| Search | Login/Register
   Home » Audio Discussions » How to get a LOT of SET power. (102 posts, 6 pages)
  Print Thread | 1st Post |  
Page 3 of 6 (102 items) Select Pages:  « 1 2 3 4 5 » ... Last »
   Target    Threads for related reading   Most recent post in related threads   Forum  Replies   Views   Started 
  »  New  More power from Melquiades? More powerful tube?..  A new Icon Audio's MB81 Mono Blocks...  Melquiades Amplifier  Forum     19  169793  09-01-2007
  »  New  NAT Audio Magna -160 Watts of Single-Ended Class A..  The KR Audio ways....  Audio Discussions  Forum     5  48562  11-12-2007
  »  New  Claude Perrier and his tube seremony..  Claude Perrier and his tube seremony...  Audio Discussions  Forum     0  9834  01-05-2008
  »  New  Getting more power from SET vs. properly distorting SS...  Sound Board...  Audio Discussions  Forum     4  31086  05-09-2011
04-16-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 41
Post ID: 16055
Reply to: 16051
GU-80/GU-81
fiogf49gjkf0d
Paul, there are those Russian 450W pentodes GU80/81.

http://www.tubes.ru/techinfo/AmateurRadio/gu-81m.html

I'm also wondering how it behaves as a triode. 450W  +/- compact and linear triode would be great!



Cheers,
Jarek
04-16-2011 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,130
Joined on 10-12-2006

Post #: 42
Post ID: 16056
Reply to: 16055
11V Direct Heater
fiogf49gjkf0d
N-Set, this tube does look interesting, in that it mostly fits your stated theoretical desiderata, save the true triode part.  It will also be interesting to see how you swing the drive voltage in only one stage, and how you will from there arrange an acceptable output impedance.

Like I keep saying, I have no idea how to tame self-effects with all the giant parts and high voltages/EMF in the same box.  The scatter from an amp like this would light a grounded flourescent bulb at 30 paces, and the rails would support the MTA.

As you might have guessed already, I've been seriously zapped before, so this certainly enters into my thinking on this subject.  This said, and more to the point here, I  have no experience that suggests to me at this point that this approach would in the end bear better sonic fruit than a more conventional design.

AC heater, anyone?

Best regards,
Paul S
04-16-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 43
Post ID: 16057
Reply to: 16056
Can it be it?
fiogf49gjkf0d
Paul, I know NOTHING about pentodes. I looked at the curves:
at 1500V/300mA/-85V loaded with 3.2k will give a whooping 190W
with +/-85Vmax swing, staying in A1. At 2.4k this drops to 155W.
G2 @500V, eating 50mA of current(!)
But the linearity...???? The knee seems soft but the data are very sparse.

I'm wondering if 6E5P, choke loaded, DC coupled can drive it?
The is theoretically no grid current and the capacitances at LF do not matter
(anyway it's a HF tube)?
Then it would be a straight, DC coupled 2 stage
amp, as Romy wanted. The supply will be a but complicated: stacked 200V and 1500V
Seems people are trying it (although with a limited power in this case):

http://www.diyaudio.com/forums/tubes-valves/183163-gu-81m-tube-amp-schematics-3.html


I'm trying to see if this little monster has any following in audio...

I think I have a DC supply to power the heaters, they seem similar to 304TL.

   






Cheers,
Jarek
04-16-2011 Post does not mapped to Knowledge Tree
haralanov


Bulgaria
Posts 130
Joined on 05-20-2008

Post #: 44
Post ID: 16058
Reply to: 16057
GU81M
fiogf49gjkf0d
N-set, a friend of mine built an amp with this monstrous USSR tube. It has 3 stages and the GU81M is driven by 6C33C.

ampa na ra4ev.jpg

Within context of driving mid or high sensitivity bass drivers, this is all you need. In comparison, his previous SET with 6C33C sounds like a toy. There is only one tube that can beat GU81M in terms of dynamics and effortlessness in the bass octaves and it is GM100.






"If you can't explain it simply, you don't understand it well enough." -A.E.
04-16-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 45
Post ID: 16060
Reply to: 16058
More info please
fiogf49gjkf0d
Haralanov, thank you for the info.
More please! Is his a dedicated bass amp?
What is the operating point and load?
Why 6c33c? How reliable GU81 is?






Cheers,
Jarek
04-16-2011 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,130
Joined on 10-12-2006

Post #: 46
Post ID: 16061
Reply to: 16058
Uuummm...
fiogf49gjkf0d
Petar, one can deduce that the PS takes up the rest of the universe that is not in this picture; but where is the output transformer?  Is it hidden by that monster tube, or is this sucker OTL?!?

Eat your heart out, George Kaye!

Best regards,
Paul S
04-16-2011 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 9,495
Joined on 05-28-2004

Post #: 47
Post ID: 16062
Reply to: 16058
GU/GM vs. 6C33C
fiogf49gjkf0d
 haralanov wrote:
N-set, a friend of mine built an amp with this monstrous USSR tube. It has 3 stages and the GU81M is driven by 6C33C. Within context of driving mid or high sensitivity bass drivers, this is all you need. In comparison, his previous SET with 6C33C sounds like a toy. There is only one tube that can beat GU81M in terms of dynamics and effortlessness in the bass octaves and it is GM100.

 In comparison to GU/GM tubes, SETs with 6C33C sounds like a toy? Haralanov, I would challenge this observation. I would challenge it not becomes I am familiar with GU/GM to well and not because I am 6C33C user as suffer from “insulted ownership”. I would challenge this observation because there is nothing in larger tubes that would be superior to 6C33C in terms of “dynamics and effortlessness”.

What I would like to offer to your friend or to you is to expend on your findings. I do not refuse what you guys say but I challenge. I did see a few times what people report that bigger tubes have  “more dynamics and effortlessness” but it really it is all about the fact that more powerful tubes sit deeper in A1 class or that power powers tube has better drivers that drive them. I do not believe that GU/GM tube is more dynamic then 6C33C if both tubes are driven respectfully identical and drive the appropriate for own power sensitivity. In fact, I think that in respectful to own power operation 6C33C will be MORE dynamic then GU/GM (lower plate R, less wire in OPT...).

Sure, we are not sitting in the same room and exchange opinion about the same Sound and it is very difficult to find any common denominators… Still I feel that what your friend reports might be incorrect.

Rgs, Romy the Cat


"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
04-16-2011 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,130
Joined on 10-12-2006

Post #: 48
Post ID: 16064
Reply to: 16057
Bee's Knees
fiogf49gjkf0d
N-set, I give lip service and a sort of generic consideration to plate curves, but in the end I am not wholly secure in my understanding of this information as it relates to sound with a given tube in a given circuit.  My usual basic tactic is to ask rather less of the tube than it might give, in terms of plate dissipation, keeping one eye on the curves.  But I agree, I can't say that these curves look like something you'd generally choose for an output tube for an audio band amp.

But you've waited this long already; so I suppose that you are not going to spring into action as a result of some internet chatter on an "underground" DIY site.  As we frequently observe around here, plenty gets done as DIY just to do it.  OTOH, to survive building an amp like this should bring bragging rights, in itself!

Best regards,
Paul S
04-17-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 49
Post ID: 16066
Reply to: 16064
GU81 considerations
fiogf49gjkf0d
Paul, don't get me wrong--normally I'm not choosing a tube by the power, but this situation is somehow special.
There are very few tubes which have the right electrical parameters (gm, mu, power, etc) to start with. In the
small tubes word one can glaze hours on the curves, trace them himself and then choose to whatever
factors. Here we have what we have. Few +/- aplicable tubes and the curves traced for class B-C opertion.
It's all a gamble. BTW, I've located a Dutch guy at DIYAudio who has been building around GU81M, and monsters like GU5b.

GU81M curves do not look that disastrous: observe that the knee (esp. 0V) seems softer.
I'm wondering how to drive it in A1. Will it need power?
I'd like a very linear triode with mu=30-33 at least and ideally 20W plate diss. SV572-30? 812A? 2x6E5P??
I'm wondering why Romy in his wishlist prohibited paralled
drivers but would happily accept paralled output tubes???

Below is something which may be breadboardable with a single 6E5P
(provided my power tranny can stand the current and I teleport 3000km).
I have no idea how to power the screen properly (500V wrt the cathode, 50mA!).
A separate supply with VR string to give 500V? Now is the RC circuit (no idea how to get C from the data).

GU81M_2.gif

With the 833 A2 idea I'm a bit stuck: i've spiced 6E5P -> EL34 (triode) -> 10:1 -> 833 and
EL34 will not give enough swing (the main limiting factor is 425V screen max, plate can stand 800V)





Cheers,
Jarek
04-17-2011 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,130
Joined on 10-12-2006

Post #: 50
Post ID: 16074
Reply to: 16066
Triode 3-Stage
fiogf49gjkf0d
N-set, starting from reasonable input, the 6E5P can swing some serious voltage, and Romy also sings its praises in terms of sound.  Maybe sim that, running it in its safe zone to a 6C33C. From there, you have all the power you'd want to drive -- whatever.  Basically, I'm supposing that linear operation, in practice, is likely to work out better than a pushed 2-stage, given the working constraints of SET power in the first place...

I have no personal experience with the 6E5P, and so I don't know how well it holds its values over time; but I can say that the 6C33C will pass on what is fed to it, as long as you can keep it tuned;  basically, "fixed bias" needs to be monitored and adjustable.

How does the Dutch guy consider/deal with self/parasitic noise?  Maybe figure on all-shielded hook-up wire bled via a dedicated ground...   Also, I recently saw (then promptly forgot about...) a "new" composite shielding material that "offers the best of steel, aluminum and Mu metal", yet it can be bent, etc., even grounded.

Did you find those caps yet?

Please correct me if I'm wrong, but it appears to me that you are in love with "transformers".  Maybe parallel tubes are a way to boost power and drop output impedance at the same time, versus bigger tranny, with its own ramifications, etc.  No Free Lunch...

Best regards,
Paul S
04-17-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 51
Post ID: 16079
Reply to: 16074
2 stage
fiogf49gjkf0d
Paul, I'm not rushing to the breadboard. I'm trying to visualize if Romy's constraints
can be satisfied. I have to gather more data on GU81 to think of doing anything.
How linear it is?? Perhaps this tube is an audio no-go as you've mentioned?

Going to 3 stage would defy the whole GU81 exercise. If alowing the 3rd stage I'd go with 304tl A1
in the first place or 833A A2. Both seem to have a +/- good reputation in audio.
And reputation is what unfortunately I have at my disposal at this stage. Same applies to 6E5P.
I agree that 6E5P seems to give just right enough gain to approach 0Vbias
 (2Vrms x 32=90Vmax), without margin,
but anyway I'd parallel
the drivers than add one more stage. I've somehow bought the Romy's sense of elegance
with 2 stages.

PS caps are seemingly not a big problem: find the Nichicon part number, google it,
put your spam-friendly email in the pages which locate parts for you, in the next 2 days you'll
receive 40 offers from Asia. Some seemingly serious, but I haven't tried so far.

I'm not in love with transformers, but not affraid of them either (so far...)
Again driven by the sense of elegance rather,
I'm attracted by a simple design with few stages and one big tube. I of course can be
very wrong, as I said at ULF small dissincronizations of paralled tubes should not
matter (no time-domain smearing). Transformer will have to handle the power anyway,
so it will be big, but the compromises with HF are absent, as have already been mentioned 1000 times.







Cheers,
Jarek
04-17-2011 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 9,495
Joined on 05-28-2004

Post #: 52
Post ID: 16080
Reply to: 16079
Only one transformer per amp.
fiogf49gjkf0d
It is unfortunate that this thread will not lead to anything practical or tangible to me personally, so any talk from me are very much worthless. Still from the purely conceptual perspective I would like to explain why I would prefer non-transformer coupling.  I am thinking about low knee decay and I would like it to be as low order as possible. The OPT will have own roll off slope, whatever order it will be there is no escape from it. A second coupling transformer will have another roll off. So, pretend that each of them have second order at bottom, it means the entire amps will have 4th order roll off. Consider that I am crazy but I would like to have as slow roll off at the bottom as possible… So I feel that it has to be only one transformer in amplifier. Do not give me Sacuma as example – his amps sound like shit. BTW, the notion of one transformer per amp was inspired in me by Dima….

The caT.


"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
04-17-2011 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,130
Joined on 10-12-2006

Post #: 53
Post ID: 16081
Reply to: 16079
Gap
fiogf49gjkf0d
What is the "reverse engineering" version of an SET circuit that would accomplish - in terms of audible results - what you want?  The reason I suggest 3 stages is only because I see a gap between required input and necessary output for the proposed driver versus the desired output tube, versus the plate curves.  YMMV, of course.

Surely, the answer lies somewhere between Ockham and Einstein: something like, "make it as simple as necessary."

Re: buying in to the sense of 2-stage elegance, don't forget the difference in gain between your own stated wants and Romy's DSET, starting from the same point.

Best regards,
Paul S
04-17-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 54
Post ID: 16082
Reply to: 16080
You've promissed...
fiogf49gjkf0d
...to assist spiritually!

Interesting view on IT. Indeed the multichannel SET with line-level crossover changes the perspective
on the coupling techniques. This rules A2 out (unless cathode follower...blah blah blah)

And paralled drivers? May I ask why you wanted to avoid them but would accept paralled output tubes?



Cheers,
Jarek
04-17-2011 Post does not mapped to Knowledge Tree
gordan
Posts 12
Joined on 01-29-2009

Post #: 55
Post ID: 16083
Reply to: 16082
There are more human ways to electrocute your pets....
fiogf49gjkf0d
Guys, you seem to be perfectly comfortable with 2kV of plate voltage in your rooms?

04-17-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 56
Post ID: 16084
Reply to: 16081
How big a gap?
fiogf49gjkf0d
 Paul S wrote:
What is the "reverse engineering" version of an SET circuit that would accomplish - in terms of audible results - what you want?  The reason I suggest 3 stages is only because I see a gap between required input and necessary output for the proposed driver versus the desired output tube, versus the plate curves.  YMMV, of course.

Surely, the answer lies somewhere between Ockham and Einstein: something like, "make it as simple as necessary."

Re: buying in to the sense of 2-stage elegance, don't forget the difference in gain between your own stated wants and Romy's DSET, starting from the same point.

Best regards,
Paul S


Paul, I assume the standard 2Vrms input.  If in A1 GU81 is still ok with 40mA, 8W drive, then 6E5P gives
just enough gain, unless I'm overlooking something. If doubled, 16W should be ample enough.
A SET would require a linear 600W triode to give 150W A1, easy to drive, high gm, medium mu.



Cheers,
Jarek
04-17-2011 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 9,495
Joined on 05-28-2004

Post #: 57
Post ID: 16085
Reply to: 16083
2kV-3kV of B+ vs. 1A of plate current
fiogf49gjkf0d
 N-set wrote:
And paralled drivers? May I ask why you wanted to avoid them but would accept paralled output tubes?

I would for sure avoid any paralleling, would it be driver or output. Still, driver provides voltage gain and output stage - current gain. The difference between the paralleling tube will be more affecting voltage gain stage then current, at least it is what I feel but I do not insist the I am right.
 gordan wrote:
Guys, you seem to be perfectly comfortable with 2kV of plate voltage in your rooms?

We are perfectly comfortable to talk about it. No one put any 2kV in anybody room. We are talking about what conceptually might be done.  I would personally be much more comfortable with parallel 6C33C, which might be even sonically preferable as they would have ultra low Rp and very high current. (Higher current = higher rise time as less sensitive to any capacitance). The problem is that now one would say what the factual result is as no one, of very-very few would experiment with all of those options.

Another subject you are losing, gordan. I do not know him personally but if you see what N-set then you recognize the he is very experienced builder. The people who do not afraid 2kV-3kV power supplies are ether ignorant or stupid. N-set neither ignorant nor stupid and he knows how to deal with what he is dealing with. I, with my experience would not touch 2kV-3kV, in fact the people who advise me explicitly prohibited me to do it. However, for well experienced person it is not a big deal. Do not forget that somebody work on 750.000.000 power lines – if to know how to do it then voltage is not a problem – ignorance is. I am very sure that a well-built 2kV power amp made to the appropriate standards will be in home as dangers as a kitchen toaster. After all in older color TV we has 36.000V, so how many people were damaged?

BTW, if I sometimes to assemble the Dima’s Zarathustra II hybrid then …it has 5A idle current in output stage. The PS for it also VERY obnoxious. It is class A so it is huge input chokes with 4 by 270.000uF per channel.  The power transformers if I remember correctly are rated for 70V/25A. I can’t not even move the box with parts of Zarathustra II power supply…

The Cat


"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
04-17-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 58
Post ID: 16087
Reply to: 16085
HAHAHAHA!
fiogf49gjkf0d
Please, guys, I'm not an experienced builder at all!!Please don't offed a real experienced builders!
 I'm rather both hopelessly stupid
and ignorant. Knowing that, I'm trying to be carefull. That's all.
My active participation here, apart from the spiritual motivation to hear an unheard trace
of the voice of God, is simply because I have an unfnished, idiotic,  lifetime project over which I've
spent countless hours, down to designing the PS transformers myself and getting
German EI laminations for them. Then the experience came from my winders, they've showed me
few tricks. Now that project may get a new, less idiotic, life now. At least on paper...

Paralleling: good points! Although this is the transformer which changes V's to A's, but true
a tube like 6c33c is somewhat closer to that function than many other tubes. Parallel drivers would
have to be matched, which with 6E5P will not be costly.

2-3kV: I look at those Wavac's ect to see how they protect the tubes. The topcaps
are fully covered, probably in teflon. If you touch whatever part, including the tube, you are
potentially safe from HV. I'd add a faraday cage around the tube. Not only for safety but also to shield
the strong fields from the powertube. Inside: HV potted transformers and chokes, RF HV cable (I have 80m of
a prime RG something, good to few kV). This also answers Paul: use RG coax for the connections.

And I'd definitely try Slagle's cancellation technique in the OPT too: an additional winding
carrying 11A of the heater current to cancell  some of the 350mA DC magnetization.
This should make the OPT even more tangible: low effective DC magnetization, no MF nor HF,
just power and inductance. The magic of DSET.



Cheers,
Jarek
04-17-2011 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 9,495
Joined on 05-28-2004

Post #: 59
Post ID: 16088
Reply to: 16087
There is nothing to laugh.
fiogf49gjkf0d
 N-set wrote:
Please, guys, I'm not an experienced builder at all!!Please don't offed a real experienced builders!
 I'm rather both hopelessly stupid
and ignorant. Knowing that, I'm trying to be carefull. That's all.
My active participation here, apart from the spiritual motivation to hear an unheard trace
of the voice of God, is simply because I have an unfnished, idiotic,  lifetime project over which I've spent countless hours, down to designing the PS transformers myself and getting
German EI laminations for them. Then the experience came from my winders, they've showed me few tricks. Now that project may get a new, less idiotic, life now. At least on paper...
You be careful with your self-assessments. I did show your proposals to people I trust and they observing your circuits inform me that you are not an idiot and that you know what you are doing. I would still be looking voice of God in different place, primary from within, but it is a free country and everyone has own rights to pursue own happiness.

The Cat


"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
04-17-2011 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 450
Joined on 01-07-2006

Post #: 60
Post ID: 16089
Reply to: 16088
The highest path...
fiogf49gjkf0d
...is to come both from within and without. And to meet in the one focal point!



Cheers,
Jarek
Page 3 of 6 (102 items) Select Pages:  « 1 2 3 4 5 » ... Last »
   Target    Threads for related reading   Most recent post in related threads   Forum  Replies   Views   Started 
  »  New  More power from Melquiades? More powerful tube?..  A new Icon Audio's MB81 Mono Blocks...  Melquiades Amplifier  Forum     19  169793  09-01-2007
  »  New  NAT Audio Magna -160 Watts of Single-Ended Class A..  The KR Audio ways....  Audio Discussions  Forum     5  48562  11-12-2007
  »  New  Claude Perrier and his tube seremony..  Claude Perrier and his tube seremony...  Audio Discussions  Forum     0  9834  01-05-2008
  »  New  Getting more power from SET vs. properly distorting SS...  Sound Board...  Audio Discussions  Forum     4  31086  05-09-2011
Home Page  |  Last 24Hours  | Search  |  SiteMap  | Questions or Problems | Copyright Note
The content of all messages within the Forums Copyright © by authors of the posts