Reply
 
Thread Tools Display Modes
  #1   Report Post  
Jon Yaeger
 
Posts: n/a
Default Using Damper Diodes for B+ Supplies


I'm contemplating using a pair of 6AU4GT damper diodes for a full-wave
bridge, with a 700 VAC CT tranny.

Yet the spec sheet for the tube has these limits:

Heater positive with respect to cathode:
DC: 100V
DC + Peak: 300V

Heater negative with respect to cathode:
DC: 900V
DC + Peak: 4500V


I assume that during an AC cycle the cathode will "see" -350 volts relative
to the heater which would normally be operating near ground.

Looks like I'd be in trouble without floating the filament at a minus
potential.

Is that a real concern in this application? And if so, is floating the
filament a -300V (using SS rectifiers) a practical work-around?

Else would I be better selecting another device or topology?

TIA,

Jon


  #2   Report Post  
Bret Ludwig
 
Posts: n/a
Default


Jon Yaeger wrote:

al work-around?

Else would I be better selecting another device or topology?



Unless you are designing for nuke war-in which simply using tubes is
not enough, you have to make use of some very specific shielding and
grounding practices or small signal tubes will start losing
filaments-you would be far better off using solid state recitifiers.

Get a good book on building DC supplies that is simple and direct,
such as that bu the late Joe Carr. First build a good DC filtered A+/-
supply with either controlled warm-up time, foldback current limiting
for the steady state requirements of the tubes used, or both. Then
sequence your HT supply from the primary (line AC) supply for inrush
limiting. Easy and simple.

Derive your C- supply (you know what that is, right?) from the heater
LT side so you have it as soon as you have heater voltage. If you have
a D+ supply (an arcane term, so I will supply it-screen voltage) that
is separate it needs to come up simultaneously with or AFTER the B
supply.

Think of it like starting a jet engine-or Tony Montana's soliloquy
about, "First, you get the power..."

  #3   Report Post  
Jon Yaeger
 
Posts: n/a
Default

in article , Bret Ludwig
at
wrote on 10/1/05 11:09 PM:


Jon Yaeger wrote:

al work-around?

Else would I be better selecting another device or topology?



Unless you are designing for nuke war-in which simply using tubes is
not enough, you have to make use of some very specific shielding and
grounding practices or small signal tubes will start losing
filaments-you would be far better off using solid state recitifiers.

Get a good book on building DC supplies that is simple and direct,
such as that bu the late Joe Carr. First build a good DC filtered A+/-
supply with either controlled warm-up time, foldback current limiting
for the steady state requirements of the tubes used, or both. Then
sequence your HT supply from the primary (line AC) supply for inrush
limiting. Easy and simple.

Derive your C- supply (you know what that is, right?) from the heater
LT side so you have it as soon as you have heater voltage. If you have
a D+ supply (an arcane term, so I will supply it-screen voltage) that
is separate it needs to come up simultaneously with or AFTER the B
supply.

Think of it like starting a jet engine-or Tony Montana's soliloquy
about, "First, you get the power..."



Thanks, Bret, that's all good general advice, but perhaps in my ignorance it
seems that your answer is somewhat peripheral to the question.

My interest in using these tubes is that they are reputed to produce the
smoothest rectification. Because the cathodes are indirectly heated, there
is an 11+ second warm-up time, which makes life easier on the cathodes
downstream. Solid state rectifiers have their advantages, but I want an
all-tube circuit here, EMP notwithstanding. With plenty of current
available but no active regulation.

You state that specific shielding is needed. Would that be different, say,
from a 5AR4, and if so, kindly elaborate . . . . .

This circuit will be using cathode bias, so C- supply considerations don't
apply. Also will have ultralinear outputs so the arcane D supply is also a
non-issue.

TIA,

Jon

  #4   Report Post  
Chris Hornbeck
 
Posts: n/a
Default

On Sat, 01 Oct 2005 22:56:52 -0400, Jon Yaeger
wrote:

I'm contemplating using a pair of 6AU4GT damper diodes for a full-wave
bridge, with a 700 VAC CT tranny.


I assume that during an AC cycle the cathode will "see" -350 volts relative
to the heater which would normally be operating near ground.


And peak of (350) x (sq-rt 2) = about 500 volts (during startup,
and lower after warmup).


Looks like I'd be in trouble without floating the filament at a minus
potential.

Is that a real concern in this application? And if so, is floating the
filament a -300V (using SS rectifiers) a practical work-around?

Else would I be better selecting another device or topology?


I'd say that it's a non-issue, but if it bothers you, using
the damper diode as slow-start device following silicon
rectifiers may be more attractive to you. Serves most of
the same purposes, but not all.

Good fortune,

Chris Hornbeck
  #5   Report Post  
Bret Ludwig
 
Posts: n/a
Default


Jon Yaeger wrote:
in article , Bret Ludwig
at
wrote on 10/1/05 11:09 PM:


Thanks, Bret, that's all good general advice, but perhaps in my ignorance it
seems that your answer is somewhat peripheral to the question.

My interest in using these tubes is that they are reputed to produce the
smoothest rectification. Because the cathodes are indirectly heated, there
is an 11+ second warm-up time, which makes life easier on the cathodes
downstream. Solid state rectifiers have their advantages, but I want an
all-tube circuit here, EMP notwithstanding. With plenty of current
available but no active regulation.

You state that specific shielding is needed. Would that be different, say,
from a 5AR4, and if so, kindly elaborate . . . . .


Well, if EMP is anticipated, you are looking at voltages generated
along a gradient so that if there is a loop formed by the layout a
large differential can be developed across, that's where it will. What
you are doing is the same basic thing done to reduce M-field
interference-only much more carefully.

I don't think smoothness of rectification is any better with a damper
diode than any other tube, and I don't think smoothness of
rectification is that critical IF you filter your supply properly. I
wouldn't even spend any extra money on HEXFREDs. I would put my power
supply on a subchassis or a whole different chassis if I were really
serious, and I would heavily bypass all my lytics and put a RF choke in
the supply path.

Your heater supply should be fully energized at least one whole minute
before ANY B+ whatsoever is applied for best life. You could use time
delay relays or separate manual switches, whichever floats your boat.

A tube rectifier of course will work fine, but if you want the maximum
theoretical life you will have to provide for its slow warm-up as well.
It can get complicated.

Of course there is the alternate view, which is to say "screw it" and
build for simplicity. You will go through more tubes but if everyone
does that way they will keep making them and the price will be lower.
With the CRT industry on the verge of stand-down in general this could
be an important point-or the reduction in materials supply may end
receiving tube production altogether. hard telling.



  #6   Report Post  
DanF
 
Posts: n/a
Default


Jon Yaeger wrote:
I'm contemplating using a pair of 6AU4GT damper diodes for a full-wave
bridge, with a 700 VAC CT tranny.

Yet the spec sheet for the tube has these limits:

Heater positive with respect to cathode:
DC: 100V
DC + Peak: 300V

Heater negative with respect to cathode:
DC: 900V
DC + Peak: 4500V


I assume that during an AC cycle the cathode will "see" -350 volts relative
to the heater which would normally be operating near ground.

Looks like I'd be in trouble without floating the filament at a minus
potential.


You are over analyzing this... In a simple full wave B+ supply (2
damper tubes with ct of transformer being ground) the cathodes are
ALWAYS positive with respect to the heater (so that means that the
heater is always negative with respect to the cathode). Heater is at
ground potential, cathode at B+ potential.
You are correct about all of the advantages that tube rectifiers have
over solid state. The only drawback to them is that most consume A LOT
of heater power; but if you have extra heater winding current to spare
(or extra space to mount another heater transformer) then by all means,
use them. They sound nice!
For more info, get a copy of "Vacuum Tube Valley" issue # 12. There is
an article about using damper tubes for rectifiers.
Hope this helps... Daniel

  #7   Report Post  
Bret Ludwig
 
Posts: n/a
Default


DanF wrote:
You are correct about all of the advantages that tube rectifiers have
over solid state. The only drawback to them is that most consume A LOT
of heater power; but if you have extra heater winding current to spare
(or extra space to mount another heater transformer) then by all means,
use them. They sound nice!


My years of serious listening experience have proven that tube
rectifiers' only "sonic advantage" is if you want a saggy supply for a
guitar amp. In hi-fi service there are no sonic advantages except in
the minds of the nostalgia-ridden. In fact if the supply is properly,
fully engineered there is no difference at all.

A fully engineered tube rectifier power supply is more work and rigor
than most people here know or will stand for. You would be far better
to buy a Kepco or Lambda rack or bench supply and use that. They are
still available despite the morons, some of them regulars here, who
butcher them to make ****ty phat toob homebrew projects. They are well
done pieces in almost all cases.

  #8   Report Post  
John Stewart
 
Posts: n/a
Default

Jon Yaeger wrote:

I'm contemplating using a pair of 6AU4GT damper diodes for a full-wave
bridge, with a 700 VAC CT tranny.

Yet the spec sheet for the tube has these limits:

Heater positive with respect to cathode:
DC: 100V
DC + Peak: 300V

Heater negative with respect to cathode:
DC: 900V
DC + Peak: 4500V

I assume that during an AC cycle the cathode will "see" -350 volts relative
to the heater which would normally be operating near ground.

Looks like I'd be in trouble without floating the filament at a minus
potential.

Is that a real concern in this application? And if so, is floating the
filament a -300V (using SS rectifiers) a practical work-around?

Else would I be better selecting another device or topology?

TIA,

Jon


I once used a pair of 6AX4's in a PS. They seemed to do just fine. The
transformer HT was 400-0-400 as I recall. Cheers, JLS

  #9   Report Post  
Ian Iveson
 
Posts: n/a
Default

Clever Bret wrote

My years of serious listening experience have proven...


To who?

...that tube rectifiers' only "sonic advantage"...


No-one has suggested a sonic advantage. You have made this up.

[snip irrelevant part]


A fully engineered tube rectifier power supply is more work and
rigor
than most people here know or will stand for.


Then perhaps we can look forward to some positive contributions from
you.

... despite the morons, some of them regulars here, who
butcher them to make ****ty phat toob homebrew projects...


Morons? Here? Hope your not talking about my mate Patrick.

Who else has made a ****ty phat toob homebrew project?

cheers, Ian


  #10   Report Post  
Adam Stouffer
 
Posts: n/a
Default

Jon Yaeger wrote:

Looks like I'd be in trouble without floating the filament at a minus
potential.

Is that a real concern in this application? And if so, is floating the
filament a -300V (using SS rectifiers) a practical work-around?


Why not float the filament completely?


Adam


  #11   Report Post  
Andy Evans
 
Posts: n/a
Default

Question - I use 6BY5GA in a hybrid bridge. I float the filament, which
also feeds the input valve. Is this, then, compromising the HK
performance of the inpit valve, and should I reference this whole 6.3v
supply to earth? And if so how? Thanks

  #12   Report Post  
Mark S
 
Posts: n/a
Default


"DanF" wrote in message
oups.com...

Jon Yaeger wrote:
I'm contemplating using a pair of 6AU4GT damper diodes for a full-wave
bridge, with a 700 VAC CT tranny.

Yet the spec sheet for the tube has these limits:

Heater positive with respect to cathode:
DC: 100V
DC + Peak: 300V

Heater negative with respect to cathode:
DC: 900V
DC + Peak: 4500V


I assume that during an AC cycle the cathode will "see" -350 volts
relative
to the heater which would normally be operating near ground.

Looks like I'd be in trouble without floating the filament at a minus
potential.


You are over analyzing this... In a simple full wave B+ supply (2
damper tubes with ct of transformer being ground) the cathodes are
ALWAYS positive with respect to the heater (so that means that the
heater is always negative with respect to the cathode). Heater is at
ground potential, cathode at B+ potential.
You are correct about all of the advantages that tube rectifiers have
over solid state. The only drawback to them is that most consume A LOT
of heater power; but if you have extra heater winding current to spare
(or extra space to mount another heater transformer) then by all means,
use them. They sound nice!
For more info, get a copy of "Vacuum Tube Valley" issue # 12. There is
an article about using damper tubes for rectifiers.
Hope this helps... Daniel


Jon,
Daniel is right. I'm planning on using the same on my next project
(someday!). I've characterized many different types for voltage drop (the
dredded "sag" refered to elsewhere in this thread) and the 6CJ3 is an
absolute beast, damn stiff for a tube. The 6AU4 is no weenie either but I
think the newer 9 or 12 pin compactrons like the CJ are newer lower
impedence designs. Good luck!
Mark


  #13   Report Post  
Jon Yaeger
 
Posts: n/a
Default

in article et, Mark S at
wrote on 10/5/05 7:23 PM:


"DanF" wrote in message
oups.com...

Jon Yaeger wrote:
I'm contemplating using a pair of 6AU4GT damper diodes for a full-wave
bridge, with a 700 VAC CT tranny.

Yet the spec sheet for the tube has these limits:

Heater positive with respect to cathode:
DC: 100V
DC + Peak: 300V

Heater negative with respect to cathode:
DC: 900V
DC + Peak: 4500V


I assume that during an AC cycle the cathode will "see" -350 volts
relative
to the heater which would normally be operating near ground.

Looks like I'd be in trouble without floating the filament at a minus
potential.


You are over analyzing this... In a simple full wave B+ supply (2
damper tubes with ct of transformer being ground) the cathodes are
ALWAYS positive with respect to the heater (so that means that the
heater is always negative with respect to the cathode). Heater is at
ground potential, cathode at B+ potential.
You are correct about all of the advantages that tube rectifiers have
over solid state. The only drawback to them is that most consume A LOT
of heater power; but if you have extra heater winding current to spare
(or extra space to mount another heater transformer) then by all means,
use them. They sound nice!
For more info, get a copy of "Vacuum Tube Valley" issue # 12. There is
an article about using damper tubes for rectifiers.
Hope this helps... Daniel


Jon,
Daniel is right. I'm planning on using the same on my next project
(someday!). I've characterized many different types for voltage drop (the
dredded "sag" refered to elsewhere in this thread) and the 6CJ3 is an
absolute beast, damn stiff for a tube. The 6AU4 is no weenie either but I
think the newer 9 or 12 pin compactrons like the CJ are newer lower
impedence designs. Good luck!
Mark


Mark,

I'm looking at the 6DW4 too. Very hefty tube, with 1.2A filament draw.

Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
FS Scully power supplies and parts MrMarksMusic Vacuum Tubes 0 November 16th 04 07:24 AM
FS Scully power supplies and parts MrMarksMusic Pro Audio 0 November 16th 04 07:23 AM
FS Scully power supplies and parts MrMarksMusic Marketplace 0 November 16th 04 07:19 AM
"The Audibility of Power Supplies" Jim Candela Vacuum Tubes 37 February 10th 04 01:41 PM
Choosing Amplifier Power Diodes gene ohara Tech 9 July 19th 03 04:23 PM


All times are GMT +1. The time now is 05:15 PM.

Powered by: vBulletin
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 AudioBanter.com.
The comments are property of their posters.
 

About Us

"It's about Audio and hi-fi"