|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
209.137.226.131
In Reply to: RE: OT: Thorsten, 2A3s, 45s and 300Bs. posted by RC Daniel on July 14, 2012 at 18:04:06
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---
Follow Ups:
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'.
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: