
"filament" for DH.
Confusion arises by saying "I run my 5687 filament on DC". Please say
"heater".
Grego's bad reaction to my zener-regging query came because of his evident
confusion of "heater". Hence, this post is dedicated to Grego, with thanks
to Dave for the kick-start.
Reid
=========================================================================
From: Grego Sanguinetti <grego@lscpdx.latticesemi.com>
Subject: Re: Zener-regging an AC heater rail?
Date: Tue, 27 Oct 1998 10:32:57 -0800
Source: Sound Digest Archive v01.n470
On Fri, Oct 23, 1998 at 10:33:25PM -0400, Reid Welch wrote:
> >YUK!
> >
> Indeed.
Yes, after a three day weekend, it still offends me!
> Evidently a misunderstanding here: I want to regulate a 6.3VAC heater
> circuit, and don't care what the AC waveform looks like. This supply is for
No misunderstanding.
If you don't care what the waveform looks like then this will work, sort of.
I say sort of because it is a very poor AC regulator. The point of the article
was to just keep the voltage under control and the zener should do that.
The basic premise of keeping the voltage stable is reasonable.
As to your distinction between "filament" only referring to a DH tube and
a "heater" only referring to an IDH tube; That is the first I have heard
of that distinction. Seems like most people use one or the other for all
tubes. But fine, why not. My comments were about the supply, without regard
for what kind of tube it was supplying.
My concern is that the output waveform will look really bad. Obviously
this is less of a concern for an IDH. It's audibility will depend on
the tube and the circuit. In your case it could be fine?
> Proud to help heave your lunch, <g>
Thanks Reid, burp.
> I successfully employ conventional zener
> shunt regulation on the _DC heater supply_ of the Acme phono preamp which
> has 5842 cathodes all grounded . There is absolutely no audible zener noise
> or sonic artifacts, although ugly sawtooth ripple is obvious on the scope.
That is also DC. The primary waveform is single polarity and therefore the
only signal that forward biases the zener junction is the AC ripple component.
I have zener regulated many a DC supply, though I employ a ripple filter
first. I assume you also have a cap or choke before the zener?
I have an extremely noisy HP regulated supply that I used to power the filament
of the 5842 input tube of my PP6B4 amp while prototyping. It also didn't seem
to notice, although the filament waveform was awful to look at.
> This preamp can handle a 100 microvolt MC cartridge so if zener noise were a
> problem I would hear a difference when unplugging the shunt unit. I don't
> hear a difference.
Good, but we are not talking about the same supply or circuit. Also the
output of the AC-zener will be quite different and alot more ugly, it
will also change polarity. My guess is that it is alot worse, but will
it be audible?
> OK. I'm mailing you a GIF -the page with diagram-. Perhaps the author was
> an opinionated incompetent and his circuit impossible.
No, he just didn't care as much about the waveform as he did about the
voltage level. The implementation was secondary to his point.
> Well, I hope the AC zener rail works as the book says. Steady voltage is the
> goal not a beautiful wave. Otherwise I can go for DC heating with zener
> shunt a' la Acme- a more complicated second choice, since I do not need a
> quiet supply.
Try it and see, right? I could very well be condeming it too quickly, but
my gut still says yuk.
I have never checked but it would also be interesting to know if there is a
reliability issue with the zener used in a full AC configuration.
You might check with an apps engineer.
- -grego
=========================================================================
From: "Reid Welch" <rewelch@earthlink.net>
Subject: RE: Zener-regging an AC heater rail?
Date: Tue, 27 Oct 1998 21:07:36 -0500
Source: Sound Digest Archive v01.n471
Grego,
Points well taken in your letter appended below.
Concerning filament terminology- nowhere in my tube-book library is a
_heater_ type tube referred to as a filament type.
The distinction is one of precision and economy. We should not throw it
away. IDH tubes have heaters and DH tubes have filaments. To use "filament"
for both IDH and DH tubes lends confusion over which type of tube is under
discussion, and to a neophyte, only offers confusion over how the respective
types are heated or applied in circuit- like asking "how much ripple is OK
for a 12AU7 filament?"
So for these reasons I say a heater should remain a heater.
While talking up terms, "bias current"... I don't see anyone on the list
using it, thank goodness, but occasionally you do encounter audio writers
speaking of tubes that run with (say) "high bias current". I know what they
mean of course, I just hate the meaningless, misleading term which could be
"idle current" or "quiescent current", "standing current" and so on.
Curiously, the Hammond transformer catalog uses "bias current" in describing
the allowable DC in their products. From a _magnetics_ viewpoint i guess
there it makes some small sense since the iron is thereby "biased" up some
portion of the saturation curve.
back to the AC zener regulator in the old book:
Have received private correspondence including a really credible analysis by
Paul Joppa- From the concensus it appears certain that Tomer's AC zener
shunt is purely bogus. It just can't work -not at all-, as shown.
Tomer's book, "Getting the Most Out of Tubes" is a 1960 first printing,
first edition SAMS item. Perhaps the author took someone's word for the
workability. What an awful mistake to publish an impossibility as tried and
true- with regulation curves, no less. Was Tomer a Jute? I don't think so-
His credentials listed on the dust jacket:
"Starting as a helper in a radio store, (Mr. Tomer) progressed to test
equipment builder, laboratory technician, production foreman, design
engineer, chief engineer, director of industrial engineering, and director
of commercial engineering. Today he is manager of field engineering for CBS
Electronics, and recognized authority on tubes, transistors.... He holds
several patents in photography and color TV. Since 1944 his articles have
appeared in numerous trade journals. In addition, he has published twelve
papers on engineering and ham topics, is author of CBS Electronics' "Tech
Tips", and has written ten major engineering bulletins o tubes and
transistor. He also is co-author of the CBS Electronics Transistor Home
Study Course."
His book is chock full of interesting insights. Only, now, I don't know how
much of Tomer to believe. Too damn bad, 'cause it's fascinating text. Some
of the chapter headings:
Catastrophic failures- glass failures, heater failures, arcing, fixed versus
cathode bias
Degenerative failures- Gas, getters, spurious emissions, interelectrode
leakage, interface resistance, cathode depletion
Subjective failures- Hum, microphonics, noise
Selected and premium tubes- How tubes are selected, results of tube
selection, premium tubes, the multi-spec tubes, standardization and
reliability, how standards are set, universal tubes, the universal
commodity, defining reliability, reliability and standardization.
Predicting tube performance- life expectancy by application, tests for
initial performance, predicting life
Special purpose tubes- filamentary tubes, low voltage tubes, phototubes,
voltage regulator tubes, thyratrons.
Methods for lengthening tube life- measuring bulb temperature, dissipation
control, voltage and current regulation, low-voltage operation, mechanical,
maintenance.
(if anyone requests I can type in particular topics of interest)
Reid
>-----Original Message-----
>From: owner-sound@lists.io.com [mailto:owner-sound@lists.io.com]On
>Behalf Of Grego Sanguinetti
>Sent: Tuesday, October 27, 1998 1:33 PM
>To: sound@deliverator.io.com
>Subject: Re: Zener-regging an AC heater rail?
>
>
>
>On Fri, Oct 23, 1998 at 10:33:25PM -0400, Reid Welch wrote:
>> >YUK!
>> >
>> Indeed.
>
>Yes, after a three day weekend, it still offends me!
>
>> Evidently a misunderstanding here: I want to regulate a 6.3VAC heater
>> circuit, and don't care what the AC waveform looks like. This
>supply is for
>
>No misunderstanding.
>If you don't care what the waveform looks like then this will
>work, sort of.
>I say sort of because it is a very poor AC regulator. The point of
>the article
>was to just keep the voltage under control and the zener should do that.
>The basic premise of keeping the voltage stable is reasonable.
>
>As to your distinction between "filament" only referring to a DH tube and
>a "heater" only referring to an IDH tube; That is the first I have heard
>of that distinction. Seems like most people use one or the other for all
>tubes. But fine, why not. My comments were about the supply, without regard
>for what kind of tube it was supplying.
>
>My concern is that the output waveform will look really bad. Obviously
>this is less of a concern for an IDH. It's audibility will depend on
>the tube and the circuit. In your case it could be fine?
>
>> Proud to help heave your lunch, <g>
>Thanks Reid, burp.
>
>> I successfully employ conventional zener
>> shunt regulation on the _DC heater supply_ of the Acme phono preamp which
>> has 5842 cathodes all grounded . There is absolutely no audible
>zener noise
>> or sonic artifacts, although ugly sawtooth ripple is obvious on
>the scope.
>
>That is also DC. The primary waveform is single polarity and therefore the
>only signal that forward biases the zener junction is the AC
>ripple component.
>I have zener regulated many a DC supply, though I employ a ripple filter
>first. I assume you also have a cap or choke before the zener?
>
>I have an extremely noisy HP regulated supply that I used to power
>the filament
>of the 5842 input tube of my PP6B4 amp while prototyping. It also
>didn't seem
>to notice, although the filament waveform was awful to look at.
>
>> This preamp can handle a 100 microvolt MC cartridge so if zener
>noise were a
>> problem I would hear a difference when unplugging the shunt
>unit. I don't
>> hear a difference.
>
>Good, but we are not talking about the same supply or circuit. Also the
>output of the AC-zener will be quite different and alot more ugly, it
>will also change polarity. My guess is that it is alot worse, but will
>it be audible?
>
>> OK. I'm mailing you a GIF -the page with diagram-. Perhaps the
>author was
>> an opinionated incompetent and his circuit impossible.
>
>No, he just didn't care as much about the waveform as he did about the
>voltage level. The implementation was secondary to his point.
>
>> Well, I hope the AC zener rail works as the book says. Steady
>voltage is the
>> goal not a beautiful wave. Otherwise I can go for DC heating with zener
>> shunt a' la Acme- a more complicated second choice, since I do not need a
>> quiet supply.
>
>Try it and see, right? I could very well be condeming it too quickly, but
>my gut still says yuk.
>
>I have never checked but it would also be interesting to know if there is a
>reliability issue with the zener used in a full AC configuration.
>You might check with an apps engineer.
>
>-grego
>
=========================================================================
From: evaguido <EvaGuido@iaehv.nl>
Subject: Re: Zener-regging an AC heater rail?
Date: Wed, 28 Oct 1998 19:06:00 +0100
Source: Sound Digest Archive v01.n472
Now I would like to add my 2 cents
The described concept cuts sharp (current)edges and therefor produces more
RF products. Not my favourite. I would go for pure ac if possible. If not
use Schottky Barier diodes as they do not suffer from charge storage (no
P-N junction)
Guido
At 10:32 27-10-98 -0800, Grego Sanguinetti wrote:
>
>On Fri, Oct 23, 1998 at 10:33:25PM -0400, Reid Welch wrote:
>> >YUK!
>> >
>> Indeed.
>
>Yes, after a three day weekend, it still offends me!
>
>> Evidently a misunderstanding here: I want to regulate a 6.3VAC heater
>> circuit, and don't care what the AC waveform looks like. This supply is
for
>
>No misunderstanding.
>If you don't care what the waveform looks like then this will work, sort of.
>I say sort of because it is a very poor AC regulator. The point of the
article
>was to just keep the voltage under control and the zener should do that.
>The basic premise of keeping the voltage stable is reasonable.
>
>As to your distinction between "filament" only referring to a DH tube and
>a "heater" only referring to an IDH tube; That is the first I have heard
>of that distinction. Seems like most people use one or the other for all
>tubes. But fine, why not. My comments were about the supply, without regard
>for what kind of tube it was supplying.
>
>My concern is that the output waveform will look really bad. Obviously
>this is less of a concern for an IDH. It's audibility will depend on
>the tube and the circuit. In your case it could be fine?
>
>> Proud to help heave your lunch, <g>
>Thanks Reid, burp.
>
>> I successfully employ conventional zener
>> shunt regulation on the _DC heater supply_ of the Acme phono preamp which
>> has 5842 cathodes all grounded . There is absolutely no audible zener
noise
>> or sonic artifacts, although ugly sawtooth ripple is obvious on the scope.
>
>That is also DC. The primary waveform is single polarity and therefore the
>only signal that forward biases the zener junction is the AC ripple
component.
>I have zener regulated many a DC supply, though I employ a ripple filter
>first. I assume you also have a cap or choke before the zener?
>
>I have an extremely noisy HP regulated supply that I used to power the
filament
>of the 5842 input tube of my PP6B4 amp while prototyping. It also didn't seem
>to notice, although the filament waveform was awful to look at.
>
>> This preamp can handle a 100 microvolt MC cartridge so if zener noise
were a
>> problem I would hear a difference when unplugging the shunt unit. I don't
>> hear a difference.
>
>Good, but we are not talking about the same supply or circuit. Also the
>output of the AC-zener will be quite different and alot more ugly, it
>will also change polarity. My guess is that it is alot worse, but will
>it be audible?
>
>> OK. I'm mailing you a GIF -the page with diagram-. Perhaps the author was
>> an opinionated incompetent and his circuit impossible.
>
>No, he just didn't care as much about the waveform as he did about the
>voltage level. The implementation was secondary to his point.
>
>> Well, I hope the AC zener rail works as the book says. Steady voltage is
the
>> goal not a beautiful wave. Otherwise I can go for DC heating with zener
>> shunt a' la Acme- a more complicated second choice, since I do not need a
>> quiet supply.
>
>Try it and see, right? I could very well be condeming it too quickly, but
>my gut still says yuk.
>
>I have never checked but it would also be interesting to know if there is a
>reliability issue with the zener used in a full AC configuration.
>You might check with an apps engineer.
>
>-grego
>
>
Tim Reese