 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
92.2.70.128
I've been building some 2a3/6B4G amps and am looking to standardise on either the 2a3 or the 6B4G. I listen to a lot of classical music and want smooth sound with good tone and detail - no hardness (I find Sovtek unacceptably hard).
In 2a3:
Prefer old biplate Chinese to Sovtek monoplates
- Haven't compared with Chinese 2a3C
- Haven't compared with NOS biplate 2a3
In 6B4G:
Prefer old biplate Svetlana 6C4C to Sovtek monoplates
- Haven't compared with NOS biplate 6B4G
I haven't directly compared the Chinese 2a3 biplates with the 6C4C - different sockets and voltage.
NOS 2a3 seem more plentiful but more expensive than 6B4G. Chinese biplates are virtually extinct. 6C4C Svetlana seem fairly plentiful in Russia still.
Can anybody give me views on which way to go here - 2a3 or 6B4G? What would you do?
Andy
Follow Ups:
I have a lifetime supply of 2A3, 6B4 and 6A5 tubes bought over several decades. While I have used National Union and RCA 2A3s in my primary system, they along with the 6B4 and 6A5s are now relegated to supplying my triode amp collection. I will only use new tubes for my "serious" listening.
So, I recently bought a quad of JJ 2A3-40s for about $500 for my new amps. I bought 4 so I would have spares in case of incident but also because I still intend to build a PP 2A3 amp. All 4 biased up about the same in my amps built expressly for the JJ 2A3-40: 290Vp-k @ 64mA for a 18.56W plate dissipation. This is 46% of the allowed dissipation according to JJ. The tube biased exactly to the 2A3 plate curves; it is NOT a 300B with a 2.5V filament as is still reported occasionally.
I am very happy with these amps and the JJ 2A3-40. While I have not had an opportunity to compare any other tube in these 2 stage amps because of the operating conditions, they do compare very favorably with my 3 stage amps running the AVVT mesh at less dissipation. Transparency has improved but the most noticeable change is in the mid range which has much more weight and authority while being cleaner and better defined.
So for the foreseeable future I will standardize on the JJ 2A3-40 and when finances permit will probably buy another spare set.
BTW just to be clear: the last time I used a Chinese tube in any of my amps was over 10 years ago and the results were not pretty...
Andy,
You can not judge optimally what Finals sound best until you operate the tubes with a dissipation that is conservative, so the tubes do not sound stressed. You tell me privately you are at 14.575 Watts, way too high.
Its silly discussing the sounds of DHTs when they are all operated near maximum dissipation, and are sounding thermally stressed!! Relax the dissipation and relax the sound.
This applies not only to the Finals, (which you need to correct first), but to the tubes that precede it in the front end.
Reduce your finals to about 10.5 Watts dissipation as a start, and then try for Golden Ratio dissipations of your front end tubes. Set the front end dissipations by EAR until they each lock into a sonic sweet spot.
Then you are cooking with the design.
Easy does it, have fun. Report back to us all.
Jeff Medwin
I'm very interested in operating points. I leave my amps on all day mostly - the TV and Radio go through them as well - everything comes from my computer > DAC > amps.
So what do we know about longevity of 2a3s at 15W, 12.5W, 10W. Any Data?
Thorsten used to say Sovtek monoplates sounded better at 45 operating points.
Andy
Dennis Fraker reports on this Forum getting 50,000 hours PLUS on AVVT meshplate 2A3s at 41-43 mills, 245 P-K, AND equally important, they sound better there. As compared to the usual 2,000 hours (to maybe 10,000 hours) and degrading continuously after about 200 hours of initial service.If you can get 50,000 hours, spending money for a EML meshplate becomes totally justifiable, or, a less expensive JJ2A3-40 is a no-brainer!! (Either of those brands readily outperform the tubes in this thread you were considering.) The JJ has a 40 Watt 300B plate, so it is in "loaf mode", long lived enough for your grandkids to enjoy hearing.
Yes, Thorsten has heard and reported about reduced dissipation also, but Dennis Fraker was designing his 2A3 amps this way in 1989. A Type 45 is usually rated and run at 10 Watts dissipation, and it is that dissipation we (mostly Dennis) suggests for the 2A3. For a Type 45, we have reported it best at 7 Watts dissipation, lower current, maintain the voltage at about 250 VDC P-K.
Cheers,
Jeff Medwin
Edits: 07/10/12
IIRC, Thorsten was disappointed with 2A3 sound in general, much preferring the 45. He tried the 2A3 at 45 operating points and preferred the result, though the Sovteks were still not up to the same general quality level as his preferred 45s. No surprise there.Although I don't always agree with Thorsten's subjective preferences, I greatly appreciate Thorsten's technical contributions and subjective descriptions over the years. His experimentation with 300B operating points (like running type 45-like I/V bias ratios), increasing loads and decreasing dissipation were most informative.
I find it curious that some folks "don't like 300Bs", yet have not tried the lower current (and dissipation) and higher load configuration and/or the more 'modern' sounding 300B tubes.
Back on topic.
Cheers.
“As long as we have any intention to be right… we should be wary. So long as words have the slightest ego attachment, they are dishonest.” Charlotte Joko Beck
Edits: 07/15/12 08/14/12
Thorsten might have appreciated AVVT and EML. Sovtek's 2A3 really is a bargain-- and that is all!
Just don't expect it to sound like a good tube, and you'll be fine!
---Dennis---
I agree Dennis. I have the understanding that when he was trying different tubes, there were few new 2A3s available and he did not try many. The marketplace is very different now.IIRC, he did like the AVVT 300B.
Cheers,
Ray
“As long as we have any intention to be right… we should be wary. So long as words have the slightest ego attachment, they are dishonest.” Charlotte Joko Beck
Edits: 07/15/12
I get a kick out of people comparing 45, 2A3 and 300B. Of course, with each type, we have differences.Here I offer some explanation-- maybe this can help some of you a bit.
The common NOS 45's were built with decent plates, a decent grid set and a "W" or "M" filament configuration. Shaping the filament into a "W" or "M", while a cheap way to assemble tubes, creates all sorts of distortions and harmonics that ARE NOT THERE on the recording that is being played back. Despite this, some people like to listen to these tubes. What do they have going for them? FORCED to run LOWER PLATE CURRENT-- which GREATLY benefits output transformers! They are also EASY TO DRIVE-- which benefits the anp's driver stage operation.
The NOS 2A3 was built as a Single-Plate, and as a "biplate". The Single-Plate is rare, because it was too expensive to build for the Depression Era.... and was dropped from production. These were under RCA, FIVRE, CUNNINGHAM, etc. All were RCA. Raytheon also did a Single-Plate, but the construction was different. This is also rare. They sounded good, and were the most accurate reproducers of their day. Today, only EML has beaten them.
NOS 2A3 Biplates are a sandwich of two "45" plates, with appropriate grid and filament arrangements. There are two "W" or "M" type filaments in these, and they are connected together, as are the grids, and plate-sandwich.
They're powerful, reliable, euphonic, and have high distortion and timing-out-of-sequence problems. These sound good on non-revealing speakers, but sound trashy on transparent speakers. Today, cheapo tube makers still make them. Chinese varieties of these are even much worse. The best-sounding of the NOS Biplates are of old Sylvania manufacture-- with Spring-Tops tensioning the filaments. You will often find these in old Philco and Silvertone (Sears) boxes.
The 300B started life at Western Electric, and was a higher-power version of the W.E. 275 telephone repeater amplifier. They were fair-sounding, having a decent vacuum and decent materials. The "M" or "W" style filament structure created hot-spots in their radiation pattern, rendering them useless for accurate musical reproduction. The CHEAP filament system was used because these were designed for telephone-repeaters, not for the ultimate in High-Fidelity. Copies abound and they're generally even worse musically. The W.E. are more reliable than the copies.
Modern EML 45, 2A3 and 300B are all the SAME kind of tube-- all built correctly. Arguing over which one of THESE is best is pointless-- all are better than any of the other parts in your amp. Of course, the DICTUM still applies-- the lower plate currents sound best.
Some people prefer the "45" even in the case of EML-- all they are benefitting from here is being forced to run the lower plate current of the 45-- which is a GOOD THING! Actually, THE best tube of all of this discussion is EML's 2A3 Mesh-- run at almost (just a little higher) "45" specs.
---Dennis---
Edits: 07/15/12 07/15/12
It's explained pretty good in the book this old post refers to:
http://db.audioasylum.com/mhtml/m.html?forum=tubediy&n=198107&highlight=round+plate+gsh&r=&search_url=%2Fcgi%2Fsearch.mpl%3Fsearchtext%3Dround%2Bplate%26b%3DAND%26topic%3D%26topics_only%3DN%26author%3Dgsh%26date1%3D%26date2%3D%26slowmessage%3D%26sort%3Dscore%26sortOrder%3DDESC%26forum%3Dtubediy
"The "M" or "W" style filament structure created hot-spots in their radiation pattern, rendering them useless for accurate musical reproduction."
So Dennis, do the electrons come straight off the filament and go straight to the plate?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Straight to what part of the plate? OK- let's suppose-- make that the nearest point on the plate for any given point of the filament, unless something changes that.Now, how far is that point (on the plate) from the plate's lead-out for the tube's common plate connection?
One can draw this out and look at the distances between the place where a certain part of the filament was closest to the plate-- and how far that part is from the common plate connection-- the plate lead-out-- and see what happens.
Remember also, that these NOS tubes have a single, long filament in some cases, and have a parallel pair of single filament runs in others.
EML is different-- there is a solid bar along the top of the filament structure-- it is a common rail, and there is another solid bar common rail along the bottom of the structure. There are 8 completely separate vertical filaments-- each is welded at both ends onto a common-rail bar.
Each is the same total length, and each has the same path length to the common rail. Each also has the same biasing as each other filament at the same vertical point in any of the 8 identical strings.. This is a great sounding filament setup. It isn't perfect-- that would be a one point filament aiming at one point on the plate.
If you look at Vintage 811A's, you'll also find something interesting. Those are High-Mu, high plate resistance triodes that run usually over 1000 volts on the plate.
Plate current isn't very high-- usually 60 ma. or so in Class B amp service. Well, on the BETTER ones, they ran two identical internal plate leads to the plate common connection on the tube's top-cap plate connection-- even though they were running relatively high voltage at low current-- one lead could certainly carry it. . Each went to an opposite side of the plate. CHEAP 811A's only used one lead from the plate-- only one side got loaded..
Which tube sounded better? The one with two leads that were equal length to the top connection, and balance-loaded the plate from two opposite sides, not just one side..
Filament setups can get more complex as manufacturers attempt to balance-out some hum in DHTs, and also try to make the thing cheap to run on automated assembly. The single, or double long filament has different bias points along its length, and some are wired in "M" and "W" shapes to boot, making machine-assembly fast and cheap..
If we have a tube with both of these defects, we have different path lengths AND different path length biases.
The JJ 2A3-40 has only one major defect. They did run the filament wire in 8 vertical runs- good-- it shows they understand that, and corrected the relationship to the plate-shape.. But there's still only one, long, single filament. The only defect here is different biases at different points along the filament length. This was done to make the tube-- which is excellent in all other ways-- affordable.
Of course, that defect also occurs in each of the 8 filaments in parallal in the EML-- but the performance anomaly is greatly lessened-- the tube is simply the best that we can actually go out and buy.
It is also NOT cheap to build!
Well, anyhow, it shows they know what they're doing. They're working very hard to build us good stuff.
---Dennis---
Edits: 07/15/12 07/15/12
The single, or double long filament has different bias points along its length
every filament has different bias points along its length.
dave
"Straight to what part of the plate? OK- let's suppose-- make that the nearest point on the plate for any given point of the filament, unless something changes that."You didn't answer my question.
Do the electrons go directly from the filament to the plate or is there a destination in between?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/16/12
When a filament is heated electrons come free from the filament coating and form a huge electron cloud known as a space charge. When B+ is applied to the plate some electrons leave the space charge and migrate to the plate creating a current path.There are so many electrons in this cloud that the plate will never deplete them and they're being replenished from the filament continually.
These electrons move reasonably quickly across the vacuum, about 1/3 the speed of light, but they move very slowly in metal, etc (the filament winding, the cathode resistor, the plate, the tube pin, etc.) about 0.0093 cm/s for 1 amp depending on wire size.
We do not listen to these electrons that have created this current path but instead the current that uses this path.
The current that uses the path, created by the electrons, moves at (almost) the speed of light.
So the differences in the path lengths, that you seem so concerned about, are meaningless.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/17/12
What's all this talk about electrons?
Electrons only matter in non audio applications. When it comes to audio amplifiers MUSIC is what flows across the tube elements, not electrons. So therefore we don't have to abide by the rules of electrons.
It's so simple!
be careful, someone might think you're serious.Here's a bit more information.
First a quote (from a different source) to set the stage.
"However, it's important to understand that electron flow and charge flow are not the same thing. It is the flow of an electric field through the wire that creates the effect we understand as electricity, not the actual movement of electrons, which is relatively slow."
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/18/12
Don't think so Gusser. Ever read "Inside The Vacuum Tube" by John Rider ??
There is a space charge of electrons.
Jeff Medwin
a
or Dennis could just read this.First a quote (from a different source) to set the stage.
"However, it's important to understand that electron flow and charge flow are not the same thing. It is the flow of an electric field through the wire that creates the effect we understand as electricity, not the actual movement of electrons, which is relatively slow."
Now a quote from the pdf,
"Current flow does not mean "one electron in - same
electron out." This happens very slowly as that
individual electron travels through the conductor.
Instead, it means "one electron in-one electron out."
This can happen almost instantaneously, as Figure 2
suggests."And the instantaneous speed (almost the speed of light) of this understanding of "current flow" is what makes all of Dennis' talk about distances absurd.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/18/12
What is going on here are several factors.
There is instantaneous electron transfer, and there is also transfer from Space Charge. In a Pure Tungsten DHT, ZERO space charge is required to effect electron transfer, although whether or not that condition exists is determined by the voltages and currents that that tube is being run at.
In most tubes that audiophiles use, there are both factors at work simultaneously.
This is Summer in Montana, and my posts will be mostly from what I'm able to deliver quickly and easily, as I am busy with irrigation systems, and a lot more.
If you wish to do further research, you will find that one needs to consider all of the facets that apply to a given situation-- not just focus on a single question-- which leaves out pertinent collaborating information.
One-sided approaches-- to anything-- usually leave something out, and that is why I didn't answer your question-- I didn't have the time to give you a complete answer.
I don't now, either, so you can start in RDH4-- Pages 2 through 5. That's not even close to a complete explanation, but at least it makes sense.
---Dennis---
I'm not asking you questions to educate myself. I'm asking you questions to educate you.You said " If you look at Vintage 811A's, you'll also find something interesting. Those are High-Mu, high plate resistance triodes that run usually over 1000 volts on the plate.
Plate current isn't very high-- usually 60 ma. or so in Class B amp service. Well, on the BETTER ones, they ran two identical internal plate leads to the plate common connection on the tube's top-cap plate connection-- even though they were running relatively high voltage at low current-- one lead could certainly carry it. . Each went to an opposite side of the plate. CHEAP 811A's only used one lead from the plate-- only one side got loaded..
Which tube sounded better? The one with two leads that were equal length to the top connection, and balance-loaded the plate from two opposite sides, not just one side.."
Only one side got loaded????? Are you kidding me?
Dennis, the only explanation I'm looking for from you is for you to explain why you thought the small path length differences will make any audible difference when it is clear that the "music" moves through our circuits at almost the speed of light?
The only real answer from you should be "Hey, now that you have shown me how these things work I see that those signal path length differences don't matter. Thanks Tre'".
You're welcome Dennis.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/19/12
They "don't matter"? What a joke that is!
Those path lengths are THE MOST important item in ALL of amplifier design! They are in tubes also, TRE......
---Dennis---
"Those path lengths are THE MOST important item in ALL of amplifier design! They are in tubes also"
In some applications starting at a few MHZ, that is quite correct. Again in some applications where timing relative to another signal matters such as radar and television.
But in base band audio, it is a joke! You are really going to make a fool of yourself if you continue this idea here.
So what? The truth of the matter is proven.
Honestly! No hard feelings, but you guys can't possibly be into High Fidelity.
Theories, yes. I do prefer performance, though-- it makes me so much more happier! (musically).
Oh Well! To each his own they say.
---Dennis---
.
'So what? The truth of the matter is proven'
Proven because you say so? Dennis you are posting misinformation again. Please validate your claim or withdraw it.
The issue here is that some innocent people who read your crap might actually believe it.
You are claiming that there's no performance advantage to a symmetrically-built vacuum tube?
That is astounding!
---Dennis---
"You are claiming that there's no performance advantage to a symmetrically-built vacuum tube?"
I don't think anyone is saying that. Of course there are a plethora of issues in vacuum tube construction - even at audio frequencies.
The issue here is your frequent claims that inches of lead length differences are audible. This is simply not possible with the physics that govern our universe.
It has been proven not only mathematically but also with countless examples over the past 100 years of electronic audio systems.
Now there is an issue with acoustic energy delay and that where I believe this confusion in audiophile circles comes about. Yes, there are "time aligned" drivers in some speakers. But one must remember that sound travels at 750mph at sea level. Electrons travel at roughly 66-68% the speed of light in wire. Quite a difference. Therefore it is somewhat critical to align distances on your speaker drivers. But that does not extend to the physical wiring.
Look at a modern computer mother board. You will no doubt see some traces zig-zagged and look like a 'sinewave' Well this is exactly what you speak of. The multiple wires in a data buss must be equal lengths at the several hundred megahertz they operate at. Even a few fractions of an inch differences can cause a problems with skew at these frequencies.
But for the last time, this has no significance with base band audio electronics.
This is a good discussion and shows first rate intelligence on your part.
I'm not perfect. However, I think you've missed something really important here. (God knows I've done that too!).
I get major changes in sound qualities in an amplifier by changing the length and diameter of conductors-- in increments as small as 1/4 inch. MAJOR. When I say "proven"-- one can hear these changes every time they're made-- there's simply no question to anyone present while the changes are being made..
That is why, when I first began posting a few years ago-- on TUBEDIY-- I rated layout and wiring of an amplifier near the top in priority. This was to allow one to get lead lengths right.
Good engineers have posted-- and I know many more who do not post-- that ground systems and ground planes in circuit board layouts are always markedly inferior to good hand-wired circuitry.
This is not always true! A really good circuit board engineer will build boards that are pretty darn good. But, sooner or later, the engineer, no matter how good, will simply run out of choices on a circuit board-- some things will have to be placed at the wrong placements and wrong lengths.
This is where the same genius can work wonders with hand wiring. He will be able to place every part, every wire- right where it belongs. He'll redesign the entire amplifier chassis and parts placements to get it. He will also be able to relate lead lengths to actual function, and then redesign layout again in order to accommodate that, ALSO accommodating all else that was learned before that.
In the end, a balance will be achieved that is the best that one can do with what's available-- and the problems and materials at hand.
That this greatly outperforms anything else is self-evident to those who are in the Lab-- observing the results-- and to users who have bought the product. To us, it's proven.
To an outside observer, you might want numbers and theories-- which you may call "science"-- but I think that most of us realize that we are all scientists-- the best of us are also empiricists-- in the full meaning of the word.-- What works is what works.
I know I keep repeating this Company ad-nauseum on here, but it's appropo nevertheless. CAT equipment-- especially dozers.
These guys are great engineers, but the way they design is they do the best they can, and then issue prototypes to the meanest and stupidest users they can find-- backed up with super-intelligent bosses who know what's going on on their jobsites.
CAT is observing. As things break, as they underperform, as they make PEOPLE MAD--- CAT fixes them as they occur. The end result is really great dozers.
Porsche does this also. The various Porsche Clubs drive, play, and race in Porsches. Over the years, the users kept refining the product. Hey! It's a blast to drive!
---Dennis---
"I get major changes in sound qualities in an amplifier by changing the length and diameter of conductors-- in increments as small as 1/4 inch. MAJOR."I don't believe you. I think you are making that up. If you really believe what you say then you are delusional.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/20/12
"The issue here is your frequent claims that inches of lead length differences are audible. This is simply not possible with the physics that govern our universe."
Give it up Gusser ! An inch of bad wire can easily kill the audio experience.
Jeff Medwin
mqracing made a valid point about building and listening to a circuit application before dismissing it.
Dennis Fraker actually builds and listens to the circuits he is talking about here.
He is not just trying to determine what a circuit will sound like based on a purely theoretical model.
dt 667
What, experimentation, empirical learning, iterative methods, questioning, taking a different approach and actually listening to the results? Sharing those results and encouraging others to experiment while protecting the details, his IP?! Assemble the mob, we have a witch in our presence!Funsies aside, I am all for Dennis' approach; however, sometimes his passion and direct-to-forum communication result in some posts that rightly spark debate (how's that for diplomacy?).
Cheers.
“As long as we have any intention to be right… we should be wary. So long as words have the slightest ego attachment, they are dishonest.” Charlotte Joko Beck
Edits: 07/21/12
"I am all for Dennis' approach"So am I.
It's been said that the wire from the plate of the driver tube to the grid of the output tube is important in Dennis' design.
So add another wire to it, twice as long. Be careful to keep it away from other wires and the chassis. We don't want stray capacitance or inductance to muddy the experiment.
Do this to just one amp.
Now listen and see if you can hear a difference.
If you think you can, please have a friend there to make sure you don't know which amp you're listen to. Can you still hear a difference?
That's how I know that it doesn't matter.
How's that for experimentation, empirical learning, iterative methods, questioning, taking a different approach and actually listening to the results?
Peace
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/21/12
Hi Tre,"How's that for experimentation, empirical learning, iterative methods, questioning, taking a different approach and actually listening to the results?"
To answer your question in brief: unconvincing. I will qualify that by stating that I used to develop medical / clinical research protocols for a living, so feel free to consider my judgement of your approach harsh if you wish.
TBH, *all* experimentation is flawed. As is our so-called Scientific Method, which even the most general health research text states. Although it is not my area of expertise or even interest these days, I think most experimentation discussed on audio forums is naive at best. Despite this, we still seem to be learning... So, let's do what is practicably possible; let’s experiment, even if our methods are imperfect. But by the same token, let us also not attribute more to the 'results' that we should. The results of our experiments are not absolute truths.
I'm not going to crap on about general experimentation any longer; books have been written about it. But I will mention placebo effects before I sign off. I find it interesting that when expectation effects are discussed, people are almost implicitly discussing the positive expectation effects. Very rarely discussed are the negative expectation effects that might be prevalent among cynics and sceptics. For example, the expectation that some factor will *not* have a causal effect. Expectation effects cut both ways. This serves as background to the point that just because you don't hear something does not mean it does not exist or that others cannot hear it.
Of course, this makes it increasingly difficult to refute fallacious claims.
Cheers.
EDIT: I am also sceptical of the claims made regarding the length of wire, at least the lengths and context we are discussing here. My comments were referring to Dennis' approach in general and I was being rather light-hearted about it, though that could be lost in translation. It was off topic and has dragged this thread out a little longer than it deserves (to my mind). Apologies for the diversion.
“As long as we have any intention to be right… we should be wary. So long as words have the slightest ego attachment, they are dishonest.” Charlotte Joko Beck
Edits: 07/21/12 07/21/12
It's not Dennis' approach that I object to. It's making repeatedly making absolutist statements when they are inappropriate.
Opinion and ideas are fine.
If he had said. "I don't agree with you gusser, I reckon I can hear differences when unequal lengths of wire are used" that would be great.
However, he has stated it is PROVEN, without giving a shred of evidence to back it up.
I know where you are coming from! I also take issue with personal preferences and experiences being promoted an absolute; qualitative as quantitative... as subjective Truths.That is not to say for a moment that I think the subjective and qualitative is unimportant - it is very important to audio if you do it for others' (and your own) enjoyment.
Misrepresented "facts", intentional or not, are open for discussion and debate. I have called out Jeff and Dennis on occasion and may again. If I do, I hope I won't make it personal, a pissing contest, or gloat if I score a win (not saying you are, BTW). The moment I do this I am part of the problem: distracting the discussion and confusing the issue, adopting logical fallacies, being generally unconstructive and affecting people in a hurtful way... at that point it becomes more about my personal neediness, not any positive outcome.
In closing, I think Jeff and Dennis are two guys I'd really quite like. Yeah, they occasionally say things, they seem a little out there at times, and their passion appears to lead to an exaggerated emphasis on certain things, but hell, I like their genuine passion and resolve. Ans so what, compared my own crap, theirs is minimal! We are all imperfect - I try to remember that and keep my self-righteousness in check.
Cheers,
Raymond
“As long as we have any intention to be right… we should be wary. So long as words have the slightest ego attachment, they are dishonest.” Charlotte Joko Beck
Edits: 07/23/12 07/23/12 07/25/12 08/14/12
.
Have Fun and Enjoy the Music
"Still Working the Problem"
You're wrestling with a pig Tre'.
Hi ! I am actually surprised you have not had the opportunity to audition real NOS USA 2.5V and 6.3V triode opt tubes, yet. Many lunar eclipses ago, I wondered as you do now. However, back in the eighties, there were no easily available Russian triode opt tubes stateside and most of the then available Chinese tubes were horrible. Reliability and sonics were not yet up to par. Having compared my used USA tubes to a friend's Russian 6B4 equivalents, I stuck with the USA tubes.
If I wanted a harder brighter sound, I used the unattached twin plates, previously known as the bi-plates. If I wanted a more natural sound, as I always prefer, I used the folded M or W plates, which I think many folks are now calling the bi-plates. If the plates touch and the plates appear like folded M or W, these are the most natural sounding to me. Regarding 2A3s, I usually prefer them over the Single Plate 2A3s. The single plate has unique qualities, but it is a harder sound in my PP amps.
Having already built 2A3 PP amps with ITs in the mid-80s, I then built some amps for the 6.3V and 7.5V tubes, using four pin sockets and high impedance opt trannies. That way, I could roll 6A3, 50, 10, VT-52 et al plus compare the sound with my 300Bs and then 252As (don't ask when and where they went)in my other SE amps. Also, having a few types of PP amps around the house, I could really compare and realize which to keep and which to flip. Well, I preferred the 6A3 as most natural, even in my 300B amps running at 5V fil ! Next preferred were the GLobe 250s and then the VT-52 tubes.
The 10 family was unique. I favored the tall plate globes and the non-gold grid varieties of tall plate STs. I actually despised the small plate ST 10s at that time, as they were way bright sounding in my amps. The Cunningham silver or black plate 210/310 were great tubes; bright sounding(and looking) but still here. Then, there were the white base 801s by RCA. These are keepers. Rugged, reliable and very nice sound. They are still available reasonably today. But, you know that.
I have a real opinion about the 2.5 vs. 6.3V tubes. Since the current is much higher on the 2.5V tubes, I believe the electrons flow more efficiently; maybe even faster. Call me crazy if you like. But, I have preferred to hear a 2A3 over a 6B4 or 6A3. Any of these win over the WE 300Bs for natural sound. Another great comparison is the 245 vs. the taller plate VT-52s. Yes; I want the 245s to stay. I still keep some VT52s around; but I sometimes wonder why. I also keep some large GL 10s as well as the RCA/Cunnings; but I rarely use them. The 801s are sufficient for me; but I only use them on occasion. When I do, I revert back to the 6A3; unless they are used in my late '20s Samson amp.
So, if I had to ditch some, and I still have many, I would move the 6B4s rather than the 6A3s. I like that four pin versatility. A strong 50 or 10 can run on 6.3V fil all day and night for many full moons. If you need octals, you will undoubtedly prefer the USA W or M plate 6B4s over the twin plate 6B4s and Russian or Chinese varieties. I have faith in your ears.
If you need some USA 6B4s, let me know. I could use the fair money. 6A3s I am reluctant to move; yet. Ditto with my quad of 250s. They have such a robust sound. I have the trannies to build a pair of Amertran 250 amps(as in Riders #1) and I suppose I will get around to that before I retire. But, for now, I am happy with my PP 245 amps. All DHT, 30-245-IT-245s. They have convinced me to move most of my 2A3 amps away. But, trying 45 STs usually disappoints me. I need my globes in my amps.
Since you seem to build or rebuild continuously, don't move out any tubes; yet. When you get to the point that you can reliably compare all the varieties within a small time frame, then you can convince yourself which stay and which go. But, comparing 2A3s with 6B4s or 6A3s is not as identical as you might think. Circuitry wise; yes. Sonically ? You have to decide, my friend. It is your listening happiness at stake...ENJOY !
Thank you for sharing generously from your experience with these tubes - experience which significantly pre-dates mine.
big j.
"... only a very few individuals understand as yet that personal salvation is a contradiction in terms."
Thank You ! I like to share the joys of sound !
I've been looking at prices for US NOS tubes and 2a3s are around $75. 6B4Gs at around $35.
6a3 vary quite a lot, because I guess there were a lot less made. Can be same as 6B4G or a bit more. Are there many 6a3 still about? May be a bit dodgy to commit to those unless there's availability.
Andy
Andy, this mystifies me! Since the EML Mesh Plate 2A3 are available, and since the JJ 2A3-40 are available, and since the JJ-2A3-40 is superior to everything out there except the EML Mesh 2A3-- WHY are you interested-- even in the slightest-- in those other old tubes that never did perform well enough to be accurate music reproducers... no matter how you ran them!
Those NOS tubes have so many design/construction mistakes in them, why would anyone use them today when people have learned, and have built better tubes?
It makes no sense! When something better comes along, why not just use it? It could save you a lot of messing around and never getting what you could have had.....
---Dennis---
Hello Dennis,
I haven't heard these tubes, unfortunately. Bear in mind that though I'm posting on the SET asylum because people here know their 2a3s, I'm actually using a PP 2a3 amp. That means FOUR output tubes! EML mesh comes to over $1,000 a quad. And so does JJ 2a3-40. I'm not looking to spend $1,000.
My amps are on all day and sometimes all night if I fall asleep to music - my Mac Mini is the source for everything - TV, radio, music etc. So I may be looking at replacing tubes once a year at least. So I'm putting a ceiling of $250 for a quad, and preferably under $200. That's perfectly possible with 6B4G, 6C4C or 6A3.
I've made several 300b SETs with excellent iron and after investing a lot of time and money in these, I find that a good 2a3 PP amp works better with my speakers which are full-range Mark Audio Alpair 10s. I like them - they're musical and detailed and easy to drive, and are in thin vertical columns so fit nicely in my front room. It's taken decades to evolve this system. I would prefer panal speakers - had Maggies and Apogee Caliper Sigs and stacked Quads previously, but I just don't have the space in my listening room. So same would apply to horns, which aren't my taste anyway. I don't need the sensitivity of horns - my Alpair 10s can be driven by a 300b SET anyway.
So back to the 6B4G variants. I prefer all the biplates I've heard to the Sovtek monoplates on classical music, which is my main listening. Friends who are musicians like me and listen to orchestral music have tried Chinese 2A3C against 6C4C and prefer those Svetlanas. I can't confirm or deny this myself - haven't tried them and would like to.
I don't doubt that mesh 2a3s sound great! Don't need the JJ 2a3-40s for power. But look at the $$$$$$.........
Best
Andy
I've had one email with another view of the JJ 2a3-40:
"I don't recommend the JJ2A3-40. A very powerfull bass and dynamic sound.
Excellent macro dynamics, but not so good in micro. It's a Naim like PRAT
tube. The even more problem is the glare in upper mids and highs. Your
fullrange speakers are high resolving and the aluminium cones are sensitive
to glare."
This goes on to recommend the black coated glass dual Monoplate 6B4G or 6A3 as transparent and a very refined sounding tube. Not harsh.
This could be something like my own limited experiences. I do like biplate old Chinese 2a3 and 6C4C Svetlana. smooth sound. I haven't found many monoplates that I like. In 2a3, Sovteks all have that glare. In 300b I don't like Sovtek or EH 300b. Haven't heard Svetlana. Best I found were the Chinese ones. In fact I prefer just about all the Chinese ones to the Soviet ones.
I haven't heard any EML.
Andy
No one's heard things properly, ANY of the tubes you mentioned at all, when operating them as you run yours, at 95% of rated dissipation. They are all STRESSED sounding, due to excessive thermal dissipations.
Fix the amps and then re-listen. Be conservative and enjoy the roses.
Jeff Medwin
Andy,The JJs 2A3-40s are probably about half the price of a EML mesh. $500 a quad, not $1,000.
I pointed out right away - your DIY 2A3 amp's over 14.5 Watt dissipation is wayyyy too high, and the cause of (1) stressed sonics and (2) short Finals tube life.
If you replace a quad of outputs twice a year, that is 180 days at 24 hours, or a tube life of 4,320 hours.
That means your amp is mis-designed, because it is running the Finals too hard.
A JJ 2A3-40, with a 40 Watt plate, will easily go 50,000 hours at the 10 to 10.5 watt operating point I suggested, and importantly, it will sound better than at 14.5+ Watts of dissipation.
So, do you want to buy TEN sets of lesser tubes at $250 a quad set ( $2,500) to get stressed-sounding and constantly degrading performance...
OR...
wouldn't it be a LOT smarter to spend $500 to $1,000 on ONE quad set, $8 on some Mills MRA-12 Rks to run them at 10.5 Watts, to obtain five years of consistent and higher-level performance?
The way I see it, you are now paying BMW prices for YUGO performance.
Increase the Rk, use better tubes, you will love the sonics, and I would suggest its cheaper and better by far !!
Jeff Medwin
Edits: 07/11/12 07/11/12
I'm not sure about the JJ 2A3-40 pricing right now, but these were about $115 USD each last time I looked.Since your amp is PP, the requirement for ultimate transparency is greatly lessened over a simple S.E. application, but you'll still need repeatability in tube characteristics to run 4 of them at once..
The Electro-Harmonix versions of the Sovteks are way better than the plain Sovteks. In S.E. service, both are unremarkable and have stunted Bottom-End. However, in Push-Pull, you're going to get Bottom-End if the amp's design allows it-- tubes are a lesser factor.
4 output tubes is a lot of complexity to run simply to listen to music, however-- but people do it everyday.
If this is working for you, perhaps the rugged reliability of the JJ-- once you find 4 that are similar to each other (that's a tall order for any output tubes) would fill your needs for a long time. They seem to be completely "Bulletproof" in our S.E. amps.
I have a friend who is running 4 of them in a Mastersound integrated stereo amp-- has been for years-- upgraded the JJ power tubes last year--they're still running OK.
The last of the JJ's I purchased cost me $105 USD each.
The Electro-Harmonix and Sovteks that I have around here were cheaper-- the Sovteks a lot so. I have quite a few-- and no use for either.
I also sold-off all NOS as soon as I knew that they're not doing the job right either-- all kinds! I probably kept a few of the best NOS-- Spring-Top Sylvania, etc. If you can use any of this, let me know.
I don't sell JJ or EML. I buy those for my amps.
---Dennis---
Edits: 07/11/12
Hi Dennis and Jeff!I was pricing the JJ as follows - $255 each at http://tubedepot.com/jj-2a3-40.html. That was the first I found on Google. My mistake - reading closer this was for a matched pair. Seems a typical price
I can see you have no use for the Sovteks, Dennis - you make a very high quality product. I don't use them either. I don't mind about the bass but I can't stand the hard treble - violins are plain unpleasant. The EH may be a bit better - haven't heard them. It's the orchestral violins that are the issue - they absolutely have to be sweet and smooth. I read a test of the Sovteks where they had between three and ten times the distortion of NOS 2a3s.
I'm still going for 'ultimate transparancy' even if it's PP though! The differences in the finals are quite audible. I agree we have a lot of complexity in my amps - three DHT stages and 4 tubes per stage, all with separate filament supplies. That's what made me build 300b SETs - exactly half the work. But the results are worth it, so heck, the music is what counts in the end. I could have used SE inputs and drivers and done the phase splitting in the Interstage, but I have a DAC with balanced output, so that made me go diff pairs right through.
I agree completely with Jeff that I should try the 2a3s at 10 watts dissipation. I'll reduce the wattage as far as my speakers are comfortable - if I can save on tube life so much the better. I don't know what the comfort level of power is exactly with my speakers. I think from experience it's about 25 watts dissipation on a 300b SET but a PP amp may sound cleaner at that power. So 12.5 watts dissipation from each 2a3 should be OK at least.
I'm surprised you don't like the NOS 2a3s, but you've clearly listened a lot in A-B tests so you have good reasons for that. If the EMLs were less expensive.......
Andy
Edits: 07/11/12 07/11/12
Let's put a little finer point on things, Andy.(1) You can get JJ 2A3-40 in singles-- if you buy 4 or more-- for about $125 USD each.
(2) This tube is large and rugged. Run it at 12.5 watts total dissipation each. If using EML Mesh, run at 11.8, if running old AVVT Mesh, run at 10.5.
(3) NOS 2A3s. A FEW of NOS (old RCA Single-Plate, Sylvania Spring-Top Biplate) sound good on most music-- but not GREAT like EML Mesh.
The WORST of these old Biplates are RCAs-- which are just plain distorted. We used to toss 'em at trash cans. Also W.E. 300B tubes. I serviced many a theatre-- these always went into Dumpsters. The "M" and "W" shaped filaments (CHEAP! or it that CHEEPIE!?) that these, and most 45 tubes used made them sickenly euphonic and decidedly inaccurate music-wise.
The Studios all knew it-- not only did they trash the W.E. gear-- they got it OUT of their theatres, and put in ALTEC. That was a MONUMENTAL improvement!
Amps got better also. I actually ran some theatres from McIntosh KT-88) amps-- and did fine with them. Today, that's not so great with me-- but in those days, when you needed a reliable amp that sounded OK-- you could go with them, and run them for years. You STARTED with new capacitors, of course. The old electrolytics would be another trash-can item.
NOS Biplate 2A3s possess a kind of musical euphony-- but not as bad as one might think. These things were two 45 tubes sandwiched together, with appropriate filament, grid and other wiring changes.
This caused two things-- (1) they were not as transparent as the 45 because the tube now had a lot of junk in it-- we got more distortion and less transparency. (2) They LOST a lot of musical detail (compared to 45).... or a GOOD 2A3 (RCA Single-Plate at that time).
The problem was the same as with the 45-- the filament assembly was JUNK-- and still is. Had they built a welded, waterfall kind of filament structure like they had in their-- too expensive to produce-- Single-Plate RCA, the sandwich 2A3 and the 45 would have been accurate music reproducers. Well, they're NOT!
Today, the only output tubes made in the world that are actually built right are EML.
JJ 2A3-40 is an orphan-- also produced by the same people. The filament structure in the EML consists of vertically-hung, welded sections just like the old RCA Single Plate-- but better, and with far better modern materials. The filament structure, grid set, and plate all line-up with each other, and we get a uniform radiation pattern-- we don't have euphonic "hot" and "cold" spots messing with our music like the NOS tubes do!
The JJ is a well-engineered cheapie. The filament is long and strung-out just like a W.E. 300B, BUT it's not draped-in like a @##$%%Mart Toaster-- it doesn't form the DREADED "M" or "W" shape that RUINS MUSIC. Instead, they simply run the whole thing up and down and across the tube's top and bottom spacers, but they were smart enough to run some filament across both the top and bottom of the spacer plates-- and got---Tah Dah!!!---- VERTICAL filament strings. Now, the radiation pattern between this structure, and the plate is the SAME SHAPE-- and now the tube works! La, La! Music! Really!
What's wrong compared to the EML? Instead of 8 equal vertical filament sections welded onto bars at the top and bottom of the tube, we have only ONE long filament.
This ONE long filament means that one end of it is at 2.5 volts, the other end is at zero. All along the filament length, we have a different bias at any given point than at another given point. The radiation pattern is not totally uniform, but it's certainly better than mis-shaping the thing as well! The other thing is that the long filament can get looser easier-- but there are a few nice, thoughtful touches done to minimize that-- somewhat.
The tube is made from really good glass, is rugged, heavy and-- for the price-- a great performer.
Can it compete with PERFECT (almost!) filaments and a Mesh Window (EML)? Nope! But, then, it's only $125 bucks, and it lasts a long time, and sounds really good in a good amp.
What's not to like on the JJ 2A3-40? Nothing-- at the price.
---Dennis---
Edits: 07/11/12
You used to chuck 300Bs at trash cans ? Sure you did, when they were bad tubes; maybe...
RCA touching plate 2A3s, specified by significantly important companies like Brook and Northern Electric, are great tubes, used correctly. The fact that you do not like NOS USA triode opt tubes is simply your wish; nobody else's fact. Well, maybe Jeff believes that as well. I am not wishing to diss you or Jeff; you know that. Why do you say, repeatedly, that SE inherently has more clarity than PP ? Can you cite scientific facts that prove such ridiculous claims ? You might prefer, as do many on this forum, the sound of SET amps. But, to blanketly say or always imply SET is better than PP; that is nonsense. Do you really believe that you, or Jeff, are enjoying your systems any more than the rest of us commoners that do not have your amps ?
On the contrary, all of us are enjoying our systems; equally. That is why we hover as inmates, here and at other forums, some of the time. The rest of our free time we are hopefully listening to and ENJOYing our systems. I love some SE amps; to be sure. I sure do love my PP amps even more. In my systems, this is my preference, most of the time. Which flowery lingo sonic description term do you care to discuss that means SE is always better ? If SE is always better sounding to you, great. Enjoy your stuff. Please stop trying to make anyone else believe their systems, or even tube brand choices, are inadequate; compared to your choices. Can't we all just get along and enjoy listening; without any denouncing ?
.....I love some SE amps; to be sure. I sure do love my PP amps even more. In my systems, this is my preference, most of the time.> >
I think more than a few of us feel this way. Many of us must have built a stack of both. I don't think there's anything mystical about SETs, though I know many do think that way!
I'm listening this morning to the Brandenburgs by Il Giardino Armonico on my PP 2a3 all-DHT amp, and I'm as deeply satisfied as any SET user - which, by the way, includes me!
Andy
I like that! ---Dennis---
No denouncing! That's not the idea, please! It's just an objective discussion of differences that exist-- choose what you like!S.E. amps have no natural Common-Mode (hum, noise & distortion) rejection.
This factor is why they must be built better, with better parts-- if they are to become anywhere near as good as a cheaper to build P/P.Superior is when you develop either type to best address the speaker you're driving with the amp.
There are, of course, differences that one should be aware of. S.E. amps tend to tune into narrow bandwidths that tend to center around the values of capacitors that are used for cathode bias and plate supplies. S.E. amps are extremely sensitive to these values. The really great S.E. amp requires-- what some may regard as extreme measures-- virtually "perfect" engineering across a wide spectrum of the amp's design and construction.
If all of the necessary challenges are met, then the S.E. is more musically revealing and sensitive to musician attitudes and emotions-- and is a better replicator of musical depth and layering-- than any other form of amp, especially if the S.E. is simple-- uses only two stages Directly-Coupled together.
So what, you could say? And I might agree-- depends....
The P/P amp has natural Common-Mode rejection-- it eliminates a lot of hum, noise and distortion automatically. It is pretty self-evident that there are artifacts that are a part of (dynamically occurring) music-- that will look the same as hum, noise, or distortion-- to the circuit-- which, being Push/Pull, will naturally reject some of it. ALL Push-Pull circuits do this.
Common-Mode rejection is a form of processing-- it PROCESSES OUT some hum, noise, distortion AND MUSIC.
A Push/Pull amp is far easier to design for a wide bandwidth as each side of the circuit tends to augment the other's bandwidth-- expand it.
Output transformers for Push/Pull circuits tend to self-cancel any tendency for the laminated core to "saturate"-- allowing easy design for wide bandwidth and power in the output trans.
Basically, the best-designed S.E. amp will be under 2 watts or so, Push/Pulls can be anything you want to build to-- in power levels.
To choose one topology over the other isn't the point here-- the point is that tube quality requirements are MUCH more stringent for the S.E. type.
So, why do we listen to S.E. amps at all? THEY DO NOT PROCESS MUSIC-- they can't unless additional circuitry is designed into them to force it. Of course, we're not going to make that mistake with S.E.-- the simpler and more natural it's built to be-- the better it is.
What to like? I use both types of amps routinely. But this is the SET forum!
---Dennis---
Edits: 07/11/12
"It is pretty self-evident that there are artifacts that are a part of (dynamically occurring) music-- that will look the same as hum, noise, or distortion-- to the circuit-- which, being Push/Pull, will naturally reject some of it. ALL Push-Pull circuits do this."
Care to explain this?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Thanks for posting that. Well said !!
Jeff Medwin
SET:...more musically revealing...more sensitive to musician's attitudes and emotions...better replicator of musical depth and layering...
PP: CMR Common Mode Rejection in push-pull amp naturally rejects artifacts of music that look like hum noise or distortion...PP processes the music...
Does anyone really think all of this is gospel ? Jeff, my old friend, do you think this is all true fact ? Come on, now, let us decipher some of this; shall we ?
The opinions about more musically revealing, more sensitive to emotion and attitude plus a better replicator of musical depth and layering, sound just like my PP Triode amps using an IT and NFB (no neg fdbk). In fact, my PP amps have more dynamic contrast (thus dynamic range), better depth and layering(much more involving 3D EFX described as holographic), plus they are clearly a better conveyor of the speed and the action of each instrument within any musical selection; compared to many SE amps used on my same systems. Of course, this is opinion. In yours or Dennis's system, your's or Dennis's opinion is what matters.
There is no inherently better topology, nor inherently better tube vs. solid state. System synergy can create fantastic sound in anyone's home. Despite this being the SET forum, there is no SEPentode, triodes, tubes vs. transistors or PP forum. So, this forum is where we often meet and greet.
I have no doubt that your topology and Dennis's topologies have many virtues, sonically. Even spec wise; though we all know many of the specs are inadequate, some of your specs could be admirable. Leaving out any discussion of CMR, a truly theoretical mumbo jumbo description originally intended to describe power supply rejection, how about distortion ? Specs being inadequate, to be sure, it is said the SE amp has higher even order distortion than PP. Okay; and this higher magnitude even order distortion came to be known as euphonic, compared to the lower odd order distortion in PP ? Euphonic; what a pleasant term. What a pleasant effect. Is it any wonder we all strive for euphonious playback ? Even if we prefer other terms, like accurate and detailed, better replicator, more musically revealing, even more flowery words, don't we all strive for musicality and harmonious music reproduced in our homes ? Yes; we all strive for euphonic playback.
Regardless of what type of distortions or the magnitudes of these distortions, none of us usually bring any distortion meters, signal generators or spectrum analyzers when we evaluate systems for the purposes of long-term listenability. We rely on our ears, most of the time; as it should be. Most importantly, any topology, even including ss amps, can create a very musical sound when synergistically chosen within a system approach. There is no reason, for any of us, to drill into anyone else's minds, that their particular approach makes any and all others not as good. Let us all enjoy the music...
Hi,
Thanks for informative subjective opinions (to all), it helps me decide which way my investigations go. And on that, nice to know that people are actually building stuff.
Andy, I can also recommend that you try the JJ 2A3 tubes, if you're concerned about matching, I can recommend Bob at Eurotubes.com, I bought a quad 5 years ago and they still all bias up the same, and they were also the most cost effective vendor, even half way round the globe.
PP vs SE, I suspect theres alot more to it that meets the average load. I did and do like 71A SE to PP 2A3 grids into 6k6p-p. Cracker-Jack. But you need a good IT. Lundahl doesn't, Bud Purvine does and no doubt theres a suitable UTC unit that fits more than well.
Foregoing the IT and the splitter and running direct coupled to the SE OP stage was better again, clearly.. but I can appreciate other could be preferred.
I also prefer Garfish to Whiting, others will disagree.. but disagree with what exactly?. Its just a preference, and thats all it is.
Liked what Dennis had to say about bias shift with Vf potential, yes. Could be lessened by some extent, at least to my thinking, by using a CT AC supply -1.25vAC - 0 - +1.25vAC.. who really wants a Rk or Ck anyway?.
At any rate, I'm sure most would agree that its nice to be interested in something that can reward you with satisfaction. And I think that less people can attest to the same as time goes by.
Hey, Thanks.
Shane
Your sense of humor and common sense is refreshing-- and fun!
One thing you might understand-- in my own case-- I've been in audio for a long, long time.
During that time, I've always had access to the best audio amps, preamps, wiring and speakers-- for at least the last 50 years.
I was into Ham Radio, built all my own equipment, ran commercial broadcast stations, designed , serviced, and built movie theatres across half the USA, and worked in L.A. recording venues..
In the early days, the first thing one did to an amp was disable the NFB. Next thing-- in those days-- was better signal handling caps. That included speaker crossovers which normally used Band-Aid packages for caps in passive crossovers. One could get monster improvements in the ALTEC theatre speakers-- for example-- by calling up Rel-Cap, (right in town!) and getting some film caps that were actually designed for speakers.
We had all kinds of amps to play with, and had the money to buy more.
It quickly became obvious that "pro" amps and "Hi-Fi" amps were usually sort of opposites. All were Push-Pull (except for the W.E. Model 91)-- some worked and some didn't. They all had to have major work to sound OK.
In the 60's, finally-- parts that were fairly good had become available to the population-- not just to "pro" users. Many people took available amps as far as they could go-- examples were Dynaco, Citation, McIntosh, Marantz, ALTEC (tube), and a whole pile of Scott, Fisher, and whatever else you want to remember.
NONE of those amps EVER performed well enough to convince a listener-- on any speaker-- no matter how good-- no matter how well modified the amp was-- not one could convince me, or others that we were listening to anything other than equipment that was struggling.
The culprit was amp power supplies-- designed to reduce hum-- and for nothing else useful at all. Those things guaranteed current starvation of all driving
elements, and even screwed-up operation of the output stages-- to some extent. Regardless of power rating, ALL of those amps had NO POWER.
That wasn't all! They also screwed-up Rhythm, Pace, and Timing-- big time.
Today, a few people still try to fix-up this stuff. The results are often far better than the originals, but they still fall very short of actually reproducing music.
A few of us got busy and studied all aspects of amp design, and we first threw-out everything that was not needed. The result was two-stage direct coupling without NFB. The next thing was to assure adequate current availability to the driver stage-- regardless of operating condition.
The next thing was to stop "snubbing" voltage-driven dynamics-- a practice that was robbing music of its "wow" and "fun-factor".. We soon learned to make circuits VERY High-Impedance, put minimal loading on them, but provide them with LOW Impedance POWER.
The rest was implementation-- good chassis design, floating parts in 3-dimensions, using really heavy, sonically excellent conductors, using the best parts, floating iron parts on brass. Using parts of the chassis as isolation platforms. Not much was overlooked!
Today, the results are spectacularly good at reproducing what's been recorded by the best Studios.
That was what we were after the whole time.
---Dennis---
...A few of us got busy and studied all aspects of amp design, and we first threw-out everything that was not needed. The result was........> >
Hi Dennis,
This sounds a very, very familiar scenario and probably one we all went through. The thing is there's a LOT of ways of finishing the sentence! My own would be:
"... putting all indirectly heated tubes back in drawers or on ebay and only using DHTs right through, only using polypropylene caps in power supplies, avoiding all caps in the signal path and as cathode bypasses (except for the output stage), using filament bias, direct coupling where possible, only using transformers or plate chokes on plates of tubes......... and so on
Andy
Nice, Andy!
---Dennis---
Reputed to sound as good as 2a3.
To infinity and beyond!!!
At 10 to 10.5 watts dissipation.Jeff Medwin
Edits: 07/10/12
I have a 6B4G SET and I now have 4 sets of NOS tubes: Sylvania, Tung-Sol and Raytheon (which are probably Sylvania-made though) 6B4G, along with unbranded 6C4C.
They're all better than the unbranded 6C4C. In my amp they all sound virtually indistinguishable but my amp is fairly inexpensive (e.g., Edcor OPT).
The advantage of 6B4G is cheaper NOS, which are still plentiful. The disadvantage is the difficuly in keeping the noise level down. Swings and roundabouts.
Heck, if you're interested, I'd consider selling you one of my NOS sets. Let me know.
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
