Sound Practices Mailing List Files - Volume 1
=========================================================================
From: "Mark Donen" <soledadd@worldnet.att.net>
Subject: la maison de l'audiophile system
Date: Sat, 6 Dec 1997 11:10:07 -0500
Source: Sound Digest Archive v01.n127
Jean-Michel salut de nouveau,
Thanks for the reply.
I did indeed hear the WE15A at lmda as well as at l'espace kiron. I also
heard the Sato system at lmda. I'd like to hear the full Onken system too
though. Still think the VOT/Dauphin/TAD/Onken tweeter system was more fun
and integrated.
However, could you advise on the crossover type and point for a 511/807
crossing to a 414 in a large bass reflex?
I would be interested to recieve any plans you might have for the dauphin
horn or similar. I think I saw you post that you had some.
Merci mille fois,
Mark
=========================================================================
From: ROBERT ROSENTHAL <RBTR@mail.rkd.snds.com>
Subject: Lambda model 29 tube regulated power supply
Date: Wed, 27 Aug 1997 14:33:41 -0500
Source: Sound Digest Archive v01.n003
Hey All,
Greetings ! I picked this all tube & oil cap Lambda #29
100-200v@100mA regulated ps w/nice honkers of
power tranny & choke & 1/8" steel faceplate! at a local
hamfest; it uses 6L6pair,6SJ7,0A4,6X5 & 5V4G.
Does anyone have any info to share on this? It looks
juicy enough to base a DHT stereo amp on.
Best regards, Rob
"US TUBE DUDES & DUDETEES GOTTA STICK
TOGETHER 'CUZ TOO MANY PEOPLE THROW OUR
TUBES AWAY, & THERE'r TOO MANY OF THOSE BAD
TRANSISTORS OUT THERE !"
=========================================================================
From: "Mark Donen" <soledadd@worldnet.att.net>
Subject: La nouvelle revue du son
Date: Thu, 9 Jul 1998 08:55:33 -0400
Source: Sound Digest Archive v01.n337
This is a multi-part message in MIME format.
- ------=_NextPart_000_000A_01BDAB17.55091CE0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
I can also translate an article from the revue du son if anyone is =
interested. Email or fax it to me. Should take longer than typing it in =
to the computer. Maybe we can eventually put up the translation and the =
scanned original pages to someone's web page?
Regards,
Mark=20
- ------=_NextPart_000_000A_01BDAB17.55091CE0
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 4.71.1712.3"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>I can also translate an article from =
the revue=20
du son if anyone is interested. Email or fax it to me. Should take =
longer than=20
typing it in to the computer. Maybe we can eventually put up the =
translation and=20
the scanned original pages to someone's web page?</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>Regards,</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 =
size=3D2>Mark </FONT></DIV></BODY></HTML>
- ------=_NextPart_000_000A_01BDAB17.55091CE0--
=========================================================================
From: "J. Gordon Rankin" <waudio@cinti.net>
Subject: Re: La nouvelle revue du son
Date: Thu, 9 Jul 98 11:26:05 -0500
Source: Sound Digest Archive v01.n337
>I can also translate an article from the revue du son if anyone is
>interested. Email or fax it to me. Should take longer than typing it in to
>the computer. Maybe we can eventually put up the translation and the
>scanned original pages to someone's web page?
Man don't those pictures say enough... Man what I would do with a few
million!
Gordon
=======> Wavelength Audio <=======
mailto:waudio@cinti.net
http://www.wavelengthaudio.com
ph.fx. 513-271-4186 USA
=========================================================================
From: "Johari Yip" <hfyip@pacific.net.sg>
Subject: Re: La nouvelle revue du son
Date: Fri, 10 Jul 1998 08:44:51 +0800
Source: Sound Digest Archive v01.n338
Hi Scott,
Maybe you can help with this?
Johari
- ----------
From: Mark Donen <soledadd@worldnet.att.net>
To: joelist <sound@deliverator.io.com>
Subject: La nouvelle revue du son
Date: Thursday, July 09, 1998 8:55 PM
I can also translate an article from the revue du son if anyone is
interested. Email or fax it to me. Should take longer than typing it in to
the computer. Maybe we can eventually put up the translation and the
scanned original pages to someone's web page?
Regards,
Mark
=========================================================================
From: "Le Cleac'h J.-M." <lecleach@cgi.ensmp.fr>
Subject: La Nouvelle Revue du Son / Special Issue No 1 / Infos
Date: Wed, 10 Jun 1998 18:11:02 +0200
Source: Sound Digest Archive v01.n308
Hello everyone,
30 fellows Joenetters have ordered to me copies of the Special Issue No 1
of La Nouvelle Revue du Son.
At the moment I could find only 20 copies of that issue. Most of them have
been posted on saturday by airmail so probably you are beginning to receive
it now.
I ordered others issues but strangely distributors don't want to order more
than 4 or 5 at one time (may be they fear to not be paid). So I have to run
several kilometers around my home to try to find the precious and rare
issues. I'll do my best to have the remaining orders satisfied as soon as
possible. (I'll receive 4 issues today)
Remember that's not a commercial deal (I am not in the audio business but
in education and research) but some help I give to you.
BTW: I yet advanced more than 2200FF (around $350), that's a lot for me!
Please, everyone, try to
pay your due as soon as possible. Thank you very much by advance!
Best regards and enjoy your Special Issue!
Jean-Michel Le Cleac'h, Paris, France
=========================================================================
From: "Ken Dangerfield" <bpyakd@mail.island.net>
Subject: Re: La Nouvelle Revue du Son / Special Issue No 1 / Infos
Date: Wed, 10 Jun 1998 19:36:57 -0800
Source: Sound Digest Archive v01.n308
On 10 Jun 98 Le Cleac'h J.-M. said:
> Remember that's not a commercial deal (I am not in the audio
> business but in education and research) but some help I give to you.
Which is very much appreciated!
> BTW: I yet advanced more than 2200FF (around $350), that's a lot for
> me! Please, everyone, try to pay your due as soon as possible. Thank
> you very much by advance!
In the mail to you tomorrow.
Thanks and regards,
Ken Dangerfield
=========================================================================
From: "P.A. & M. Almeflo" <sd@ljusdal.se>
Subject: RE: La Nouvelle Revue du Son / Special Issue No 1 / Infos
Date: Thu, 11 Jun 1998 22:03:53 +0200
Source: Sound Digest Archive v01.n309
Hi,
Le Cleac'h J.-M. wrote:
>
> Hello everyone,
>
.....
> 30 fellows Joenetters have ordered to me copies of the Special Issue No 1
> of La Nouvelle Revue du Son.
> Remember that's not a commercial deal (I am not in the audio business but
> in education and research) but some help I give to you.
Thank you very much for this praiseworthy initiative.
> BTW: I yet advanced more than 2200FF (around $350), that's a lot for me!
> Please, everyone, try to
> pay your due as soon as possible. Thank you very much by advance!
>
> Best regards and enjoy your Special Issue!
> Jean-Michel Le Cleac'h, Paris, France
I sent the money today and guess what, also received the Special Issue
today!
It was really full of inciting articles and photos. The text will be a
challenge though.
Thanks,
Per Arne
=========================================================================
From: Hartmut <Hartmut.Quaschik@mch.sni.de>
Subject: Re: La Nouvelle Revue du Son / Special Issue No 1 / Infos
Date: Mon, 15 Jun 1998 10:44:24 +0200
Source: Sound Digest Archive v01.n311
Jean-Michel,
I have got that special issue. It is great. Maybe I will try the
latest Kaneda Preamp. We have already tried the 1995 Kaneda preamp
(like presented in MJ June 95) and found it very favourable.
Most folks here in Munich changed their old L'Audiophile style
Kaneda to the newer one. I myself changed that 1995 circuit to
the topology of Jonathan Carr's Connoisseur preamp and found
that even better.
It seems that most of the installations described in the
special issue are inspired by Maison de
L'Audiophile. The high percentage of Onken and Goto drivers
hints at that. Never heard of Dupertuis turntables. They look
marvellous.
I would like to see _your_ installation ...
I have sent you - between two postcards - 100FF plus 1$ to
cover the 105FF costs of you.
Merci mille fois,
Hartmut from Munich
PS:
- - good news: I was at a Tory Amos live concert on
Saturday in Bonn and it was wonderful
- - bad news: I crashed my car on the motorway yesterday
and it is totally damaged.
=========================================================================
From: Philippe <Cadique@worldnet.fr>
Subject: Re: La Nouvelle Revue du Son / Special Issue No 1 / Infos
Date: Sun, 21 Jun 1998 10:42:30 +0200
Source: Sound Digest Archive v01.n317
Hi,
The bookshop I'm used to go, have two special issues La nouvelle revue
du son.
If somebody is interrested or if Jean-Michel need it, mail me.
Has I'm very busy at this time, be patient for the answer.
Philippe Camus
=========================================================================
From: Jeremy Epstein <70664.154@compuserve.com>
Subject: Large Advent Re-Suspension Project
Date: Fri, 19 Sep 1997 15:32:35 -0400
Source: Sound Digest Archive v01.n032
Joes:
I dumpster-dove a pair of Large Advents, the suspension is shot on both woofers. I see Parts Connect
ion has suspension repair kits for
20 bucks or so.
Is this worth it? I had a pair of these once, they were what I would think handy to have for a spare
system, given the investment hurdle
we're talking here.
Anybody ever re-suspension old speaks here? Is it a fairly reasonable DIY project?
- -j
PS : Is there something better I could concoct with these drivers, oh speaker gurus?
=========================================================================
From: Andre Jute <andre@indigo.ie>
Subject: Large stash of gennie 300B
Date: Sun, 12 Oct 1997 11:15:57 +0100 (BST)
Source: Sound Digest Archive v01.n065
Found on RAT:
- --------------------
Subject:
W.E. 300B's
Date:
12 Oct 1997 03:36:54 GMT
From:
ldaniel102@aol.com (LDaniel102)
Organization:
AOL http://www.aol.com
Newsgroups:
rec.audio.tubes
I recently bought six government surplus P.A. ampliers. Each had four engraved
base W.E. 300B's. Of the 24 tubes six had loose bases or indicated weak
emission on my tester. The other 18 seem strong. Can anyone help with
putting a value on these tubes as a lot?
- -----------------------
I'm just forwarding it from a public forum, so write directly to Mr Daniel
if you want to make an offer.
Andre
Andre Jute
andre@indigo.ie
Communication Jute
the main index to the pages we support for lovers of classical music, for
writers, and for audiophiles is at
http://www.foundmark.com/ComJute/ComJuteF1.html
=========================================================================
From: Call-NOW@1-602-267-9688
Subject: (Last Call) -Big Year End Sale!!!-
Date: Sun, 21 Dec 1997 17:18:18 -0600 (CST)
Source: Sound Digest Archive v01.n139
*********************************
Linko -- P.O. Box 66781 -- Phoenix, AZ 85082
Phone (602)267-9688 Fax (602)275-1868 1-11pm (MST)
- LAST CALL -- LAST CALL -- LAST CALL -
***************************************
Greetings Business Owners & Managers,
We welcome you to our "Years End Holiday Special"...." Sale of all
Sales". We are slashing the prices to the bare bones on all of our
Commercial Bulk E-mail Services and the prices of all our E-mail
lists. The prices stated below will remain in effect until midnight
January 1, 1998. If you would like to order please call Linko at
(602)267-9688 to discuss the details. We accept checks by phone or
fax. We will be raising our prices slightly in 1998. Order now!!!
********************
Commercial E-mail Service Rates:
We will mail your two page sales letter,(80 lines total), to our
best lists for the prices stated below. If you want us to send a
longer message please add 20% for each extra page.
Bulk e-mail to 100,000 General List.. Reg. $79.95.. Sale price $49.95
Bulk e-mail to 50,000 General List... Reg. $59.95.. Sale price $29.95
Bulk e-mail to 100,000 Biz-owners.. Reg. $199.95.. Sale price $129.95
Bulk e-mail to 50,000 Biz-owners....Reg. $129.95.. Sale price $69.95
Bulk e-mail to 50,000 MLM List... Reg. $139.95.... Sale price $79.95
Bulk e-mail to 25,000 MLM List... Reg. $99.95..... Sale price $59.95
Bulk e-mail to 5,000 MLM List.... Reg. $39.95..... Sale price $24.95
Bulk mail to custom lists is also available in some cases.
Please call to discuss building of special custom lists.
***** NOBODY BUT NOBODY BEATS OUR COMMERCIAL E-MAIL PRICES ****
*****************************************************************
*****************************************************************
GOOD QUALITY E-MAIL LISTS:
100,000 GENERAL E-MAIL ADDRESSES REG. $19.95.... Sale price $9.95
500,000 GENERAL E-MAIL ADDRESSES REG. $59.95.... Sale price $29.95
10,000 BIZ OWNERS E-MAIL ADDRESSES REG. $39.95... Sale Price $19.95
25,000 BIZ OWNERS E-MAIL ADDRESSES REG. $59.95... Sale Price $29.95
50,000 BIZ OWNERS E-MAIL ADDRESSES REG. $79.95... Sale Price $39.95
5,000 MLM E-MAIL ADDRESSES REG.. $29.95......... Sale Price $14.95
25,000 MLM E-MAIL ADDRESSES REG. $99.95......... Sale Price $49.95
50,000 MLM E-MAIL ADDRESSES REG. $139.95........ Sale Price $69.95
*ALL LISTS ARE MAILED U.S.P.S. WITH ONE ON A LINE IN .TXT FORMAT*
*** 1.4 MG - 3-1/2" DISKETTE ***
***** DON'T FORGET.... THIS SALE ENDS January 1, 1998 *****
****** YOUR BUSINESS IS APPRECIATED...... THANKS *******
******************************************************************
=========================================================================
From: "Mark Donen" <soledadd@worldnet.att.net>
Subject: Re: last message from a single
Date: Wed, 8 Jul 1998 14:11:31 -0400
Source: Sound Digest Archive v01.n336
Congrats, Hartmut,
Guess you'll be going converting from single ended to push pull now!
Hah, Hah
Regards,
Mark
PS I prefer PP myself (being married myself!)
=========================================================================
From: Grover Gardner <groverg@postoffice.att.net>
Subject: Re: last message from a single
Date: Wed, 8 Jul 1998 14:31:34 -0500
Source: Sound Digest Archive v01.n336
At 6:44 PM +0200 7/8/98, Hartmut wrote:
>Folks,
>
>as some of you already know, I am gonna marry on Friday.
>I will be on email again on July 20th and you will see,
>whether I will be the old dude or not ;-)
>
>Don't worry too much, my SO is a jazz music lover, so she
>is quite positive about my time comsuming DIY tube,
>loudspeaker and analoge interests.
>(But she likes JanGarbarek, which I don't 8-( )
>
>I have a lot of projects in schedule, mostly with speakers
>and tube pre and power amps. When I visited Thomas Mayer
>last Sunday, I liked his 211 Sakuma style power amps
>very much.
>
>Have a nice weekend,
>kind regars,
>
>Hartmut from Munich - soon married
Congratulations and best wishes! I'll forgoe the hoary old marriage
jokes--have a great time!
Grover Gardner
groverg@postoffice.att.net
=========================================================================
From: Roscoe Primrose <roscoe@aiko.com>
Subject: Re: last message from a single
Date: Wed, 08 Jul 1998 15:08:04 -0400
Source: Sound Digest Archive v01.n336
Mark Donen wrote:
>
> Congrats, Hartmut,
>
> Guess you'll be going converting from single ended to push pull now!
>
> Hah, Hah
>
Nahh, that's pushed & pulled!!!
>
> PS I prefer PP myself (being married myself!)
Me too, at least the married part ;-)
Peace
- --
Roscoe Primrose -- mailto:roscoe@aiko.com --
http://www.aiko.com/roscoe
"Once in a while you get shown the light
In the strangest of places if you look at it right." Robert Hunter
"640K ought to be enough for anybody" , Bill Gates, 1981
=========================================================================
From: Hartmut <Hartmut.Quaschik@mch.sni.de>
Subject: last message from a single
Date: Wed, 08 Jul 1998 18:44:46 +0200
Source: Sound Digest Archive v01.n336
Folks,
as some of you already know, I am gonna marry on Friday.
I will be on email again on July 20th and you will see,
whether I will be the old dude or not ;-)
Don't worry too much, my SO is a jazz music lover, so she
is quite positive about my time comsuming DIY tube,
loudspeaker and analoge interests.
(But she likes JanGarbarek, which I don't 8-( )
I have a lot of projects in schedule, mostly with speakers
and tube pre and power amps. When I visited Thomas Mayer
last Sunday, I liked his 211 Sakuma style power amps
very much.
Have a nice weekend,
kind regars,
Hartmut from Munich - soon married
=========================================================================
From: Bob Chernofsky <bobc@mci2000.com>
Subject: Last Tubes - Was 'Nuvistors'
Date: Wed, 05 Aug 1998 16:05:17 -0700
Source: Sound Digest Archive v01.n377
- -----Original Message-----
From: Ken Kessler <KSQUARE@compuserve.com>
To: Bob Chernofsky <bobc@mci2000.com>
Date: Wednesday, August 05, 1998 4:44 AM
Subject: Re: Last Tubes - Was 'Nuvistors'
OK, guys, I get the point. I should have written, 'possibly one of the last
tubes ever designed'.
Embarrassingly,
KK
=========================================================================
From: "Frank Deutschmann" <fdeutsch@blackrock.com>
Subject: Re: Last Tubes - Was 'Nuvistors'
Date: Wed, 5 Aug 1998 16:35:06 -0400
Source: Sound Digest Archive v01.n377
On Aug 5, 4:05pm, Bob Chernofsky wrote:
> From: Ken Kessler <KSQUARE@compuserve.com>
> To: Bob Chernofsky <bobc@mci2000.com>
> OK, guys, I get the point. I should have written, 'possibly one of the last
> tubes ever designed'.
No, you shouldn't have: we have well established that there is absolutely no
possibility that the 6CW4 nuvistor was one of the last tubes ever designed.
(For further clarification, I would direct you to our resident grammarian.)
For marketing hype, I would suggest something along the lines of, "nuvistors
are perhaps the epitome of vacuum tube technology" though that sort of thing
generally makes me quite ill when I see it....
- -frank
=========================================================================
From: Grego Sanguinetti <grego@lscpdx.latticesemi.com>
Subject: Re: Last Tubes - Was 'Nuvistors'
Date: Wed, 5 Aug 1998 15:48:01 -0700
Source: Sound Digest Archive v01.n378
> For marketing hype, I would suggest something along the lines of, "nuvistors
> are perhaps the epitome of vacuum tube technology" though that sort of thing
> generally makes me quite ill when I see it....
>
> -frank
Or, in my language;
Nuvisters are interesting little thingys that emitted from the junction
of a tube and a transistor. With the linearity of a tube but the bad
skin of a transistor, they have the encumbrance of a filament yet
wholly lack its cheery glow. They can still get up to the plate to bat
but can never get on base, yet are a must have for the serious collector.
- -grego
gee, should I add a smiley face?
=========================================================================
From: Paul Joppa <pdj@ISDNseattle.net>
Subject: Re: late widtwe
Date: Thu, 16 Jul 1998 09:35:55 -0700
Source: Sound Digest Archive v01.n345
Maybe I missed something earlier, but ... what is the load impedance
and current rating of the output (transformer, right?)? This is of
great interest to me, as my current setup runs 1200 ohms cathode
resistor and 55mA at about the same voltages (into a 3kohm output that
Mikey insists shoudn't have more than 60mA in it).
I note with interest that your max 800 ohm cathode resistor is only a
little lower than the 880 in a WE Model 91...
- -Paul Joppa
Danielak, Robert M wrote:
>
> hi guys,
>
> a bit late, but what (little) i did this past weekend:
>
> finally finished my stereo se 300B amp.
>
> this one uses the following tube line-up:
>
> 1/2 12sl7 => 1626 => 300B (Svetlana !)
>
> all cap coupled.
>
> 420V B+
>
> originally set it up for fixed bias and it sounded real good.
> real power and authority. figure up in the 8W range.
> (this was a big step from my usual 0.7W!!!!!!!)
>
> anyway, i decided to try cathode bias to see if it could be improved
> at the expense of reduced power level.
>
> i did have an unused 6.3V winding on the power tranny,
> so i hooked it in series with the primary (correct polarity)
> to boost the B+ a bit. think i got another 20V or so.
>
> a pair of 1k 25 w resistors were handy. these were obviously too high.
> plate current was around 60 -65mA, if i remember correctly.
>
> the sound was rather flat in comparison.
> so i decided to play with the cathode resistance.....
>
> i also had some 600 ohm 25W WW on hand.
>
> this was the other extreme, but i LIKED it.
>
> about 90 - 95mA thru the buggars. including cathode voltage and
> OPT DC drop i wound up with about 330V across the sv300B's
>
> this "low 'n' hot" operating point is
> apparently not a problem for these tubes.
>
> what a sound....
> now i understand what this L&H philosophy is all about!! ;)
>
> later i tried increasing the cathode resistance incrementally
> to see where the point of diminishing returns was for my system.
>
> i wound up with something like 800 ohms as the max "tolerable"
> cathode resistance.
>
> next experiment would be to try fixed bias again, but i think i've re-
> confirmed my preference for cathode bias.
>
> in any event, i have found that this 300B amp sounds *almost*
> as good as my 1626 amps, but with a bit more power ....
>
> maybe it has something to do with the 1626 driver ??? ;)
>
> gotta run,
>
> bob.d.
=========================================================================
From: "Danielak, Robert M" <robert.m.danielak@lmco.com>
Subject: late widtwe
Date: Thu, 16 Jul 1998 10:49:53 -0400
Source: Sound Digest Archive v01.n345
hi guys,
a bit late, but what (little) i did this past weekend:
finally finished my stereo se 300B amp.
this one uses the following tube line-up:
1/2 12sl7 => 1626 => 300B (Svetlana !)
all cap coupled.
420V B+
originally set it up for fixed bias and it sounded real good.
real power and authority. figure up in the 8W range.
(this was a big step from my usual 0.7W!!!!!!!)
anyway, i decided to try cathode bias to see if it could be improved
at the expense of reduced power level.
i did have an unused 6.3V winding on the power tranny,
so i hooked it in series with the primary (correct polarity)
to boost the B+ a bit. think i got another 20V or so.
a pair of 1k 25 w resistors were handy. these were obviously too high.
plate current was around 60 -65mA, if i remember correctly.
the sound was rather flat in comparison.
so i decided to play with the cathode resistance.....
i also had some 600 ohm 25W WW on hand.
this was the other extreme, but i LIKED it.
about 90 - 95mA thru the buggars. including cathode voltage and
OPT DC drop i wound up with about 330V across the sv300B's
this "low 'n' hot" operating point is
apparently not a problem for these tubes.
what a sound....
now i understand what this L&H philosophy is all about!! ;)
later i tried increasing the cathode resistance incrementally
to see where the point of diminishing returns was for my system.
i wound up with something like 800 ohms as the max "tolerable"
cathode resistance.
next experiment would be to try fixed bias again, but i think i've re-
confirmed my preference for cathode bias.
in any event, i have found that this 300B amp sounds *almost*
as good as my 1626 amps, but with a bit more power ....
maybe it has something to do with the 1626 driver ??? ;)
gotta run,
bob.d.
=========================================================================
From: "Danielak, Robert M" <robert.m.danielak@lmco.com>
Subject: RE: late widtwe
Date: Thu, 16 Jul 1998 13:26:46 -0400
Source: Sound Digest Archive v01.n345
hi paul,
no you didn't miss anything.
in my haste to get to a boring meeting to discuss the next generation
military satellite power system
i forgot to mention these details ;)
the OPT is a one-electron UBT-2 (4.8K nominal).
currently using the 16 ohm tap for a 2.4K equivalent load.
maybe i have this all mixed up,
but this would seem to be the best of both worlds (see caveats below)
given the higher primary inductance provided for the higher 4.8K primary
(like 40H). then utilizing the entire secondary (16 ohm tap)
pays you back in reduced leakage inductance....
this would apply to tapped-secondary OPT's, like the UBT's or tango 808s.
michael and brian's fine trannies generally use series/parallel secondary
connections which always take advantage of the entire secondary.
the UBT-2 is rated for 110mA DC. i believe the primary inductance is spec'ed
at this current level.
my only complaint with these trannies is the (understandably high)
DCR of about 460 ohm.
losing >15% of your B+ here is tough.
another 20% on a cathode resistor.....
so far the apparent volume from the low n hot cathode bias
isn't too far from the 8W fixed bias (at around 5W output).
i don't think i'd run my 300B's at almost 100mA indefinitely.
but it was an interesting experience.
i just checked my notes. i'm currently using 820 ohms.
btw: i'm sharing 410 ohms, equivalent, between two channels,
same CT filament,
with one bypass cap !!!!!! ;)
take care,
bob.d.
> ----------
> From: Paul Joppa[SMTP:pdj@ISDNseattle.net]
> Reply To: NO_SPAM_pdj@ISDNseattle.net_NO_SPAM
> Sent: Thursday, July 16, 1998 12:35 PM
> To: Danielak, Robert M
> Cc: sound@deliverator.io.com
> Subject: Re: late widtwe
>
> Maybe I missed something earlier, but ... what is the load impedance
> and current rating of the output (transformer, right?)? This is of
> great interest to me, as my current setup runs 1200 ohms cathode
> resistor and 55mA at about the same voltages (into a 3kohm output that
> Mikey insists shoudn't have more than 60mA in it).
>
> I note with interest that your max 800 ohm cathode resistor is only a
> little lower than the 880 in a WE Model 91...
>
> -Paul Joppa
>
> Danielak, Robert M wrote:
> >
> > hi guys,
> >
> > a bit late, but what (little) i did this past weekend:
> >
> > finally finished my stereo se 300B amp.
> >
> > this one uses the following tube line-up:
> >
> > 1/2 12sl7 => 1626 => 300B (Svetlana !)
> >
> > all cap coupled.
> >
> > 420V B+
> >
> > originally set it up for fixed bias and it sounded real good.
> > real power and authority. figure up in the 8W range.
> > (this was a big step from my usual 0.7W!!!!!!!)
> >
> > anyway, i decided to try cathode bias to see if it could be improved
> > at the expense of reduced power level.
> >
> > i did have an unused 6.3V winding on the power tranny,
> > so i hooked it in series with the primary (correct polarity)
> > to boost the B+ a bit. think i got another 20V or so.
> >
> > a pair of 1k 25 w resistors were handy. these were obviously too high.
> > plate current was around 60 -65mA, if i remember correctly.
> >
> > the sound was rather flat in comparison.
> > so i decided to play with the cathode resistance.....
> >
> > i also had some 600 ohm 25W WW on hand.
> >
> > this was the other extreme, but i LIKED it.
> >
> > about 90 - 95mA thru the buggars. including cathode voltage and
> > OPT DC drop i wound up with about 330V across the sv300B's
> >
> > this "low 'n' hot" operating point is
> > apparently not a problem for these tubes.
> >
> > what a sound....
> > now i understand what this L&H philosophy is all about!! ;)
> >
> > later i tried increasing the cathode resistance incrementally
> > to see where the point of diminishing returns was for my system.
> >
> > i wound up with something like 800 ohms as the max "tolerable"
> > cathode resistance.
> >
> > next experiment would be to try fixed bias again, but i think i've re-
> > confirmed my preference for cathode bias.
> >
> > in any event, i have found that this 300B amp sounds *almost*
> > as good as my 1626 amps, but with a bit more power ....
> >
> > maybe it has something to do with the 1626 driver ??? ;)
> >
> > gotta run,
> >
> > bob.d.
>
=========================================================================
From: Paul Joppa <pdj@ISDNseattle.net>
Subject: Re: late widtwe
Date: Thu, 16 Jul 1998 13:56:35 -0700
Source: Sound Digest Archive v01.n346
Thanks, that helps. Some comments below:
Danielak, Robert M wrote:
> ...snip...
> the OPT is a one-electron UBT-2 (4.8K nominal).
>
> currently using the 16 ohm tap for a 2.4K equivalent load.
>
> maybe i have this all mixed up,
> but this would seem to be the best of both worlds (see caveats below)
> given the higher primary inductance provided for the higher 4.8K primary
> (like 40H). then utilizing the entire secondary (16 ohm tap)
> pays you back in reduced leakage inductance....
1) Good approach - the One Electron units are a little light in the
loafers when it comes to henries. It moves the leakage inductance
problem down an octave, though these units are said to be well
interleaved so that might be OK. It also moves the parasitic
capacitance problem up an octave, which might be worthwhile ...
2) OK, your new operating point (350v/90mA if I remember) is about
right by my first-cut analysis: E/I = 3900 ohms = 5.2 rp , I say 5. RL
= 3.2 rp, I say 3. (both based on 750 ohms rp). The lower current
points would have greater distortion; they have greater efficiency but
at fixed voltage just use it to dissipate less rather than put more
out.
For an interesting comparison, you could try the 8-ohm tap (make the
load 4800 ohms) and go back to 55-60 mA plate current. This would be
closer to the same low distortion. I would predict, from a distortion
basis, that the lowest current for good sound would be lower than at
2400 ohms, and that the 60mA point would now sound as good as 90mA at
2400 ohms load - but data trumps theory every time, so I don't claim
this as a truth (yet!).
> ...more snip...
> my only complaint with these trannies is the (understandably high)
> DCR of about 460 ohm.
> losing >15% of your B+ here is tough.
> another 20% on a cathode resistor.....
Also it doubles the losses of the transformer ... which is already 0.8
dB or so, now closer to 1.5dB... still, if it SOUNDS better, that's a
small price to pay...
- -Paul Joppa, still sniffling )^;
=========================================================================
From: "Danielak, Robert M" <robert.m.danielak@lmco.com>
Subject: RE: late widtwe
Date: Thu, 16 Jul 1998 17:22:26 -0400
Source: Sound Digest Archive v01.n346
paul joppa wrote:
> Thanks, that helps. Some comments below:
>
> Danielak, Robert M wrote:
> > ...snip...
> > the OPT is a one-electron UBT-2 (4.8K nominal).
> >
> > currently using the 16 ohm tap for a 2.4K equivalent load.
> >
> > maybe i have this all mixed up,
> > but this would seem to be the best of both worlds (see caveats below)
> > given the higher primary inductance provided for the higher 4.8K primary
> > (like 40H). then utilizing the entire secondary (16 ohm tap)
> > pays you back in reduced leakage inductance....
>
> 1) Good approach - the One Electron units are a little light in the
> loafers when it comes to henries. It moves the leakage inductance
> problem down an octave, though these units are said to be well
> interleaved so that might be OK. It also moves the parasitic
> capacitance problem up an octave, which might be worthwhile ...
>
i would have thought the reverse.
using all of the secondary would have decreased the leakage inductance.
then, depending upon the interleaving arrangement, it would tend
to increase the capacitance
> 2) OK, your new operating point (350v/90mA if I remember) is about
> right by my first-cut analysis: E/I = 3900 ohms = 5.2 rp , I say 5. RL
> = 3.2 rp, I say 3. (both based on 750 ohms rp). The lower current
> points would have greater distortion; they have greater efficiency but
> at fixed voltage just use it to dissipate less rather than put more
> out.
>
yes. my L&H operating point was like 320V @ 95mA fwiw.
that's still way under the 40W rating. and within my 80% derating, too...
> For an interesting comparison, you could try the 8-ohm tap (make the
> load 4800 ohms) and go back to 55-60 mA plate current. This would be
> closer to the same low distortion. I would predict, from a distortion
> basis, that the lowest current for good sound would be lower than at
> 2400 ohms, and that the 60mA point would now sound as good as 90mA at
> 2400 ohms load - but data trumps theory every time, so I don't claim
> this as a truth (yet!).
>
yes indeed. i've tried the 4.8K load for the lower current cathode bias
(lower voltage) operating point as well as the lower current higher
voltage (fixed-bias) operating point.
so far in all cases i liked the 2.4k load.
funny, because i've always been a proponent of higher than "normal"
loading. i would have expected that i'd be one of the 5K-loaded-300B
crowd....
but let's not get into that argument again......
my mathcad/p-spice/excel spreadsheet analyses all show higher distortion
for the L&H case with the 2.4K. but i still think it sounds better.
non-technical ears agree... ;)
> > ...more snip...
> > my only complaint with these trannies is the (understandably high)
> > DCR of about 460 ohm.
> > losing >15% of your B+ here is tough.
> > another 20% on a cathode resistor.....
>
> Also it doubles the losses of the transformer ...
>
yeah, that's a pretty big hit when using a 2.4k load !
> which is already 0.8
> dB or so, now closer to 1.5dB... still, if it SOUNDS better, that's a
> small price to pay...
>
> -Paul Joppa, still sniffling )^;
>
oh no!
hope you don't have that funky chest cold that's running aroung the
west coast.
i just got back from SF yesterday. only stayed a couple days,
but it seemed like everyone i met was under the weather
(which was nice for a change -the weather, that is...)
bob.d.
=========================================================================
From: Acrosound@aol.com
Subject: Re: RE: late widtwe
Date: Thu, 16 Jul 1998 18:08:58 EDT
Source: Sound Digest Archive v01.n346
In a message dated 7/16/98 4:35:31 PM, you wrote:
Paul Joppa wrote:
>> > my only complaint with these trannies is the (understandably high)
>> > DCR of about 460 ohm.
>> > losing >15% of your B+ here is tough.
>> > another 20% on a cathode resistor.....
>>
>> Also it doubles the losses of the transformer ...
and Bob D. added;
>yeah, that's a pretty big hit when using a 2.4k load !
Two points about the losses...Paul points out that this high level of primary
DCR increases the DC voltage (the plate supply) across the winding...but it
also, if I may point out, increases the losses of the AC (the music) across
the primary as well....real power going up in heat.
Traditionally for a 5K ohm load you would want the primary resistance to be in
the range of 250 ohms or less....this will give you roughly a .5DB loss.
Double the resistance and double your insertion losses...and we haven't even
looked at the secondary copper circuit yet....
and going down to a 2.4K primary will further stress the efficiency of the
primary copper circuit...since at a lower primary impedance at any given powe
level your AC signal current will be much larger than at 4.8K. So if the DCR
leads to high losses at the nominal impedance of the unit....ratioing the
trans and cutting the effective primary impedance in half will again double
(as an approximation) your losses.
Again, as a reference, and to use a product other than my own. Tango in their
top of the line series generally sport insertion losses in the .25 to .3 db
range. Even their more "economy" U 808 has a loss figure (doing this from
memory) in the .65db range.
And this is not simply a matter of burning off a few volts here and there. If
you sweated bullets over building a low impeance power supply...and thought
that was important...if you discriminated btwn two different chokes solely on
the basis of which had the lower DCR (and the power supply filter choke does
not pass AC signals as it's dayjob :=) then perhaps you might also want to pay
some attention to these losses (and now they do raise the impedance of the AC
signal path).
And to be clear, no one in my opinion should make a purchase decision for an
ouptut based solely on which unit has the lowest DCR...and actually....one way
to get a really low DCR is to run the shit out of the unit by designing the
primary with very, very few turns....so, perhaps as Aristotle hinted at nearly
2 millineium ago, avoid either extremes or at least be in a position to
evaluate the whole of the design.
Mikey
=========================================================================
From: Acrosound@aol.com
Subject: Re: Re: RE: late widtwe
Date: Thu, 16 Jul 1998 18:29:47 EDT
Source: Sound Digest Archive v01.n346
In a message dated 7/16/98 5:20:52 PM, MSL wrote:
>Two points about the losses...Paul points out that this high level of primary
>DCR increases the DC voltage (the plate supply) across the winding..
hey folks, of course, minus the typo, the above should have read:
Two points about the losses.....Paul points out that this high level of
primary DCR increases the DC voltage *losses* (the plate supply) across the
winding......
sorry for the error.
Mikey
=========================================================================
From: An-shyang Chu <achu@sandia.gov>
Subject: Re[2]: late widtwe
Date: 17 Jul 1998 07:54:52 -0600
Source: Sound Digest Archive v01.n347
- ---------------------------- Forwarded with Changes ---------------------------
From: An-shyang Chu at po897cc2
Date: 7/16/98 12:23PM
To: robert.m.danielak@lmco.com at hubsmtp
Subject: Re[2]: late widtwe
- -------------------------------------------------------------------------------
The amp that I have taken up to Tom Ronan's place (companied by Steve Stanger)
also uses 820 ohms on the cathode (with Vp ~250V , Vk ~ 50V and Ip ~ 60mA). It
work really well with the new WE300B that he had. Not a lot of the power, but
nice sound.
hopper
_______________________________________________________________________________
......<snip>....
i just checked my notes. i'm currently using 820 ohms.
.
> > bob.d.
>
=========================================================================
From: "GREGORY MONFORT" <WINGRACER@classic.msn.com>
Subject: LD/CD Players
Date: Fri, 3 Oct 97 18:33:57 UT
Source: Sound Digest Archive v01.n056
I just bought a Pioneer Elite CLD-79 LD/CD player. The LD player is considered
to be one of the better ones, but what about the CD player? Other than my
CD/ROM I haven't listened to CD's.
Personal experience/opinions?
TIA,
GM
=========================================================================
From: "A.B." <bonaguid@dada.it>
Subject: Re: LD/CD Players
Date: Sat, 04 Oct 1997 16:09:01 +0100
Source: Sound Digest Archive v01.n056
Hi Greg
As I say in other circumstances, the sound
is a subjective question as the pleasantness of it.
If You are satisfied then ....it is a good sound.
If You have some doubt then You must do some comparison
with other CD player.
My opinion (and other opinions...) about Pioneer LC/CD
could be wrong while Your could be right. :-)
Andrea
- --------------------------------------------------------
At 18.33 03/10/97 UT, GREGORY MONFORT wrote:
>I just bought a Pioneer Elite CLD-79 LD/CD player. The LD player is
considered
>to be one of the better ones, but what about the CD player? Other than my
>CD/ROM I haven't listened to CD's.
>
>Personal experience/opinions?
>
>TIA,
>
>GM
>
>
=========================================================================
From: slagle@dtlweb.com (Dave Slagle)
Subject: LDs' Amp update (long)
Date: Tue, 16 Dec 1997 11:22:55 -0500 (EST)
Source: Sound Digest Archive v01.n136
>I burn up the tubes at 25-28 mA.
> Bias on the cathode is 2.1 VDC.
good for you, what shade of orange is that??
>I tried Kotei bias with a Nicad. This was real close to
>cathode bias arrangement; however, I couldn't quite get the operating
>point that I wanted. I had to play with the input taps quite a bit.
> Ended up at about 160VDC for B+ and the plate dissipation was 4.5 watts.
> This was quieter; however, in this application but it didn't have quite
>the same magic and space.
did you try a 2V lead acid cell... they are D cell sized with solder tabs,
and will give you 2.3V or so and let you cook the tubes good! I seem to
remember reid mentioning something from takuji about a preference for
nicad's but the voltage is right....
has anyone ever tried running a 5842 at say 240V and 20ma??? maybe stack 2
D cells and add B+ to get around 5W of plate diss... the -2.7V would be
nice to avoid the contact potenetial... and allow you more swing.... I
would also be interested to hear of the tube life... but the 5842's are
cheap enough... specially as a pototo... at $5 a pop you could go through
25 pair before you even approach a 300B....
dave
=========================================================================
From: ldmoore@lucent.com
Subject: RE: LDs' Amp update (long)
Date: 16 Dec 97 13:11:00 -0500
Source: Sound Digest Archive v01.n136
Shade of orange huh? Well, I really don't know. I know it is not the
shade of orange on those old guitar amps. Maybe I'll call it "retro lava
lamp" orange. Yea, that's the ticket.
I'll try the D cells tonight.
I'll try more voltage and less current, too. Hey, I buy these by the
handfulls, you bet I'll burn them up and try everything in the process.
L.D. Moore
-----Original Message-----
From: owner-sound [SMTP:owner-sound@lists.io.com]
Sent: Tuesday, December 16, 1997 11:23 AM
To: sound
Subject: LDs' Amp update (long)
>I burn up the tubes at 25-28 mA.
> Bias on the cathode is 2.1 VDC.
good for you, what shade of orange is that??
>I tried Kotei bias with a Nicad. This was real close to
>cathode bias arrangement; however, I couldn't quite get the operating
>point that I wanted. I had to play with the input taps quite a bit.
> Ended up at about 160VDC for B+ and the plate dissipation was 4.5
watts.
> This was quieter; however, in this application but it didn't have quite
>the same magic and space.
did you try a 2V lead acid cell... they are D cell sized with solder
tabs,
and will give you 2.3V or so and let you cook the tubes good! I seem to
remember reid mentioning something from takuji about a preference for
nicad's but the voltage is right....
has anyone ever tried running a 5842 at say 240V and 20ma??? maybe stack
2
D cells and add B+ to get around 5W of plate diss... the -2.7V would be
nice to avoid the contact potenetial... and allow you more swing.... I
would also be interested to hear of the tube life... but the 5842's are
cheap enough... specially as a pototo... at $5 a pop you could go through
25 pair before you even approach a 300B....
dave
=========================================================================
From: "Mark Donen" <soledadd@worldnet.att.net>
Subject: LE5-2
Date: Fri, 23 Jan 1998 08:18:44 -0500
Source: Sound Digest Archive v01.n168
Hi,
Does anyone one have TS parameters handy for JBL LE5-2 five inch midrange
drivers.
Any advice on upper and lower crossover points using 6db per octave slopes
in a three way?
TIA,
Mark
=========================================================================
From: Finn Hammer <f-hammer@post5.tele.dk>
Subject: Re: Leak _Through_line tuner
Date: Thu, 23 Oct 1997 21:40:35 +0200
Source: Sound Digest Archive v01.n082
To all who replied to this post:
Thanks for the enlightenment, both technical and
pronounciation/spelling-wise.
When I get the photofact sheet from Ed, It is going to a proper shop,
for adjustment. Don`t want to mess, (and wreck) with this one.
But one last thing: The insides of this unit looks so as to suggest that
it has spent a considerable amount of its time in a heavy smokers denn.
Any ideas how to best remove the goo, without stuffing the thing in the
dishwasher (and wreck it).(well, both, probably)
- --
HOME OF
Tractrix Horns, Tube Amps,
Fieldcoil Loudspeakers and Computer Graphics
http://home5.inet.tele.dk/f-hammer/
=========================================================================
From: davidbarnett@aristotle.net (David N. Barnett)
Subject: Re: Leak _Through_line tuner
Date: Fri, 24 Oct 1997 01:23:33 GMT
Source: Sound Digest Archive v01.n083
On Thu, 23 Oct 1997 21:40:35 +0200, Finn Hammer
<f-hammer@post5.tele.dk> wrote:
>Thanks for the enlightenment, both technical and
>pronounciation/spelling-wise.
And yet you STILL got it wrong -- that's "Trough," not "Through." <G>
- --dnb
=========================================================================
From: Thomas Danley <tom@ppci.com>
Subject: Re: Leakage Inductance and HF transfer in audio transformers.
Date: Sun, 30 Nov 1997 21:03:24 -0600
Source: Sound Digest Archive v01.n122
Finn Hammer wrote:
>
> Posted this one friday night, never saw it again, so I`l try again:
>
> I spent the last week in Sahara, it`s a great place to clear the mind:
> On the way back, I felt I suddenly understood why Leakage Inductance is
> such an ugly gremlin.
>
> I wrote it down, in case my mind should clutter by entrance to
> civilisation, and to check if it makes sense:
>
> Leakage Inductance and HF transfer in audio transformers.
>
> In the ideal transformer, the coupling between primary and secondary is
> total, so that a resistor across the secondary is transformed, thru` the
> turns ratio squared, into a resistive load at the primary.
>
> In a real transformer, the coupling btw. pri. and sec. is less than
> perfect, and the sum of the parts of the pri., and sec. inductances that
> don`t couple to the opposite, is called the leakage inductance.
>
> Since the leaking part of the pri. inductance is unable to _see_ the
> secondary, it does not take part in the transfer of the signal, which
> leaks, is lost, as part of the insertion loss.
>
> ----------------- But now the gremlin steps in ------------------------
>
> The real disadvantage of the leakage inductance, however, is that since
> it cannot see the secondary, it is also unable to obtain the resistive
> nature of the secondary`s resistive termination, and therefore it stays
> inductive.
>
> It doesn`t take part in the transfer of the signal, but acts as a choke
> in series with the signal.
>
> A reactance in series with the signal, with an impedance rising with
> frequency, attenuating HF, and in conjunction with the winding self
> capasitance, which acts from signal to ground, as R in an R-C filter,
> cutting off RF transfer in the transformer.
>
> Is this it, or am I missing something?
>
> Yours, finn
> --
> HOME OF
> Tractrix Horns, Tube Amps,
> Fieldcoil Loudspeakers and Computer Graphics
>
> http://home5.inet.tele.dk/f-hammer/
Hi All
This is a nice summary, I would add that leakage inductance is roughly
proportional to the "magnetic distance" between the primary and
secondary and this is why multi-filar windings have less leakage
inductance (primary and secondary are closer together) and also why low
impedance and/or higher frequencies are harder to deal with in this
regard.
Normally it is the leakage inductance that sets the high corner on
output transformers but parallel capacitance can be the limiting factor
too, in a large (250 watt) massivly multi-filar output transformer I
wound once the leakage L was very small but the parallel capacitance
dominated and set the corner. This was an interesting Xfm as the core
could be removed above 3KHZ with NO change in response, all coupling was
then wire to wire. The ideal transformer would have the series L and
parallel C. such that with the source/load impedance they both set the
rolloff corner, this would result in the maximum bandwidth (for that
volume of conductor).
Thomas Danley
Intersonics Technology Corp
=========================================================================
From: Finn Hammer <f-hammer@post5.tele.dk>
Subject: Leakage Inductance and HF transfer in audio transformers.
Date: Sun, 30 Nov 1997 22:22:39 +0100
Source: Sound Digest Archive v01.n122
Posted this one friday night, never saw it again, so I`l try again:
I spent the last week in Sahara, it`s a great place to clear the mind:
On the way back, I felt I suddenly understood why Leakage Inductance is
such an ugly gremlin.
I wrote it down, in case my mind should clutter by entrance to
civilisation, and to check if it makes sense:
Leakage Inductance and HF transfer in audio transformers.
In the ideal transformer, the coupling between primary and secondary is
total, so that a resistor across the secondary is transformed, thru` the
turns ratio squared, into a resistive load at the primary.
In a real transformer, the coupling btw. pri. and sec. is less than
perfect, and the sum of the parts of the pri., and sec. inductances that
don`t couple to the opposite, is called the leakage inductance.
Since the leaking part of the pri. inductance is unable to _see_ the
secondary, it does not take part in the transfer of the signal, which
leaks, is lost, as part of the insertion loss.
- ----------------- But now the gremlin steps in ------------------------
The real disadvantage of the leakage inductance, however, is that since
it cannot see the secondary, it is also unable to obtain the resistive
nature of the secondary`s resistive termination, and therefore it stays
inductive.
It doesn`t take part in the transfer of the signal, but acts as a choke
in series with the signal.
A reactance in series with the signal, with an impedance rising with
frequency, attenuating HF, and in conjunction with the winding self
capasitance, which acts from signal to ground, as R in an R-C filter,
cutting off RF transfer in the transformer.
Is this it, or am I missing something?
Yours, finn
- --
HOME OF
Tractrix Horns, Tube Amps,
Fieldcoil Loudspeakers and Computer Graphics
http://home5.inet.tele.dk/f-hammer/
=========================================================================
From: John Levreault <jlevro@shore.net>
Subject: Re: Leak/Garrard deal
Date: Fri, 20 Feb 1998 08:54:59 -0500
Source: Sound Digest Archive v01.n197
At 09:58 PM 2/20/98 +1000, alan cameron wrote:
>
>
>And also your thoughts on SME 3009 3 with 'spare carrying arm'.
>
>
Gorgeous looking arm. Sets up very easily. Removable arm tube is a real
convenience if you change cartridges often.
Sound is, to be kind, uninspiring, especially with low compliance/high mass
(i.e. most MC) cartridges. It might be OK with a low mass/high compliance
cartridge like a Shure, but then again....
JL
=========================================================================
From: alan cameron <alan@thehub.com.au>
Subject: Leak/Garrard deal
Date: Fri, 20 Feb 1998 21:58:01 +1000
Source: Sound Digest Archive v01.n197
joes,
I'm going to look at some stuff tomorrow;
Garrard 301 (no arm or cartridge specified)
Leak TL-12 monoblock
Leak Varioslope preamp
all in "fine original condition", to quote the ad.
How much should I offer, IY(H)O. (In the currency of your choice)
Opinions on the preamp, of which i know nothing.
And also your thoughts on SME 3009 3 with 'spare carrying arm'.
Currently using TD124/SME3009 2 (SME bearings in less than showroom
condition); shocking sand preamp; Quad II.
Thanks for BW, TIA for advice.
alan
=========================================================================
From: "Mark Donen" <soledadd@worldnet.att.net>
Subject: Leak's -fb theory from the 40s
Date: Sun, 4 Jan 1998 18:17:32 -0500
Source: Sound Digest Archive v01.n150
Hi,
Thought this might amuse if not enlighten.
Found a site with an essay on the Leak point one from the forties and the
great advantages of -fb.
Let us know what you think.
Regards,
Mark
http://werple.net.au/~kiewavly/PointOne2.html
=========================================================================
From: "Mark Donen" <soledadd@worldnet.att.net>
Subject: Re: Leak's -fb theory from the 40s
Date: Mon, 5 Jan 1998 11:25:02 -0500
Source: Sound Digest Archive v01.n151
I took a closer look at this Leak's pamphlet. He plays everything on the
one card of feedback -- his output stage is nontheless triode connected.
Then I asked why Leak developed such a great attachment to -fb. To give him
the benefit of the doubt: maybe there was no way of measuring higher order
distortion spectra in the 1940s?
However, I also notice that in the listening tests Leak's triode amp with
feedback was not compared to a triode amp without feedback. His proof is
only that his amp sounds pretty good compared to "live". It is also ironic
to notice that the audio "public" were pretty wise in the 1940s. The
accepted opinion of the time was that such low IM figures were irrelevant!
Regards,
Mark
=========================================================================
From: Scott Grammer <tubedude@cdc.net>
Subject: Leak Stereo 50 and Point One Pre
Date: Wed, 13 Jan 1999 21:08:26 -0500
Source: Sound Digest Archive v01.n573
Folks:
I would like to know your opinions of the value of the above mentioned
gear, in dead mint condition, with cable to connect them. All new tubes,
(Tesla).
Thanks!
S.G.
- --
TubeDude's Webpage of Vacuum Tube Lore! http://www.cdc.net/~tubedude/
Smoke Free Youth! http://www.cdc.net/~tubedude/smokefreeyouth.html
Nothing is so fairly distributed as common sense; No one thinks they
need any more than they already have.
- --Thomas Jefferson.
=========================================================================
From: Grover Gardner <groverg@postoffice.att.net>
Subject: Re: Leak Stereo 50 and Point One Pre
Date: Wed, 13 Jan 1999 22:12:20 -0500
Source: Sound Digest Archive v01.n574
At 9:08 PM -0500 1/13/99, Scott Grammer wrote:
>Folks:
>I would like to know your opinions of the value of the above mentioned
>gear, in dead mint condition, with cable to connect them. All new tubes,
>(Tesla).
The Stereo 50 is a lovely amp, sweet, smooth "British" sound (if I may be
permitted a generalization). The Point One is not up to today's preamp
standards--has a thin, peaky sound that detracts from the system. But if
you need a phono stage with curious, old-fashioned eq settings, you can use
it for that in a pinch. Unless the seller is asking a phenomenal price,
I'd pick up both, store the preamp for later resale or nostalgic reasons
and use the amp with pleasure.
Grover Gardner
groverg@postoffice.att.net
=========================================================================
From: Peter Allen <ptrallen@melbpc.org.au>
Subject: Leak Stereo 60 on the operating table
Date: Tue, 06 Jan 1998 10:17:15 +1100
Source: Sound Digest Archive v01.n151
Where is THE ULTIMATE LEAK STEREO 60 mod.?
Despairing in his sacred quest to achieve audio nirvana our hero stumbles
(AS IF by chance but IN ACTUALITY through synchronicity) across a VIRGIN
Leak Stereo 60. Blissed, bewitched, befuddled and bemused by this
fortuitous happenstance he seeks guidance from the Hi Fi Harpies. How
shall/or should/or can he, a mere audio mortal, transform this delicimo
tubo into the pulsating purist power amp of his dreams?
Contributions are invited. What needs must we do to transform this ancient
but noble audio artifact into a modern tube lover's delight? We have no
qualms about sacrificing some of its primal 30 watts rms, perhaps even 50%,
to attain QUALITY. Tube Gurus the time is NOW!. This tube loving novice
would appreciate your joint efforts to arrive at a consensus as to THE
ULTIMATE LEAK STEREO 60 mod and advise him accordingly.
Peter Allen, buddy of Matt Jelicich
matt.jelicich.ARL.bigpond.com.au
Peter Allen ptrallen@melbpc.org.au ( 8*{)
Member, Melbourne Audio Club http://www.vicnet.net.au/~macinc
=========================================================================
From: Finn Hammer <f-hammer@post5.tele.dk>
Subject: Leak Thruline tuner
Date: Wed, 22 Oct 1997 21:57:13 +0200
Source: Sound Digest Archive v01.n080
I have just found, and bought named tuner, and the first listening
impression of it is, that it keeps time very well. The rythm really is
extraordinary. This is in comparison to a Bang & Olufson Beomaster 5000
(the old one from the late 60`s, the one that came with the Beolab 5000
amplifier).
But the treble is awfull, like a tape that needs de-essing.
I need to be pointed in the right direction, towards someone who knows
this unit, and can supply me with the right printed matter, so that I
can have it adjusted properly.
Any takers? Finn
- --
HOME OF
Tractrix Horns, Tube Amps,
Fieldcoil Loudspeakers and Computer Graphics
http://home5.inet.tele.dk/f-hammer/
=========================================================================
From: davidbarnett@aristotle.net (David N. Barnett)
Subject: Re: Leak Thruline tuner
Date: Wed, 22 Oct 1997 23:21:24 GMT
Source: Sound Digest Archive v01.n080
On Wed, 22 Oct 1997 21:57:13 +0200, Finn Hammer
<f-hammer@post5.tele.dk> wrote:
>I have just found, and bought named tuner...
Uh, that's "Troughline." You know, "trough," like a feeding trough.
Not "through." The trough in question is a big copper ground bus
thingy.
- --dnb
=========================================================================
From: Scott Grammer <gesic@cdc.net>
Subject: Re: Leak Thruline tuner
Date: Thu, 23 Oct 1997 00:47:42 -0400
Source: Sound Digest Archive v01.n081
Finn Hammer wrote:
(snip)
>
> But the treble is awfull, like a tape that needs de-essing.
>
> I need to be pointed in the right direction, towards someone who knows
> this unit, and can supply me with the right printed matter, so that I
> can have it adjusted properly.
>
> Any takers? Finn
Finn:
I am not thouroghly familiar with this unit, although I know it by
reputation. Your choice of descriptive wording, specifically "de-essing"
makes me think the discriminator transformer (usually the last shielded
rf transformer before the audio section begins) is not centered on the
IF frequency. If you have a "twiddler" that fits the transformer
- -perfectly- then you might try first marking the core's original
position and then gently tweaking it about 10 degrees one way or the
other while listening. One way or the other should improve matters.
DON"T FORCE THE CORE! It is very fragile and if cracked the unit will
not be alignable.
Good luck!
S.G.
- --
"To save every cog and wheel is the first
precaution of intelligent tinkering." -Aldo Leopold.
Check out TubeDude's Webpage of Vacuum Tube Lore!
http://www.cdc.net/~gesic
=========================================================================
From: "Lynn T. Olson" <lynno@teleport.com>
Subject: Re: Leak Thruline tuner
Date: Thu, 23 Oct 1997 02:11:10 -0700
Source: Sound Digest Archive v01.n081
Don't forget that European/UK FM tuners use 50 uSec de-emphasis networks,
while North American & Japanese tuners use 75 uSec de-emphasis. If you use
an unmodified UK tuner in the US (aside from the 110/220 AC conversion),
it'll be a few dB bright in the treble, exaggerating an misalignment of the
IF strip or the discriminator. There is an RC circuit following the
discriminator, and that's the de-emphasis circuit. Find the correct values
for UK vs US usage.
Lynn T. Olson
E-mail: lynno@teleport.com
Web: http://www.teleport.com/~lynno/Ariel.htm
Editor, Valve & Tube News, and Associate Editor, Positive Feedback Magazine
=========================================================================
From: Scott Grammer <gesic@cdc.net>
Subject: Re: Leak Thruline tuner
Date: Thu, 23 Oct 1997 07:43:44 -0400
Source: Sound Digest Archive v01.n081
alan cameron wrote:
(snip)
A final thought comes to me - faulty alignment has not been a
significant
> issue with any radios I've repaired (probably only 30 or so, not a pro),
> even my WW II HRO comms receiver was quite adequately aligned 50 years
> later. Unless the radio had been 'fixed' before I got to it. Maybe youu've
> got some frequency determining components drifted in the IF or detector.
>
> Greetings from downtown Bris-Vegas,
> Queensland
> alan
Alan, Finn, et al:
Alan is most probably right. The twiddling of cores is not a procedure
for the faint of heart or heavy of hand. I think he will also agree that
faulty tubes can cause "misalignment" problems as well. Someone else
mentioned the 50uS eq of UK tuners. This could also cause a harsh sound.
I said all that to say this: Finn, you might be better off not twiddling
until you have tried all the other options. Sometimes, late at night,
when it's cold outside and the wind blows hard, I give advice that's
skewed towards the pro technician who may have performed similar
operations hundreds of times, and I forget that the person (s) I am
addressing might not have ever done it before. My Bad.
S.G.
- --
"To save every cog and wheel is the first
precaution of intelligent tinkering." -Aldo Leopold.
Check out TubeDude's Webpage of Vacuum Tube Lore!
http://www.cdc.net/~gesic
=========================================================================
From: alan cameron <alan@thehub.com.au>
Subject: Re: Leak Thruline tuner
Date: Fri, 24 Oct 1997 17:34:19 +1000
Source: Sound Digest Archive v01.n081
At 0:47 -0400 23/10/97, Scott Grammer is rumored to have typed:
> I am not thouroghly familiar with this unit, although I know it by
> reputation. Your choice of descriptive wording, specifically "de-essing"
> makes me think the discriminator transformer (usually the last shielded
> rf transformer before the audio section begins) is not centered on the
> IF frequency. If you have a "twiddler" that fits the transformer
> -perfectly- then you might try first marking the core's original
> position and then gently tweaking it about 10 degrees one way or the
> other while listening. One way or the other should improve matters.
> DON"T FORCE THE CORE! It is very fragile and if cracked the unit will
> not be alignable.
>
> Good luck!
> S.G.
Hi Scott, Finn, Joes
NB 'Mark The Original Position'. Can't think of how many radio's I've come
across that have been 'fixed' by enthusiastic twiddling of the IF slugs,
till the things crack, and fall out.Now a radio with (at least) two
problems. A phenomenon I'm sure Scott also is most familiar with.
The ideal 'twiddler' is a plastic rod with a spot of brass on one end and
a bit of ferrite on the other. By inserting one or other end into the IF
tranny, you can raise(ferrite) or lower (brass) the coil inductance,
teling you if either is infact an improvement. (such devices readily
available through any RF oriented electronics store)
May I reccomend the chapters of the ARRL or RSGB handbooks relating to FM
reception as a guide to IF tuning if you want to do that. Ideally you'd
have an RF sig generator and RF voltmeter to do this.
A final thought comes to me - faulty alignment has not been a significant
issue with any radios I've repaired (probably only 30 or so, not a pro),
even my WW II HRO comms receiver was quite adequately aligned 50 years
later. Unless the radio had been 'fixed' before I got to it. Maybe youu've
got some frequency determining components drifted in the IF or detector.
Greetings from downtown Bris-Vegas,
Queensland
alan
=========================================================================
From: "Bottlehead" <Bottlehead@prodigy.net>
Subject: Re: Leak Thruline tuner
Date: Thu, 23 Oct 1997 09:11:18 -0700
Source: Sound Digest Archive v01.n082
>
If you have a "twiddler" that fits the transformer
> > -perfectly- then you might try first marking the core's original
> > position and then gently tweaking it about 10 degrees one way or the
> > other while listening. One way or the other should improve matters.
> > DON"T FORCE THE CORE! It is very fragile and if cracked the unit will
> > not be alignable.
> >
> > Good luck!
Hey, let's all get the techno-terminology straight.
The plastic thingie is a diddlestick, not a twiddler.
The do-dad with the brass and iron ends is called, I think, a magic wand.
Anybody know?
Doc B.
=========================================================================
From: Scott Grammer <gesic@cdc.net>
Subject: Re: Leak Thruline tuner
Date: Thu, 23 Oct 1997 12:32:23 -0400
Source: Sound Digest Archive v01.n082
Bottlehead wrote:
>
(snip)
> Hey, let's all get the techno-terminology straight.
> The plastic thingie is a diddlestick, not a twiddler.
> The do-dad with the brass and iron ends is called, I think, a magic wand.
> Anybody know?
>
> Doc B.
Now Doc, surely you know that "Doodad" is spelled with two o's and is
not hyphenated! You are, in the immortal words of the great classic
thespian, Curly Howard, "Moidering de King's English"!
:-)
Where you been, anyhoo?
S.G.
- --
"To save every cog and wheel is the first
precaution of intelligent tinkering." -Aldo Leopold.
Check out TubeDude's Webpage of Vacuum Tube Lore!
http://www.cdc.net/~gesic
=========================================================================
From: Bob Danielak <rdaniela@ew.lmms.lmco.com>
Subject: Re: Learner designer - VT-52 amp
Date: Fri, 29 Aug 1997 17:21:23 -0400
Source: Sound Digest Archive v01.n005
hi bart
se comments below:
BART wrote:
>
> Hi Guys,
>
> I have put my toes in the water to design (as opposed to build) my first
> amp.
>
> I am planning a VT52 amp, direct coupled at first and maybe later
> transformer coupled.
>
> I would like the simplest possible circuit with a single tube voltage
> amplifier / driver.
>
have you seen jc's "micro". uses parallel 6sl7 direct coupled to 2a3.
this is about as simple as you're gonna get. this could be applied
to the VT52. let me know if you need the schematic.
but see further comments below...
> The candidates I have found as the VA are 12AX7 / 7025, 6SL7 or 6F5 / 6SF5.
> The first two seem obvious choices to give the required gain and be able to
> swing the required volts. Any reason not to parallel the twin tubes in a
> monoblock?
>
the main perceived problem with these tubes is that with
high gain comes high plate resistance. this and the relatively
low plate currents coupled with the relatively high input capacitance
of power triodes (near 100pF or more) can lead to poor hf response
related to
slew rate. this should be most noteable at higher drive levels.
i must say that i've experimented with this type of driver into
a 2a3 and didn't notice a lack of high end, per se.
just wasn't thrilled with the overall sound.
i later changed to a choke loaded direct coupled driver
using parallel 6sl7's and a 90H/25mA choke. this was pretty
good. didn't even notice that bass was rolled off below 36hZ...
that's another issue, which is the subject of one of your next
questions...
> I am surprised I haven't heard more about the 6F5 before. Any comments from
> anyone or advice on other alternatives? What about a paralleled 6SC7?
>
probably because everyone lumps the 6F5 in with 12ax7.
these are generally not considered "hi-fi" for various reasons.
a lot of it is the distorting guitar-amp stigma. but for the reasons
cited above, it generally does not make a good driver stage.
it is ok for low level input stages, but then you have to worry
about noise if it is a very low level stage -or- input headroom
if it is a higher (line) level stage...
the 6sc7 is similar to the 6sl7. the 'sl7 looks a little more linear
to me, and it's rp is a wee bit higher, but i may be splitting
hairs.
> I am thinking of a 100K pot input with a Nicad bias and a choke load. I am
> thinking of using a TFA204 output trannie as the load. Is this suitable? It
> will only have between 1 to 3ma through it so I guess I could use any old
> PP output trannie. Should I terminate the secondary with a load to use a
> trannie as an audio choke? 8ohms?
>
is that ni-cad in the grid circuit or in the cathode?
in any event, most hi-mu tubes are normally biased with <1V
at plate voltages < 150V or so. (refer to 6sc7 curves for example)
i guess you could do Ebo=100V, Ibo=0.5mA, and Eco=-1.2.
but that's awfully low current (even doubled for parallel operation)
to drive a power triode. it's also still in the curved regions
of the plate curves, fwiw.
to get a more respectable operating point, you'd need 150V or more
to get about 1.25mA per section....
this is an issue when it comes to direct coupling, which i'll explain
in a few...
as far as choke loading is concerned...
in order to get good low frequency response you need high inductance
and/or low plate resistance. the low frequency cut-off will be
at f = rp /(2*pi*L)
so for paralleled 6sc7 (rp ~ 50K) the effective rp = 25K
if you want f = 20Hz, then L req'd is about 200H!
don't think the primary of that tranny is that high.
remember, i said before that parallel 6sl7 (22K) into 90H gave me
a cut-off of about 36Hz...
a BIG pp tranny might be close to 200H with no DC (don't have
a real good feel here, since primary inductance is generally not
specified for pp OPT's)...
i plan to try it soon (although i'm doing a pp amp, and will be
using the primary as a split choke)...
as for loading the secondary, my first instinct would be NOT to
load it (or maybe with a relatively high impedance >>8ohms).
if you load it with 8 ohms it will reflect that back to the primary
as a 5K (or whatever the normal impedance is) load on the driver
tube. this sorta defeats the purpose of choke loading. the choke
is supposed to present a very high ac impedance load to the driver
tube. this results in a flatter load-line (constant-current)
and greater linearity.
btw: i have toyed with the idea of <selectively> loading the secondary
of a transformer used as a choke load. for example,
i came up with an RIAA network that seems to work this way.
don't know how it would work in practice. probably wreak havoc
with the poor driver tube -load impedance dancing all over the place!
> If I want 250 volts on the plate would the choke drop only about 5volts
> from the B+? Given this I plan to arrange to float the VT-52 filament at
> 195 volts
> ( I dont have any curves or figures for this tube - can anyone supply
> please?
> At this stage I am assuming a bias of about -55 volts for example.
> Would I then need about 470 on the plate. I plan to use an FS030 output
> trannie so would I need about 480 B+?
>
something's a little mixed up here, but 1st things 1st:
the tranny shouldn't drop hardly anything with 3mA thru it
maybe 1 to 1.5V max...
ok... now
if you're direct coupling, you don't have to "float" the filament.
it will find it's own level depending on the grid voltage (equal
to the driver plate dc voltage) and the required output bias voltage.
you must subtract the negative bias voltage to the grid (driver plate)
voltage
(looks like you added...)
so if your driver plate voltage is 250V and your vt52 bias
winds up being -55V, your cathode voltage is 250 - (-55) =305V!
now you must add the desired vt52 plate voltage to come up with the B+
requirement. i'd guess you want to run them at 250V to 275V.
so B+ of 555 to 580V is required.
now you see why most direct coupled amps run the input stage at
90V or so.
although, 150V on the driver plate would result in a B+ of about
450V to 480V just like you mention.
and yes, -50 to -55V is a good number for 45 bias.
i have the '45 data sheet if you need them (.gif format).
doesn't someone have them out on the web (Dan Cheever?
try: http://web.mit.edu/cheever/www/45curve.jpg
i'll e-mail mine to you if you wish...
> I am thinking of a 5R4 rectifier . No figures for this either (I only have
> an RC-19) How many volts will this drop? I am thinking of a CLC filter with
> 100mic caps and a big choke. Should I really be thinking choke input? Am I
> right in thinking about a 1,000V CT power trannie? Or more for choke input?
>
5r4 is grossly overrated for this application, but that might
be a good thing! ;) a 5y3 or 5v4 will do just fine.
the 5r4 drops 65V at 250mA. you'll be drawing < 50mA per channel
so figure 30V max. that's probably a good number for any of these
tubes.
for a cap input supply the raw dc (before rectifier drop)
will be 1.414 * Vin for each phase.
i.e., Vin = 500V for a 1000VCT (500-0-500) power tranny.
if you run the 45's at 250V and grid is at 150V and bias is -50V
then B+ is 150 - (-50) + 250 = 450V
you need to make allowance for rect drop (30V) and some filter
drop (say 20V or less if C-L-C is used) so the "raw" output
is 450 + 30 + 20 = 500V
the tranny voltage rating is then 500V/1.414 = 354V
so a 350-0-350 (700VCT) jobbie should do.
for a choke input the same rules apply except now Vraw = 0.9*vin
so 500/0.9 = 555V . here you need an 1100VCT power tranny.
> I am thinking of DC supplies for the fils / heaters or will a humbucker pot
> on the VT-52 solve the hum problem? If I need to go DC I am thinking of CLC
> filters using 47K mic caps and small very low R chokes. What type of choke
> can I use and where from? I guess the R needs to be less than 1 ohm.
i'd try the pot and ac. should be no problem.
> I need
> separate supplies for the VT-52's but if I built a stereo amp (would
> monoblocks be much better?) could I use a common filament supply for the
> voltage amps? I was thinking common up till after the first cap then
> separate chokes and second caps? Any point? I am avoiding active
> regulation.
sounds fine to me. separate LC's ought to give you good channel
isolation.
i prefer stereo amps. mono-blocks may have
their advantages, but i've never noticed much difference...
just have a beefy supply for the both...
> I guess a big question is: If I am using a voltage divider to get the VT-52
> fil float voltage and the VA B+ then how much will all the volts interact
> when the output draws more current? Isn't this a major drawback of the high
> impedance of a tube rectifier? I guess everything goes up an down together
> sort of proportionally. How much harm does this do? Would it be worth the
> effort or overkill to arrange a totally separate supply for the VA /
> driver? The VA has about 200 volts dropped from the output B+ and then a
> 100mic decoupler (polyprop or cerafine?? bypass it?)and it only draws one
> or two milliamps. Surely this is pretty isolated?
>
here you're going amiss...
forget about this divider - we don't need no steeenking devidur ;)
all you need is
a bigger than normal cathode resistor.
1st step: decide on plate voltage of the input/driver.
this is equal to your output's grid voltage. (say it's 150V)
2nd step: decide on output tube operating point. 250V/32mA/-50V
is one.
3rd: now you know the cathode voltage is 150V - (-50) = 200V
4th: the cathode resistance is then the cathode voltage divided
by the cathode (plate) current:
Rk = 200V / 32mA = 6250
the power dissipated in this resistor is 200V * 32mA = 6.4W
so best to use a 20W resistor...
> Seems to me in thinking about this that with such a simple amp (almost just
> two tubes / two coils) the power supply is the most important part and is
> critical to the level of performance possible to be achieved (assuming that
> both tubes have well considered operating points).
>
yes, this is still true, because you still need to balance the supply
voltage of the first stage just right to get the plate at your
target (say 150V). you have to bias the input stage just right
for the proper current draw, then select the proper dropping
resistor in the power supply leg. the semi-fixed ni-cad bias
removes one variable that you might have had (i.e., the cathode
resistor) to adjust this setting. furthermore, the long-term
variation in this bias voltage could have a substantial effect on
the balance. for example, if the bias shifts by 0.1V, your plate
voltage can shift by 7 V or more. then your output stage shifts
accordingly....
it's tricky. you hit the nail on the head.
> I have heard criticisms of direct coupled circuits due to DC levels being
> upset by transientsand taking time to recover (overshoot, ringing?). Does
> anyone have any facts on this or even listening experience. I won't have
> much power so I guess relatively I'll have plenty of transients.
>
well, there are arguments either way. lesser of two evils?!
coupling cap/no coupling cap
recovery time mismatches...?!
best advice i can give is try as many topologies as you can
and find out what you like best.
i know you are thinking single stage here, and later i.t.coupling.
but i wouldn't rule out a nice simple cascade 6sn7, etc ...
it might sound best
but i know how you feel: it's not as fun to design ;(
> Lastly, the VT-52 shouldn't draw grid current should it?? I guess if my
> speakers aren't efficient enough?? If it does, I'm stuffed! Can I get 3
> watts?
>
don't think these things were meant to do grid current.
but even if they were, this type of driver (single stage
high gain low current) isn't going to be able to provide it.
maybe this is a good thing because the grid can't be slammed
so hard that it does not have much to recover from....?
figure on 2W or less. about 1.5W at 250V.
> All in all I really should go choke input for its improved regulation
> shouldn't I?
being a class A design, the variation in current demand from the supply
should not be that great. so a cap input should be fine.
>How much should I over-rate the power trannie HT current
> capability?
really depends. by rights you should be able to run both channels
from a 100mA tranny. some say to over-rate the tranny by at least
2X. imho this does you no good unless you bleed the other 100mA
(or some portion of it). the bleeder make the power supply look
bipolar (source/sink current). the rectifier generally
blocks reverse current flow so the caps have to do all the work
on that half of the cycle. the bleeder evens the response to ac
load. but again, this is not as much an issue with a Class A design.
>I guess I'm really getting beyond the 500 volt Cerafines as
> filters aren't I or are these conservatively rated??
close. reql close.
i usually use at least an 80% derating on caps. but if you use
a 700VCT tranny and a tube rectifier, only your surge voltage
will get up to about 495V. as the load kicks in (warm-up)
you'll drop to about 450...
>
> All help appreciated. Sorry about so many issues and questions - thanks for
> the bandwidth - at least this is on-topic! Nothing like trying a real
> design to put you in fast learning mode.
>
> regards,
>
> Bart
good luck bart. hope this helps.
always a pleasure...
bob.d.
=========================================================================
From: BART <bartonw@ozemail.com.au>
Subject: Learner designer - VT-52 amp
Date: Sat, 30 Aug 1997 03:18:42 +1000
Source: Sound Digest Archive v01.n005
Hi Guys,
I have put my toes in the water to design (as opposed to build) my first
amp.
I am planning a VT52 amp, direct coupled at first and maybe later
transformer coupled.
I would like the simplest possible circuit with a single tube voltage
amplifier / driver.
The candidates I have found as the VA are 12AX7 / 7025, 6SL7 or 6F5 / 6SF5.
The first two seem obvious choices to give the required gain and be able to
swing the required volts. Any reason not to parallel the twin tubes in a
monoblock?
I am surprised I haven't heard more about the 6F5 before. Any comments from
anyone or advice on other alternatives? What about a paralleled 6SC7?
I am thinking of a 100K pot input with a Nicad bias and a choke load. I am
thinking of using a TFA204 output trannie as the load. Is this suitable? It
will only have between 1 to 3ma through it so I guess I could use any old
PP output trannie. Should I terminate the secondary with a load to use a
trannie as an audio choke? 8ohms?
If I want 250 volts on the plate would the choke drop only about 5volts
from the B+? Given this I plan to arrange to float the VT-52 filament at
195 volts
( I dont have any curves or figures for this tube - can anyone supply
please?
At this stage I am assuming a bias of about -55 volts for example.
Would I then need about 470 on the plate. I plan to use an FS030 output
trannie so would I need about 480 B+?
I am thinking of a 5R4 rectifier . No figures for this either (I only have
an RC-19) How many volts will this drop? I am thinking of a CLC filter with
100mic caps and a big choke. Should I really be thinking choke input? Am I
right in thinking about a 1,000V CT power trannie? Or more for choke input?
I am thinking of DC supplies for the fils / heaters or will a humbucker pot
on the VT-52 solve the hum problem? If I need to go DC I am thinking of CLC
filters using 47K mic caps and small very low R chokes. What type of choke
can I use and where from? I guess the R needs to be less than 1 ohm. I need
separate supplies for the VT-52's but if I built a stereo amp (would
monoblocks be much better?) could I use a common filament supply for the
voltage amps? I was thinking common up till after the first cap then
separate chokes and second caps? Any point? I am avoiding active
regulation.
I guess a big question is: If I am using a voltage divider to get the VT-52
fil float voltage and the VA B+ then how much will all the volts interact
when the output draws more current? Isn't this a major drawback of the high
impedance of a tube rectifier? I guess everything goes up an down together
sort of proportionally. How much harm does this do? Would it be worth the
effort or overkill to arrange a totally separate supply for the VA /
driver? The VA has about 200 volts dropped from the output B+ and then a
100mic decoupler (polyprop or cerafine?? bypass it?)and it only draws one
or two milliamps. Surely this is pretty isolated?
Seems to me in thinking about this that with such a simple amp (almost just
two tubes / two coils) the power supply is the most important part and is
critical to the level of performance possible to be achieved (assuming that
both tubes have well considered operating points).
I have heard criticisms of direct coupled circuits due to DC levels being
upset by transientsand taking time to recover (overshoot, ringing?). Does
anyone have any facts on this or even listening experience. I won't have
much power so I guess relatively I'll have plenty of transients.
Lastly, the VT-52 shouldn't draw grid current should it?? I guess if my
speakers aren't efficient enough?? If it does, I'm stuffed! Can I get 3
watts?
All in all I really should go choke input for its improved regulation
shouldn't I? How much should I over-rate the power trannie HT current
capability? I guess I'm really getting beyond the 500 volt Cerafines as
filters aren't I or are these conservatively rated??
All help appreciated. Sorry about so many issues and questions - thanks for
the bandwidth - at least this is on-topic! Nothing like trying a real
design to put you in fast learning mode.
regards,
Bart
"Resistance is futile".
Bart Shepherd
2 Rosemount Ave., Summer Hill, N.S.W 2130, Australia
612 98081475 bartonw@ozemail.com.au
=========================================================================
From: "Hugh R. Dean" <aspen@alphalink.com.au>
Subject: Re: Learner designer - VT-52 amp
Date: Sat, 30 Aug 1997 12:52:44
Source: Sound Digest Archive v01.n005
Hi Bart, you wrote:
>I am thinking of a 100K pot input with a Nicad bias and a choke load. I am
>thinking of using a TFA204 output trannie as the load. Is this suitable? It
>will only have between 1 to 3ma through it so I guess I could use any old
>PP output trannie. Should I terminate the secondary with a load to use a
>trannie as an audio choke? 8ohms?
Your questions were a terrific challenge, and I hope this response is helpful:
Do you mean a nicad on the grid circuit with a largish resistor (220K) from
negative terminal to grid? A nicad on the cathode circuit would need to
pass 30mA which would mean it would run flat fairly quickly.
Input triodes work better if the grid leak resistor is high. 100K is
regarded as too low for ultra hi-fi, but an ideal terminating resistance
for an input signal. Further, a 100K pot in this scenario would need to
cope with the DC current of the bias from the battery, which would need to
pass through the 220K grid resistor, compromising the bias. Thus, a good
input cap might be necessary. It would need to be teflon dielelectric at
about 0u47 and just a few volts. Ouch!!! That could be rather expensive.
But then, so is an input tranny.
>If I want 250 volts on the plate would the choke drop only about 5volts
>from the B+? Given this I plan to arrange to float the VT-52 filament at
>195 volts
DCR of low current chokes is very small - typically 100 ohms - and if you
are passing 20mA the voltage drop is chicken feed - say 2 volts.
>( I dont have any curves or figures for this tube - can anyone supply
>please?
>At this stage I am assuming a bias of about -55 volts for example.
>Would I then need about 470 on the plate. I plan to use an FS030 output
>trannie so would I need about 480 B+?
The smaller tubes typically carry much smaller bias than the biggies; I'd
say about 25 volts would be right.
>I am thinking of a 5R4 rectifier . No figures for this either (I only have
>an RC-19) How many volts will this drop? I am thinking of a CLC filter with
>100mic caps and a big choke. Should I really be thinking choke input? Am I
>right in thinking about a 1,000V CT power trannie? Or more for choke input?
How about a GZ34? Cheaper and very sweet sounding. Any joenetters got
curves on this one??
>I am thinking of DC supplies for the fils / heaters or will a humbucker pot
>on the VT-52 solve the hum problem?
I would definitely go AC with a humbucker on the VT52 filament. Why?
Apparently DC on these tubes sound bland and unexciting. It is related to
the effect on filament emission from varying DC bias as you progress from
the negative end to the positive end. This starves a portion of the
cathode of its quota of electrons. There was a discussion on this about a
fortnight ago, as I remember.
If I need to go DC I am thinking of CLC
>filters using 47K mic caps and small very low R chokes.
In the designs I have seen outrageous capacitance has been used to filter
the filament supplies; typically 100,000 uF. This is expensive and stuffs
the sound; the humbucker pot with AC is the way to go. Should hum arise
in the VA stage, then I see no reason why a humbucker could not be used
there, too.
>but if I built a stereo amp (would
>monoblocks be much better?)
I'd say the monoblock for premium quality because the Zout of this amp will
be quite high and the shorter the cable run the better the sonics.
However, the allegedly successful Ongaku is a stereo amp, and this
construction would definitely be a lot cheaper. A major problem is the
correct placement of the chokes and transformers for minimal interference,
and to this end a compromise using a separate power supply chassis and a
stereo amp chassis might be the way to go. That is sufficiently novel to
be well regarded from the marketing angle, and I think would be a lot
cheaper than monoblocks.
BTW, you can integrate the supplies somewhat. The filaments can be run
from the one supply with no penalty whatever - providing the umbilical
cord is well twisted and shielded. But the B+ should be separated for each
channel at a later stage; my advice would be GZ37, 25uF, 10H, 100uF, 2 x
3H with a bypassed 16uF at the amp chassis on each channel - all
polypropylene. And I differ slightly on the multiplier used by Bob for the
secondary voltage calculation; in my experience I use 1.2 times the AC
rating less the voltage drop on the GZ37 (about 50 volts at 70mA I think).
>I guess a big question is: If I am using a voltage divider to get the VT-52
>fil float voltage and the VA B+ then how much will all the volts interact
>when the output draws more current? Isn't this a major drawback of the high
>impedance of a tube rectifier?
The voltage divider of which you speak is defined another way as a series
dropper and a bleeder attached. If the interstage tranny has, say, an 80H
inductance on the primary then the current swing on the stage will be less
than 10%, so a large bleeder is unnecessary. A decent bypassed
polypropylene cap is mandatory, however, and should be of the order of 10uF
per mA of stage current, including the bleeder, which here should pass
about the same current as the VA. You will probably use less than 200
volts on this stage since almost all will appear on the VA plate, so if you
then ground the cathode of the output tube (this gives maximum power, an
important consideration, for fixed bias) then the secondary of the
interstage can attach the top to the grid of the VT52 and the bottom to the
fixed bias voltage, maybe about 30 volts.
There are arguments for and against fixed bias; generally cathode bias
produces about 20% less audio power but sounds softer and more rounded.
Fixed bias offers maximal efficiency but is a little sharper. I have heard
it said that fixed bias is more dynamic; for the proper transmission of
'lifeforce', I would recommend fixed bias.
In the interests of avoiding churlish flames, I will simply say that's my
opinion. Nuff said.
> I guess everything goes up an down together
> sort of proportionally. How much harm does this do? Would it be worth the
>effort or overkill to arrange a totally separate supply for the VA /
>driver? The VA has about 200 volts dropped from the output B+ and then a
>100mic decoupler (polyprop or cerafine?? bypass it?)and it only draws one
>or two milliamps. Surely this is pretty isolated?
Yup, in a properly designed amp the B+ for all stages and the output tube
bias should track reasonably well. Thus the bias supply should be well
filtered but not regulated, so that if the B+ should rise with mains
peturbations, so can the bias rise with it, keeping output tube plate
current much the same. To this end, time constants in both supplies should
match so that these voltage peturbations can stay in step. Matching these
time constants is best done empirically, using a variac.
>Seems to me in thinking about this that with such a simple amp (almost just
>two tubes / two coils) the power supply is the most important part and is
>critical to the level of performance possible to be achieved (assuming that
>both tubes have well considered operating points).
Yes. Makes good sense to me. It must supply energy a near constant
voltage, it must serve as a transparent AC path to ground, and it must
remove hash and other undesirables from the mains supply.
>I have heard criticisms of direct coupled circuits due to DC levels being
>upset by transientsand taking time to recover (overshoot, ringing?). Does
>anyone have any facts on this or even listening experience. I won't have
>much power so I guess relatively I'll have plenty of transients.
This is a slew rate argument, Bart. It is based around the grid
capacitance of tubes, which in the 6SN7 is of the order of 10pF - and in a
2A3 rises to about 100pF. If the tube driving a grid only pulls 1mA plate
current, at very high frequencies the argument goes that the driving tube's
output impedance will be sufficiently high to wilt under the grid charge
regime required by the following, direct coupled tube. The Zout of the
first tube forms a reactive voltage divider with the capacitive reactance
of the grid of the following tube. Thus the second tube sees a steadily
reducing portion of the first tube's output as frequency rises. In fact,
slew induced distortion is nastier than that, and generates rather more
than just reducing volume.
>Lastly, the VT-52 shouldn't draw grid current should it?? I guess if my
>speakers aren't efficient enough?? If it does, I'm stuffed! Can I get 3
>watts?
Need the curves to say definitively, but my understanding is that all these
smaller output tubes were designed to operate in A1.
>All in all I really should go choke input for its improved regulation
>shouldn't I? How much should I over-rate the power trannie HT current
>capability? I guess I'm really getting beyond the 500 volt Cerafines as
>filters aren't I or are these conservatively rated??
No, not necessarily. I'd go for CLC followed by two legs of LC for each
amplifier.
If I have made mistakes, I'd appreciate constructive correction. Bart's a
great guy and I would hate to put him wrong.
Cheers,
Hugh R. Dean
ASPEN Amplifiers
Melbourne, Australia
Resistance is futile, impedance rules...
=========================================================================
From: BART <bartonw@ozemail.com.au>
Subject: RE: Learner designer - VT-52 amp
Date: Sat, 30 Aug 1997 14:29:10 +1000
Source: Sound Digest Archive v01.n006
Hi Bob,
Thanks for your considered and detailed reply.
I have read it and will consider it further.
Learning in public like this is a humbling experience but I can handle it
and it provides the opportunity for others to chip in and further
accelerate my progress. Those bored with it can easily skip these posts.
Oops , I did get the bias around backwards in the calc.
Actually Shishido uses both a 12AX7 and a voltage divider in a DC 2A3 amp.
I see what you mean about the choke. Might forget it or use a new design.
The 6SL7 can be biased at -2V.
I guess "float" wasn't the term to use. I meant set at a voltage above
ground.
I guess the reason for a divider versus a simple cathode R hasn't sunk in
with me yet. Dividers seem to be often used in these circuits. What are the
pro's and cons? Don't they make the setting more fixed and less current
dependant? You dont like them - why?
I guess I allowed myself the luxury of any voltages I like. Dont see this
affecting the costs too much but maybe means less cap choice with 630V
caps. All the voltages seem OK after starting off with 211 Amps!
Thanks again for your help Bob. It's obvious I'm learning but like to try
to be a bit innovative at the same time. I know this is dangerous but it's
fun and speeds the learning.
I need to pursue my understanding of Miller effect on load a bit more. The
big problem is I have no data. Can you or anyone recommend a source of tube
data and curves that I can buy. My little Philips and RCA RC-19 reprint
don't have much of the information I need. i think the VT-52 is not
identical to a 45 and uses a lower load and can produce more power. Hence I
hope for more than 2 watts.
My head hurts,
Bart
"Resistance is futile".
Bart Shepherd
2 Rosemount Ave., Summer Hill, N.S.W 2130, Australia
612 98081475 bartonw@ozemail.com.au
- ----------
From: Bob Danielak[SMTP:rdaniela@ew.lmms.lmco.com]
Sent: Saturday, 30 August 1997 7:21
To: BART
Cc: 'JoeList'
Subject: Re: Learner designer - VT-52 amp
hi bart
se comments below:
BART wrote:
>
> Hi Guys,
>
> I have put my toes in the water to design (as opposed to build) my first
> amp.
>
> I am planning a VT52 amp, direct coupled at first and maybe later
> transformer coupled.
>
> I would like the simplest possible circuit with a single tube voltage
> amplifier / driver.
>
have you seen jc's "micro". uses parallel 6sl7 direct coupled to 2a3.
this is about as simple as you're gonna get. this could be applied
to the VT52. let me know if you need the schematic.
but see further comments below...
> The candidates I have found as the VA are 12AX7 / 7025, 6SL7 or 6F5 /
6SF5.
> The first two seem obvious choices to give the required gain and be able
to
> swing the required volts. Any reason not to parallel the twin tubes in a
> monoblock?
>
the main perceived problem with these tubes is that with
high gain comes high plate resistance. this and the relatively
low plate currents coupled with the relatively high input capacitance
of power triodes (near 100pF or more) can lead to poor hf response
related to
slew rate. this should be most noteable at higher drive levels.
i must say that i've experimented with this type of driver into
a 2a3 and didn't notice a lack of high end, per se.
just wasn't thrilled with the overall sound.
i later changed to a choke loaded direct coupled driver
using parallel 6sl7's and a 90H/25mA choke. this was pretty
good. didn't even notice that bass was rolled off below 36hZ...
that's another issue, which is the subject of one of your next
questions...
> I am surprised I haven't heard more about the 6F5 before. Any comments
from
> anyone or advice on other alternatives? What about a paralleled 6SC7?
>
probably because everyone lumps the 6F5 in with 12ax7.
these are generally not considered "hi-fi" for various reasons.
a lot of it is the distorting guitar-amp stigma. but for the reasons
cited above, it generally does not make a good driver stage.
it is ok for low level input stages, but then you have to worry
about noise if it is a very low level stage -or- input headroom
if it is a higher (line) level stage...
the 6sc7 is similar to the 6sl7. the 'sl7 looks a little more linear
to me, and it's rp is a wee bit higher, but i may be splitting
hairs.
> I am thinking of a 100K pot input with a Nicad bias and a choke load. I
am
> thinking of using a TFA204 output trannie as the load. Is this suitable?
It
> will only have between 1 to 3ma through it so I guess I could use any old
> PP output trannie. Should I terminate the secondary with a load to use a
> trannie as an audio choke? 8ohms?
>
is that ni-cad in the grid circuit or in the cathode?
in any event, most hi-mu tubes are normally biased with <1V
at plate voltages < 150V or so. (refer to 6sc7 curves for example)
i guess you could do Ebo=100V, Ibo=0.5mA, and Eco=-1.2.
but that's awfully low current (even doubled for parallel operation)
to drive a power triode. it's also still in the curved regions
of the plate curves, fwiw.
to get a more respectable operating point, you'd need 150V or more
to get about 1.25mA per section....
this is an issue when it comes to direct coupling, which i'll explain
in a few...
as far as choke loading is concerned...
in order to get good low frequency response you need high inductance
and/or low plate resistance. the low frequency cut-off will be
at f = rp /(2*pi*L)
so for paralleled 6sc7 (rp ~ 50K) the effective rp = 25K
if you want f = 20Hz, then L req'd is about 200H!
don't think the primary of that tranny is that high.
remember, i said before that parallel 6sl7 (22K) into 90H gave me
a cut-off of about 36Hz...
a BIG pp tranny might be close to 200H with no DC (don't have
a real good feel here, since primary inductance is generally not
specified for pp OPT's)...
i plan to try it soon (although i'm doing a pp amp, and will be
using the primary as a split choke)...
as for loading the secondary, my first instinct would be NOT to
load it (or maybe with a relatively high impedance >>8ohms).
if you load it with 8 ohms it will reflect that back to the primary
as a 5K (or whatever the normal impedance is) load on the driver
tube. this sorta defeats the purpose of choke loading. the choke
is supposed to present a very high ac impedance load to the driver
tube. this results in a flatter load-line (constant-current)
and greater linearity.
btw: i have toyed with the idea of <selectively> loading the secondary
of a transformer used as a choke load. for example,
i came up with an RIAA network that seems to work this way.
don't know how it would work in practice. probably wreak havoc
with the poor driver tube -load impedance dancing all over the place!
> If I want 250 volts on the plate would the choke drop only about 5volts
> from the B+? Given this I plan to arrange to float the VT-52 filament at
> 195 volts
> ( I dont have any curves or figures for this tube - can anyone supply
> please?
> At this stage I am assuming a bias of about -55 volts for example.
> Would I then need about 470 on the plate. I plan to use an FS030 output
> trannie so would I need about 480 B+?
>
something's a little mixed up here, but 1st things 1st:
the tranny shouldn't drop hardly anything with 3mA thru it
maybe 1 to 1.5V max...
ok... now
if you're direct coupling, you don't have to "float" the filament.
it will find it's own level depending on the grid voltage (equal
to the driver plate dc voltage) and the required output bias voltage.
you must subtract the negative bias voltage to the grid (driver plate)
voltage
(looks like you added...)
so if your driver plate voltage is 250V and your vt52 bias
winds up being -55V, your cathode voltage is 250 - (-55) =305V!
now you must add the desired vt52 plate voltage to come up with the B+
requirement. i'd guess you want to run them at 250V to 275V.
so B+ of 555 to 580V is required.
now you see why most direct coupled amps run the input stage at
90V or so.
although, 150V on the driver plate would result in a B+ of about
450V to 480V just like you mention.
and yes, -50 to -55V is a good number for 45 bias.
i have the '45 data sheet if you need them (.gif format).
doesn't someone have them out on the web (Dan Cheever?
try: http://web.mit.edu/cheever/www/45curve.jpg
i'll e-mail mine to you if you wish...
> I am thinking of a 5R4 rectifier . No figures for this either (I only
have
> an RC-19) How many volts will this drop? I am thinking of a CLC filter
with
> 100mic caps and a big choke. Should I really be thinking choke input? Am
I
> right in thinking about a 1,000V CT power trannie? Or more for choke
input?
>
5r4 is grossly overrated for this application, but that might
be a good thing! ;) a 5y3 or 5v4 will do just fine.
the 5r4 drops 65V at 250mA. you'll be drawing < 50mA per channel
so figure 30V max. that's probably a good number for any of these
tubes.
for a cap input supply the raw dc (before rectifier drop)
will be 1.414 * Vin for each phase.
i.e., Vin = 500V for a 1000VCT (500-0-500) power tranny.
if you run the 45's at 250V and grid is at 150V and bias is -50V
then B+ is 150 - (-50) + 250 = 450V
you need to make allowance for rect drop (30V) and some filter
drop (say 20V or less if C-L-C is used) so the "raw" output
is 450 + 30 + 20 = 500V
the tranny voltage rating is then 500V/1.414 = 354V
so a 350-0-350 (700VCT) jobbie should do.
for a choke input the same rules apply except now Vraw = 0.9*vin
so 500/0.9 = 555V . here you need an 1100VCT power tranny.
> I am thinking of DC supplies for the fils / heaters or will a humbucker
pot
> on the VT-52 solve the hum problem? If I need to go DC I am thinking of
CLC
> filters using 47K mic caps and small very low R chokes. What type of
choke
> can I use and where from? I guess the R needs to be less than 1 ohm.
i'd try the pot and ac. should be no problem.
> I need
> separate supplies for the VT-52's but if I built a stereo amp (would
> monoblocks be much better?) could I use a common filament supply for the
> voltage amps? I was thinking common up till after the first cap then
> separate chokes and second caps? Any point? I am avoiding active
> regulation.
sounds fine to me. separate LC's ought to give you good channel
isolation.
i prefer stereo amps. mono-blocks may have
their advantages, but i've never noticed much difference...
just have a beefy supply for the both...
> I guess a big question is: If I am using a voltage divider to get the
VT-52
> fil float voltage and the VA B+ then how much will all the volts interact
> when the output draws more current? Isn't this a major drawback of the
high
> impedance of a tube rectifier? I guess everything goes up an down
together
> sort of proportionally. How much harm does this do? Would it be worth
the
> effort or overkill to arrange a totally separate supply for the VA /
> driver? The VA has about 200 volts dropped from the output B+ and then a
> 100mic decoupler (polyprop or cerafine?? bypass it?)and it only draws one
> or two milliamps. Surely this is pretty isolated?
>
here you're going amiss...
forget about this divider - we don't need no steeenking devidur ;)
all you need is
a bigger than normal cathode resistor.
1st step: decide on plate voltage of the input/driver.
this is equal to your output's grid voltage. (say it's 150V)
2nd step: decide on output tube operating point. 250V/32mA/-50V
is one.
3rd: now you know the cathode voltage is 150V - (-50) = 200V
4th: the cathode resistance is then the cathode voltage divided
by the cathode (plate) current:
Rk = 200V / 32mA = 6250
the power dissipated in this resistor is 200V * 32mA = 6.4W
so best to use a 20W resistor...
> Seems to me in thinking about this that with such a simple amp (almost
just
> two tubes / two coils) the power supply is the most important part and is
> critical to the level of performance possible to be achieved (assuming
that
> both tubes have well considered operating points).
>
yes, this is still true, because you still need to balance the supply
voltage of the first stage just right to get the plate at your
target (say 150V). you have to bias the input stage just right
for the proper current draw, then select the proper dropping
resistor in the power supply leg. the semi-fixed ni-cad bias
removes one variable that you might have had (i.e., the cathode
resistor) to adjust this setting. furthermore, the long-term
variation in this bias voltage could have a substantial effect on
the balance. for example, if the bias shifts by 0.1V, your plate
voltage can shift by 7 V or more. then your output stage shifts
accordingly....
it's tricky. you hit the nail on the head.
> I have heard criticisms of direct coupled circuits due to DC levels being
> upset by transientsand taking time to recover (overshoot, ringing?). Does
> anyone have any facts on this or even listening experience. I won't have
> much power so I guess relatively I'll have plenty of transients.
>
well, there are arguments either way. lesser of two evils?!
coupling cap/no coupling cap
recovery time mismatches...?!
best advice i can give is try as many topologies as you can
and find out what you like best.
i know you are thinking single stage here, and later i.t.coupling.
but i wouldn't rule out a nice simple cascade 6sn7, etc ...
it might sound best
but i know how you feel: it's not as fun to design ;(
> Lastly, the VT-52 shouldn't draw grid current should it?? I guess if my
> speakers aren't efficient enough?? If it does, I'm stuffed! Can I get 3
> watts?
>
don't think these things were meant to do grid current.
but even if they were, this type of driver (single stage
high gain low current) isn't going to be able to provide it.
maybe this is a good thing because the grid can't be slammed
so hard that it does not have much to recover from....?
figure on 2W or less. about 1.5W at 250V.
> All in all I really should go choke input for its improved regulation
> shouldn't I?
being a class A design, the variation in current demand from the supply
should not be that great. so a cap input should be fine.
>How much should I over-rate the power trannie HT current
> capability?
really depends. by rights you should be able to run both channels
from a 100mA tranny. some say to over-rate the tranny by at least
2X. imho this does you no good unless you bleed the other 100mA
(or some portion of it). the bleeder make the power supply look
bipolar (source/sink current). the rectifier generally
blocks reverse current flow so the caps have to do all the work
on that half of the cycle. the bleeder evens the response to ac
load. but again, this is not as much an issue with a Class A design.
>I guess I'm really getting beyond the 500 volt Cerafines as
> filters aren't I or are these conservatively rated??
close. reql close.
i usually use at least an 80% derating on caps. but if you use
a 700VCT tranny and a tube rectifier, only your surge voltage
will get up to about 495V. as the load kicks in (warm-up)
you'll drop to about 450...
>
> All help appreciated. Sorry about so many issues and questions - thanks
for
> the bandwidth - at least this is on-topic! Nothing like trying a real
> design to put you in fast learning mode.
>
> regards,
>
> Bart
good luck bart. hope this helps.
always a pleasure...
bob.d.
=========================================================================
From: BART <bartonw@ozemail.com.au>
Subject: RE: Learner designer - VT-52 amp
Date: Sat, 30 Aug 1997 14:52:58 +1000
Source: Sound Digest Archive v01.n006
Hi Hugh,
You wrote:
>>
Hi Bart, you wrote:
>I am thinking of a 100K pot input with a Nicad bias and a choke load. I am
>thinking of using a TFA204 output trannie as the load. Is this suitable?
It
>will only have between 1 to 3ma through it so I guess I could use any old
>PP output trannie. Should I terminate the secondary with a load to use a
>trannie as an audio choke? 8ohms?
Your questions were a terrific challenge, and I hope this response is
helpful:
Do you mean a nicad on the grid circuit with a largish resistor (220K) from
negative terminal to grid? A nicad on the cathode circuit would need to
pass 30mA which would mean it would run flat fairly quickly.
>>
I meant in the cathode circuit of the voltage amp.
I thought there would only be about 2.3ma here and the battery would try to
charge not discharge.
>>
Input triodes work better if the grid leak resistor is high. 100K is
regarded as too low for ultra hi-fi, but an ideal terminating resistance
for an input signal. Further, a 100K pot in this scenario would need to
cope with the DC current of the bias from the battery, which would need to
pass through the 220K grid resistor, compromising the bias. Thus, a good
input cap might be necessary. It would need to be teflon dielelectric at
about 0u47 and just a few volts. Ouch!!! That could be rather expensive.
But then, so is an input tranny.
>>
Dont need a cap if battery in the cathode circuit, do I?
Many circuits seem to use a 100K pot here. Should I try a 250K?
>>
>If I want 250 volts on the plate would the choke drop only about 5volts
>from the B+? Given this I plan to arrange to float the VT-52 filament at
>195 volts
DCR of low current chokes is very small - typically 100 ohms - and if you
are passing 20mA the voltage drop is chicken feed - say 2 volts.
>( I dont have any curves or figures for this tube - can anyone supply
>please?
>At this stage I am assuming a bias of about -55 volts for example.
>Would I then need about 470 on the plate. I plan to use an FS030 output
>trannie so would I need about 480 B+?
The smaller tubes typically carry much smaller bias than the biggies; I'd
say about 25 volts would be right.
>I am thinking of a 5R4 rectifier . No figures for this either (I only have
>an RC-19) How many volts will this drop? I am thinking of a CLC filter
with
>100mic caps and a big choke. Should I really be thinking choke input? Am I
>right in thinking about a 1,000V CT power trannie? Or more for choke
input?
How about a GZ34? Cheaper and very sweet sounding. Any joenetters got
curves on this one??
>I am thinking of DC supplies for the fils / heaters or will a humbucker
pot
>on the VT-52 solve the hum problem?
I would definitely go AC with a humbucker on the VT52 filament. Why?
Apparently DC on these tubes sound bland and unexciting. It is related to
the effect on filament emission from varying DC bias as you progress from
the negative end to the positive end. This starves a portion of the
cathode of its quota of electrons. There was a discussion on this about a
fortnight ago, as I remember.
If I need to go DC I am thinking of CLC
>filters using 47K mic caps and small very low R chokes.
In the designs I have seen outrageous capacitance has been used to filter
the filament supplies; typically 100,000 uF. This is expensive and stuffs
the sound; the humbucker pot with AC is the way to go. Should hum arise
in the VA stage, then I see no reason why a humbucker could not be used
there, too.
>but if I built a stereo amp (would
>monoblocks be much better?)
I'd say the monoblock for premium quality because the Zout of this amp will
be quite high and the shorter the cable run the better the sonics.
However, the allegedly successful Ongaku is a stereo amp, and this
construction would definitely be a lot cheaper. A major problem is the
correct placement of the chokes and transformers for minimal interference,
and to this end a compromise using a separate power supply chassis and a
stereo amp chassis might be the way to go. That is sufficiently novel to
be well regarded from the marketing angle, and I think would be a lot
cheaper than monoblocks.
BTW, you can integrate the supplies somewhat. The filaments can be run
from the one supply with no penalty whatever - providing the umbilical
cord is well twisted and shielded. But the B+ should be separated for each
channel at a later stage; my advice would be GZ37, 25uF, 10H, 100uF, 2 x
3H with a bypassed 16uF at the amp chassis on each channel - all
polypropylene. And I differ slightly on the multiplier used by Bob for the
secondary voltage calculation; in my experience I use 1.2 times the AC
rating less the voltage drop on the GZ37 (about 50 volts at 70mA I think).
>>
I thought the VT-52 fils would need to be separate so as to retain separate
bias adjustment?
>>
>I guess a big question is: If I am using a voltage divider to get the
VT-52
>fil float voltage and the VA B+ then how much will all the volts interact
>when the output draws more current? Isn't this a major drawback of the
high
>impedance of a tube rectifier?
The voltage divider of which you speak is defined another way as a series
dropper and a bleeder attached. If the interstage tranny has, say, an 80H
inductance on the primary then the current swing on the stage will be less
than 10%, so a large bleeder is unnecessary. A decent bypassed
polypropylene cap is mandatory, however, and should be of the order of 10uF
per mA of stage current, including the bleeder, which here should pass
about the same current as the VA. You will probably use less than 200
volts on this stage since almost all will appear on the VA plate, so if you
then ground the cathode of the output tube (this gives maximum power, an
important consideration, for fixed bias) then the secondary of the
interstage can attach the top to the grid of the VT52 and the bottom to the
fixed bias voltage, maybe about 30 volts.
There are arguments for and against fixed bias; generally cathode bias
produces about 20% less audio power but sounds softer and more rounded.
Fixed bias offers maximal efficiency but is a little sharper. I have heard
it said that fixed bias is more dynamic; for the proper transmission of
'lifeforce', I would recommend fixed bias.
In the interests of avoiding churlish flames, I will simply say that's my
opinion. Nuff said.
>>
I have been describing a direct coupled amp not the trannie coupled one. Do
I miss your point here. I may be a bit obtuse so perhaps we can continue
this bit offline.
>>
> I guess everything goes up an down together
> sort of proportionally. How much harm does this do? Would it be worth the
>effort or overkill to arrange a totally separate supply for the VA /
>driver? The VA has about 200 volts dropped from the output B+ and then a
>100mic decoupler (polyprop or cerafine?? bypass it?)and it only draws one
>or two milliamps. Surely this is pretty isolated?
Yup, in a properly designed amp the B+ for all stages and the output tube
bias should track reasonably well. Thus the bias supply should be well
filtered but not regulated, so that if the B+ should rise with mains
peturbations, so can the bias rise with it, keeping output tube plate
current much the same. To this end, time constants in both supplies should
match so that these voltage peturbations can stay in step. Matching these
time constants is best done empirically, using a variac.
>Seems to me in thinking about this that with such a simple amp (almost
just
>two tubes / two coils) the power supply is the most important part and is
>critical to the level of performance possible to be achieved (assuming
that
>both tubes have well considered operating points).
Yes. Makes good sense to me. It must supply energy a near constant
voltage, it must serve as a transparent AC path to ground, and it must
remove hash and other undesirables from the mains supply.
>I have heard criticisms of direct coupled circuits due to DC levels being
>upset by transientsand taking time to recover (overshoot, ringing?). Does
>anyone have any facts on this or even listening experience. I won't have
>much power so I guess relatively I'll have plenty of transients.
This is a slew rate argument, Bart. It is based around the grid
capacitance of tubes, which in the 6SN7 is of the order of 10pF - and in a
2A3 rises to about 100pF. If the tube driving a grid only pulls 1mA plate
current, at very high frequencies the argument goes that the driving tube's
output impedance will be sufficiently high to wilt under the grid charge
regime required by the following, direct coupled tube. The Zout of the
first tube forms a reactive voltage divider with the capacitive reactance
of the grid of the following tube. Thus the second tube sees a steadily
reducing portion of the first tube's output as frequency rises. In fact,
slew induced distortion is nastier than that, and generates rather more
than just reducing volume.
>>
Well explained. I think I understand. But given an output impedance of less
than 22k for a parallel 6SL7 dont the sums for 100pf end up out of the
audio range?
>>
snip
Thanks Hugh,
Bart
p.s. Did you get my FAX of this??
=========================================================================
From: MHuber@t-online.de (Manfred Huber)
Subject: Re: Learner designer - VT-52 amp
Date: Sat, 30 Aug 97 11:20:50 -0100
Source: Sound Digest Archive v01.n006
Bart,
>I am planning a VT52 amp, direct coupled at first and maybe later
>transformer coupled.
Ahhh YOU are the guy buying all the VT-52s before I find them <g>
Very nice tube!
My first Amp ever is working for a few days now and it sounds really nice
to me. While using a VT-52 (Senco at present) it's a completely different
animal from what you plan to do. It has three stages with RC coupling.
First stage is a 1LE3 with DC filament supply and two NiCds in the
cathode circuit for bias. Second stage is a type 26 with ac filament supply
and fixed bias. Third stage is the VT-52 also with AC heating.
>I plan to use an FS030 output trannie
I'm using the Tango XE-20s at present. It is set up for Ra=5K into 8 Ohms now.
>I am thinking of a 5R4 rectifier . No figures for this either
I'm using the 5V4. It is indirectly heated and has much less resistance then
the 5R4. Drops only 20V in my setup. Notice that I need only 315V of supply voltage.
>I am thinking of a CLC filter with 100mic caps and a big choke.
>Should I really be thinking choke input?
Don't know. Never tried the PI filter. I'm using a choke input filter right now.
10H and 200uF/350V Cerafine common to both channels. Second filter stage is
10H/200uF Cerafine again. The second stage is built two times. One for each channel.
>I am thinking of DC supplies for the fils / heaters or will a humbucker pot
>on the VT-52 solve the hum problem?
It did for me. I use the following configuration on the cathode:
- ------------------------+
6.3Vac |
- --------+-- filament ---+
| |
+-- 1000R pot --+
| | |
22R 100R 22R
| | |
+-------+-------+
|
GND
Hum on the anode of the VT-52s is 140mV. With my 95dB efficient Tannoy speakers
this is very barely audible. Note that not all of the Hum is from the VT-52 heaters.
the 26 in the second stage is also ac heated and there is very little hum from the B+
also.
>I'm stuffed! Can I get 3 watts?
I get 2.5W into 8 Ohms before clipping. This is with an 5K output transformer.
The FS030 will give more power. I used a 5K output because the VT-52 has an ra of
1600 Ohms and I wanted to have some damping. The impedance of my Tannoys varies
considerably with frequency. The VT-52 is cooking at 305V/48mA with -55V bias.
I used an operating point with considerably less anode current first but when
going up to about 50mA the amp came to life. If your trannie has considerably more
primary inductance than mine (18H) you my get by with less current. I recommend you
choose the operating point for sound and not for power. 3W to 2.5W means a loss of
0.8dB. Do you think it matters? The difference in sound I noticed was astonishing!
Good luck and have fun! I hope my comments are useful.
Best Regards
Manfred
(excuse me butchering your language)
- ------------------
Manfred Huber
MHuber@t-online.de
- ------------------
=========================================================================
From: Jim de Kort <jdekort@xs4all.nl>
Subject: Re: Learner designer - VT-52 amp
Date: Sun, 31 Aug 1997 00:28:56 +0200
Source: Sound Digest Archive v01.n007
Hi,
>My first Amp ever is working for a few days now and it sounds really nice
>to me. While using a VT-52 (Senco at present)
Senco was made by Hytron, so chances are you are using Hytrons...
>Hum on the anode of the VT-52s is 140mV.
With 10H/200uF/10H/200uF ? That's a lot, are you sure you don't have some
kind of ground problem ?
>>I'm stuffed! Can I get 3 watts?
Max 4 Watts can be pulled from the VT52, but don't ask how and how long it
will last.
Jim de Kort
jdekort@xs4all.nl
- -----------------------------------------------
For DHT tube data and tubes
for sale, check out the
PhaetoN HomePage
http://www.xs4all.nl/~jdekort/
- -----------------------------------------------
=========================================================================
From: BART <bartonw@ozemail.com.au>
Subject: RE: Learner designer - VT-52 amp
Date: Sun, 31 Aug 1997 19:10:05 +1000
Source: Sound Digest Archive v01.n007
Hi Manfred,
Thank you for your advice and comments.
Your English is excellent.
For all I know your approach may be superior to the one I am contemplating.
Your design sounds very nice.
I will try all A/C filaments.
Which Tannoys do you have? How do you like them? What sort of cabinets are
they in? Do you have enough power for a punchy loud dynamic sound?
Are they coloured e.g. to warm in the upper bass or cuppy in the mid-range?
"Resistance is futile".
Bart Shepherd
2 Rosemount Ave., Summer Hill, N.S.W 2130, Australia
612 98081475 bartonw@ozemail.com.au
- ----------
From: Manfred Huber[SMTP:MHuber@t-online.de]
Sent: Saturday, 30 August 1997 22:20
To: JoeNet
Subject: Re: Learner designer - VT-52 amp
Bart,
>I am planning a VT52 amp, direct coupled at first and maybe later
>transformer coupled.
Ahhh YOU are the guy buying all the VT-52s before I find them <g>
Very nice tube!
My first Amp ever is working for a few days now and it sounds really nice
to me. While using a VT-52 (Senco at present) it's a completely different
animal from what you plan to do. It has three stages with RC coupling.
First stage is a 1LE3 with DC filament supply and two NiCds in the
cathode circuit for bias. Second stage is a type 26 with ac filament supply
and fixed bias. Third stage is the VT-52 also with AC heating.
>I plan to use an FS030 output trannie
I'm using the Tango XE-20s at present. It is set up for Ra=5K into 8 Ohms
now.
>I am thinking of a 5R4 rectifier . No figures for this either
I'm using the 5V4. It is indirectly heated and has much less resistance
then
the 5R4. Drops only 20V in my setup. Notice that I need only 315V of supply
voltage.
>I am thinking of a CLC filter with 100mic caps and a big choke.
>Should I really be thinking choke input?
Don't know. Never tried the PI filter. I'm using a choke input filter right
now.
10H and 200uF/350V Cerafine common to both channels. Second filter stage is
10H/200uF Cerafine again. The second stage is built two times. One for each
channel.
>I am thinking of DC supplies for the fils / heaters or will a humbucker
pot
>on the VT-52 solve the hum problem?
It did for me. I use the following configuration on the cathode:
- ------------------------+
6.3Vac |
- --------+-- filament ---+
| |
+-- 1000R pot --+
| | |
22R 100R 22R
| | |
+-------+-------+
|
GND
Hum on the anode of the VT-52s is 140mV. With my 95dB efficient Tannoy
speakers
this is very barely audible. Note that not all of the Hum is from the VT-52
heaters.
the 26 in the second stage is also ac heated and there is very little hum
from the B+
also.
>I'm stuffed! Can I get 3 watts?
I get 2.5W into 8 Ohms before clipping. This is with an 5K output
transformer.
The FS030 will give more power. I used a 5K output because the VT-52 has an
ra of
1600 Ohms and I wanted to have some damping. The impedance