Rerurn to Romy the Cat's Site

Audio Discussions
Topic: Have the cake and eat it

Page 1 of 5 (107 items) 1 2 3 4 5 »


Posted by Romy the Cat on 02-26-2006

I was listening toady Rossini’s  “Sonata per Archi” and suddenly a strange feeling stroke me: I become suddenly queries about “more power”. It was not that something directly encouraged me in the Sound that I am getting in here. I never heard the Super Melquiades LF channel clipping in my room and most likely my windows would fly away before it happen but still…. I was wondering what is beyond of juts “being more then enough” power.

Something “interesting” lives in the super high gain and in the super excessive power.  I wonder if speakers (I am talking about juts LF section) are made properly, sound properly and could be driven perfectly fine by I would say 20W then HOW they would sound is the very same gain amplifier would drive them but thie time the amp would be able to output I would say 400W-500W?

Let live aside the high power SS amps. It is not difficult to make them 1000W but they will do different bass. I am talking about as crazy as 500W of A1-A2 operating triodes, preferably in A1? There are a number of the high power amps out there: 572,  813, 833, the Svetlana triodes with 350W of plate dissipation, some transmitting high voltage EIMAC tubes but they all still 100W-150W of A1. How can we get more, preferably a lot of more?

It looks that with contemporary cooling methods it would not be a big deal to submerge such a tube into a circulating-cooling oil and let the amplifier besides the driving the speakers also to deep-fry chicken…. I’m taking about some kind of two states, SET 500W monster,  with 500 pounds PS….

It would be interesting is to hear this amp, purely in metaphysical sense, and to hear what it would be able to do, and particualy  if it were able to go down a few Hz at it's full power….

I wonder if someone has it. I would fly ower to hear it…..

Rgs,
Romy the Cat


Posted by Romy the Cat on 02-26-2006

I have received a few emails today informing me that this idea of a super high power low frequency SET would not be possible and it would be “pushing the things WAY too far”.

Well, perhaps…. I am not planning to build it but I would like to encourage someone to make it and if the person will not kill himself then I would have chance to listen it. I really would love to hear it….

Initially I thought to "drive hard" the Svetlana RB300-3CX or 3CX300A1 but it would not be enough power in there. Then I was thinking about something like GM100 but this tube is too "flimsy" for this project.

http://www.goodsoundclub.com/PDF/GM100.pdf

My primary attention as now is lying on the Russian made GM2. They are not expansive, available in two versions: air cooling and water cooling and initially meant to be use as voltage regulator. It looks to me that it is very good tube for the project: 6.3V on filament at 30A, plate dissipation up to 2.800W, gain 55…

Any crazy people out there?
The Cat

Posted by Romy the Cat on 02-27-2006

It is a very strange feeling, but I really feel how this entire idea should sound. I mean the 500W SET of the REAL SET driving a 12-14 cubic feet of sealed enclosure loaded with North Leviathan Driver.

In past when I played with SETs driving Aura 18” motors it gave a completely surreal feelings. We are not accustomed to SET power, we not accustomed to the underhanded bass excursion with a pure paper suspension… It was very different feeling and I remember that some things that I experienced during those days were very “interesting”, to say the least. If you remember the Doctor Gizmo’s idea of 1000W OTL via witch he indented to “hear the voice of God”? It is not exactly my intention but I remember at that time I experienced some very “interesting thighs” that would be a fun to replicate in a REAL full scale within an environment with a real-life reverberation time.

I never was able in my past to run the full “scale bass scenario”. When in past I intended to hear my “real bass prototype” I loaded the old version (not the new crappie one) of Aura 18” into an insultingly small 9 cubic feet enclosure and locking myself in my small bathroom I was driving it with Lamm, lowering it’s plate current ML2 (there was no Melquiades at that time).  The sound  of SET driving this thing was something very special, something that is completely not heard under the normal circumstances.

Really, it would not be so outrageously difficult to make an operational prototype around the GM2 tube. GM2 with 2500V on plate could make a beautiful 400W-500W SET. It can work perfectly submerged. If you remember Von Gaylord from California made his PP 200W amplifiers around a liquid cooling.

Unfortunately in his case the amplifiers did not perform well and as I could judge in the Gaylord’s case the liquid-cooling was make explicitly for photogenic sake and served only marketing purpose.

The 2500V on plate sound like more or less manageable as well. The guys out there wind 4000kV-5000kV transformers and for the project, considering that it should handle juts <50Hz it will not have many complicated isolated sections. Sure, this amp with crazy LF objectives will demand enormous inductance and the OPT might be a size of a refrigerator…. The nobler would be if someone would build this thing…

I wonder… how far would I fly to listen this amp…
Romy the Cat


Posted by Romy the Cat on 09-28-2008

A visitor of my site sent an email about Australian Patrick Turner who is geting more SET power by paralleling the 845 tubes. It is not uncontroversial amps but itis 55W of SET…

http://www.turneraudio.com.au/monobloc845se55.html

The Cat

Posted by el`Ol on 05-23-2009
fiogf49gjkf0d
I just visited the Highend in Munich and had the chance to listen to two of those monster SETs.
Cessaro showed up with monstrous 833 amps and my thought was that there is something wrong with the music.
I know it is common in this forum to describe what one hears in a more refined way, but unfortunately I can´t in this case.
The KR monsters mainly lived from the resolution of the ESLs and the stunning dynamics of the Analysis magnetostats they drove.
Tonality was different than that of the best solid state amps, but not better.
The only ones who showed that tubes are still ahead of solid state in tonality to my ears were Leben, in combination with Blumenhofer speakers.
And this was an EL34 push-pull amp!


Posted by Romy the Cat on 05-24-2009
fiogf49gjkf0d

 el`Ol wrote:
I just visited the Highend in Munich and had the chance to listen to two of those monster SETs.

Sure, whatever you heard at the show is not the definition of sound of whole topology but the entire question is very interesting: is anything fundamentally wrong or fundamentally advantageous in high power SETs. The question is very interesting and I did not hear any credible chatter about it.

Surely if we have a tube with 10W plate dissipation and 200W plate dissipation then the difference would be in size of plate and in voltage and current that would be applied to plate.  Certainly it will not be it. It will be zillion of other moments of the tube construction not to mention the very different amps the use those very different tubes. I would not even mention the fact that 1W SET and 60W SET would drive very different speakers. It is like you bring to me now a GM70 amp with 45W then what would I do with it? I mean a higher power imply a completely different topology of loudspeakers. Even if we use high poser SET juts for bass then we immediately dive into DSET topology with very different rules of the game.

So, I think it is very difficult, if aver possible to equate the generals cones or pros for high power and low power SETs. I have see people who build/sell high power advocated some point and some claims that high power has some benefits. I have seen some low power people did the same. Why I said that there was no chatter then? Because I did not like arguments of neither camp and I did not find their points of view credible.

I still have no view about high power vs. low power SETs. I know that I have bias against high power SETs in case they use high voltages but it is about it… Unfortunatelyall amps that have a lot of dissipation on anode use high voltages…

The Cat

Posted by el`Ol on 05-25-2009
fiogf49gjkf0d
Hello Romy,

for large SETs price seems to go up exponentially with the power handling (I am sure you know better than me why), so these amps are under strong pressure of having to justify their existence, other than small SETs or pentode push-pull amps. There are few companies that explore this territory and few tube types that are available in numbers, and of course not all of them can be heard by someone who just visits a single show once a year, but when I listen to a pentode push-pull-amp that is so close to the small SETs I have heard the existence of the big SETs becomes quesionable for me.
I have also listened to the Wavac 833 amp combined with Kharma speakers some years ago and found it adds some shining or radiance to the sound I haven´t heard from other amps. For some People who listen to that slow and relaxing audiophile music and rather bath in sound than follow a musical performance this may be a very musical feature. Well, maybe this justifies the existence of something that horribly expensive, I needn´t understand everything.

Posted by Romy the Cat on 03-21-2011
fiogf49gjkf0d

This post is a logical continuation of the post: My “radar of sensations” about the “good honk”:

http://www.GoodSoundClub.com/TreeItem.aspx?PostID=15814

I wonder if any manufactures would be interested to undertake bass DSET project.

It might be an even a business opportunity for somebody. Just try to imagine. You are a manufacturer who builds one single moonoblocks with let say 2 stages bass DSET with GU-48 in the end. The GU-48 runs at 2500V and GU-48 will yield 35 times amplification – it is a LOT of gain for out triode.  It is direct heated 10A filament, so let to heat it with 10-20mHz AC. The amps is furnished with DSET output transformer that with no section and with huge inductance, let say the transformer has 5-200Hz response. It I will be large but manageable moonoblock.

After you build it you deeply the amp to people – little give it up to somebody, let it to sit there for a week or two and then send it to next person. Literally: create web site, where people will be able to sight in for amp receiving and testing in home conditions.

GU48_Tube.jpg

I feel that it might be a divert array of people who would like to hear what good DSET can do with bass instead SS amp. Wail until this type of DSET hit the market of those expensive box-speakers users what bi-amp with Levinson, Lamms, Krells and Classes… People generally do not have knowledge about good bass DSETs with limitless power in A1. You can easy pump 7W to GU-48’s grid and get out ot of it near 200W, with good driver buffer you can drive it to A2 if you wish…

The advantage of this bass DSET venture is that there is no competition for this. No one does it and for many people it would be novelty. Whatever full range set out there they all roll of at 20Hz at full power, regardless what they claim, and in really the distort like crazy at bottom end. A powerful LF DSET might be very interesting education tool for many audio people and it might be a lucrative no competition enterprise with somebody. Think about it.

If I had an interest in audio business then I would do something like thins myself – built a few demo amps – send them out to circle the listening and rooms and just pick up the orders…. If somebody would interest to do something like this then get in touch with me and I will help you to source GU-48 at much more attractive price.

The Cat

Posted by N-set on 03-21-2011
fiogf49gjkf0d
I have a 1500V supply with a 872A MV bridge somewhere.
It was meant for 304TL
You've even manged to call it a fucking mastodont IIRC Big Smile
I'm not a manufacturer of any kind but I could lend this PS
for experimentation if somebody picks it from western part
of Poe-land.

Posted by Romy the Cat on 03-21-2011
fiogf49gjkf0d
What it to have DC SS amp, in the way I have it but do not drive ULF from an amps but instead to put between amp output stags and ULF an 1:1 transformer, soft of a galvanic buffer. Would it be possible that galvanic connection would do some softness and harmonic that SET is able to do? It is very possible that it is not about tube but about transformer core. Did anybody try anything in this direction?

Posted by IslandPink on 03-21-2011
fiogf49gjkf0d
Romy
How about some sort of scaled-up PassLabs F4 or F5 amp ?
ie. Class A solid-state with no feedback .
You'd get most of the characteristics of valve (tube ) SE in terms of spatiality and low-level info ( esp. in the bass ) but with less of the (daunting) problems of trying to get something like a PSE 833 amp working ( transformers being the big issue there as I see it ) .

You could trial the idea if someone can lend you an F4 or F5 ....

MJ

Posted by Romy the Cat on 03-21-2011
fiogf49gjkf0d
 IslandPink wrote:
…. transformers being the big issue there as I see it ….
IslandPink,

We are so accustom to sink inside the box of full range SET amplifiers that we afraid of everything. This is the whole beauty of  DSET concept – the output transformer is pretty much out of the picture of complexity or any responsible limitations.

Let look into this. What is the problem to wind a transformer with high ratio and high voltage? Listening my explanation you need to be informed that they come from a person who never winded any transformer in my life but it does not bother me too much. So, watt would be deference between 600V rated 25:1 transformer and 6000V rated of the same transformer? It will be no difference except the fact that 6000V rated transformer will have MUCH more dialectic insulation. More insulation means more space taken by insulation and it means more capacitance and less effective coupling. So, to wind high voltage transformer people go for some expensive techniques like winding in vacuum and with some complex and expensive impregnator to keep the capacitance lower and feeling tighten. However, all of those complex sectioning, widening techniques and care about capacitances are necessary to facilitate the HF knee of transformer. However, if we just do not care about HF and presume that our transformer will not care anything let say above 500Hz then to make it high voltage able is no difference then to make power transformer for high voltage. It will be just one primary and one secondary in high voltage DASET bass transformer and it will be pretty much irrelevant how they wind. Bass BSET transformer run in inductance and it come from turns per core mass and minus gap. The cost of this transformer will be pretty much the cost of the cost of core. So, I do not see any reason to be afraid of DSET out transformer.

The Cat

Posted by N-set on 03-21-2011
fiogf49gjkf0d
I have huge AMCC800A cores for two such transformers
and good winders with HV experience. I also have a lot of
quiality coax cable, good to few kV for the chassis hook-up.
Again, if somebody is willing to take care of that, I can lend.
I'm myself 3000km away from the Mastodont.
The signal trafo is "easy", the biggest problem is the PS. 2.5kV of clean power costs.
If one wants a FW  rectifier then the voltage swing on the PS trafo would be
huge. Bridge may be an option but:
1) higher effective resistance; some people "report" it matters on bass, but no 1st hand exp.
2) implementing 4 rectifier tubes is a PITA; I have separate heater transformers
for 4x872A, all the sockets, etc. I'm not sure how my caps are rated, some are 6kV I think.
Getting HV caps in EU is a PAIN. I think my PS chokes should stand the voltage.
One would have to either rewind my tranny for 2.5kV (my winders would do that)
or wind a new one. I have also 120W DC heater supplies (choke input I think).

Ok, I've found the diagrams.
Here is what I have modelled (for one channel so far, and R-loaded VR in the bias). It's all huge, but perhaps can be optimized.

outputPS.jpg


filaments.jpg

bias.jpg

bridge_glow.gif


Any takers??

Posted by N-set on 03-22-2011
fiogf49gjkf0d
The price to pay for it's high mu is the A2 operation. Zero crossing, big driver,  etc.
Why not 304tl? mu of 12, Rp of 1-1.5k, so a very good current efficiency and not so demanding
induction-wise. It would require a 3-stage design though...




Posted by Romy the Cat on 03-22-2011
fiogf49gjkf0d
I do not have any special love with GU-48/833 and I named them only as examples. If somebody do it then I would root for GU-48 instead of 833 as the GU-48 are much cheaper, easy to get and Russian tubes tend to be more bassy then western tubes (and less refine in upper knee). I also read comment of the guy who have both GU-48 and 833 who in a way confirmed my prejudices. Still, I have no special feeling for any of them and it might be any powerful tube that can give let say 150Win A1.

The Cat

Posted by Paul S on 03-22-2011
fiogf49gjkf0d
Yes, we are all adults here, but I hope and pray that readers just following along and dreaming about "ultimate" systems will not take it upon themselves to actually try any of this stuff at home!  Be advised, all, that the voltages being bandied about here are LETHAL!  This stuff is for experts in HV field only!  You will not "learn a lesson" if you screw up with these voltages; you will DIE!

Paul S

Posted by Romy the Cat on 03-22-2011
fiogf49gjkf0d
 Paul S wrote:
Yes, we are all adults here, but I hope and pray that readers just following along and dreaming about "ultimate" systems will not take it upon themselves to actually try any of this stuff at home!  Be advised, all, that the voltages being bandied about here are LETHAL!  This stuff is for experts in HV field only!  You will not "learn a lesson" if you screw up with these voltages; you will DIE!
Sure high-voltage has own toll, not only it can kill but it can explode very nicely but if the things are done properly then there are no problem with high voltage. We have panty amps on 211/GM70 out run 1000V. We have many people direct couple electrostats running them at 2000-2500V from transmission tubes. To build thus amps is dangers, have own complexity and have to be done by people who know what they do. To use a properly and securely made amp is not a big deal and it is no different than anything else. In fact with a proper design it might be as secure and a kitchen toaster… Yes, you would not take it with you in bathtub or in bad but anything else is very much acceptable.

The Cat

Posted by N-set on 03-22-2011
fiogf49gjkf0d
Actually I've been trying to encourage somebody to continue my design.





Posted by Paul S on 03-22-2011
fiogf49gjkf0d

This is not a safe "project" for the average DIY guy, period.  Of course, the copycats are the last to realize this.  Dreamers, if you have to ask what a veristat is, do not even consider this.

N-set, not everyone has your education and background, as you well know.

Paul S


Posted by N-set on 03-23-2011
fiogf49gjkf0d
...as I am this has been +/-  my first tube project.
I have neither a proper background (neither me nor anyone from my
surrounding have ever worked with HV) and my education is very-very poor.
I'm desperately trying to educate myself, mostly by mistakes.
If I fry myself...well, we all have to die somehow :-)


Posted by Wojtek on 03-23-2011
fiogf49gjkf0d
I have a suitable NOS surplus power transformer  from OSborne; 110/220V primary and 2825V-2000V-0-2000V-2850V at 0.5A secondary.Should be a good base for a bass SET. Not, that I  want to encourage anybody to undertake that project.

Posted by IslandPink on 03-23-2011
fiogf49gjkf0d
There's no zero-crossing with A2 .
833 or equivalent are a good choice for this extreme project and bass performance is excellent, I have heard 833 SE briefly and it has great authority.
The driver would best be some sort of small/medium output valve driving 833 via a step-down interstage. Then the gain of the 833 works in your favour . Various people have done this .
The problem remains the huge size and power consumption of the whole amp , and the core size and gapping required to get sufficient inductance down to 5-10Hz ( for ULF ) while also being gapped for eg.300ma to cover 2x 833 . I'm sure Bud would know how to do this as he works in the power transformer game . He can give you the bad news about how big it will be ;o)

MJ

Posted by Romy the Cat on 03-23-2011
fiogf49gjkf0d
What you guys are missing is a grand scale of the concept. You are talking about voltages, implementation, electric chairs and the rest of irrelevant crap but I am talking about the pure inspiration from the concept. No one runs to solder anything. If they do then I do not mind somebody got blown away by a 5kV capacitor – I hope it will be a republican.  The point is not about a factual amplifier but the concept of getting bass that was not heard before.

I am sure not a lot of you hear sound of more or less proper playback with “interestingly” reproduced upper or midbass. I would not even mention the lower bass. However if we mention it then I would like to point out that we do not have any tool to amplifier bass except PP SS amps that are not truly able to do bass properly due to crossover and thermodynamic distortions. I am not talking about the fact that we have no speakers or very few speakers that are able to go for proper bass.
 
If some of you still feel that you know something about bass then let me ask you: how many of you even heard a playback with “interestingly” reproduced region of 10-20Hz by whatever acoustic system it was and driven by a single ended, class A1 amplification free from SS thermodynamic distortions?   I do not think I will hear a lot of rise head and you will not see my hand raised as well. But I would like to hear it.

The concept is that I would l would like to hear let say DC-coupled Milq with 8 parallel 6C33C driving 1.5:1 out transformer gaped 2.5A with core able to get 2Hz of first clip at 160W. I just would like to hear what it is, even that that I intellectually understand that sub 20Hz is very problematically distorted on the recordings. Still, I would like to hear it as I would like to get a reference of what is possible by mean of audio. Sure I do not build anything like this but I would like to say, and no one else does to the best of my knowledge but I would like to declare that I have interest in it. In one way or other I would be pushing the concept, not implementation but the concept. Some of my observations in the neighboring fields suggest that there is a lot of to fish in the ULF world if to use it creatively. Who knows, perhaps another day someone would be inspired with the idea, knock in my door and will drag into my room a refrigeration-size amplifier with a pair of fire extinguisher built-in on the chassis…  If do not dream about it then no one will knock at your door….

Rgs, Romy the Cat

Posted by N-set on 03-24-2011
fiogf49gjkf0d
Some people bias 833 at 0V. Some in A1 but to get teh stated power
it must go into A2. Do you or anyone have
normal plate curves for 833 instead of that multiamp
idiocy usually published? Say up to 200mA. Same for 304tl.
Outpur tranny--this is the function of 833's big
internal resistance. So I'd still lean towards e.g. 304tl.
More current efficient, the parellel structure some purist bark at
should not matter at bass, if it matters at all in any sense.
I think OT is an easy things here. Just the iunsulation.
Any good kV winder can do it, once you calculate the tranny.

Inspiration--there must be a balance between inspiration
and technicalities. One without the other is useless.
I'm much better in vocabulasing technicalities
unfortunately...The Voice Of God would be my inspiration!
I don't need more words.


Posted by N-set on 03-26-2011
fiogf49gjkf0d
This thread has triggered some old stupid sentiments. Feel free to remove this post
if irrelevant. I've been thinking how to rationalize my old design in the present context
and make it less idiotic.

First, know the enemy. I don't know the enemy. I assume the bass arrays are very low impedance, 0.5R and require
good controll (damping factor). This should narrow down the tube. Perhaps I'm wrong, but
833/GU-48 does not too look good on paper with their high Rp to achieve good damping,
so I'd stick with 304tl. Apart from Rp at 1.5k level it has one more interesting
degree of freedome to play with: 304th variant with highre Rp if needed. 
Biased at 2500V/-250V/120mA and loaded with 7k or so, 304tl should give around
100W, perhaps 120W with a reasonable (?) damping factor, if I'm not mistaken.
The good point is that 304tl must operate at or close to max, otherwoise the tube life is reduced.
5Hz would require 50H primary ind. Doable at 150mA with AMCC800A cores,
1mm gapped, I think 70H too.
To drive I'd go for something non DHT not to complicate:
6E5P DC to, say, EL34 triode connected@400V/60mA and then 1.5:1 stepdown.
Stepdown with 80-100H@60mA primary should be doable too. This should drive 304tl into mild A2 if needed.
PS: I went for the bridge due to the trafo constraints (I had too small lamination), but now I'd do
3kV-0-3kV trafo then 872A FW. For the input and the driver: stacked Romy style supply with mega uF etc.
The most complication is fixed bias...as I see it, due o the DC, each stage would require it's own bias
supply...pain in the ass
THis is something I'd try at this moment, but again--I don't know the enemy, so I'm blind. 


Page 1 of 5 (107 items) 1 2 3 4 5 »