Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › The How and Why of Tubes
New Posts  All Forums:Forum Nav:

The How and Why of Tubes

post #1 of 14
Thread Starter 

I've been enjoying anew listening to my favorite music now that I've gotten a lightly used pair of Sennheiser HD 600's hooked up to a Little Dot MKIII. I recently took the dive into tube rolling and purchased a pair of Voshkod 6ZH1P-EV as they came highly recommended in Dept_of_Alchemy's post on tube rolling for the MKIII. I really like the results. Penchum's review of the MKIII mentions the Voshkods and then goes on to say the GB-5654 Sylvania Gold Brands are the best EF95's he's used. I realize there's a law of diminishing returns going on here and that because I paid 'x' amount for my Voshkods doesn't mean paying '3x' for the GB-5654's will give me sound that's three times better. Clearly that's not even really possible. I have realized however I really lack some basic understanding concerning the equipment and I really should educate myself so as not to be the blundering idiot who pays a substantial amount for something just because it's being offered for a substantial amount - I may well end up disappointed. With that in mind, some questions (some rather stupid):

 

 

 

- What is the best layman's description for the how and why of the circuitry of tube amps? No need to get too detailed, just start with the input signal and proceed with each major component. Again, layman's.

 

- What are all the contacts (metal legs sticking out the bottom) for? Does each translate into a section of the sound spectrum? Obviously we're dealing with current here but I'm wondering how it all works.

 

- Are the power tubes that important? Or I guess a better question is do they have that big an influence other than gain and they'll mess everything up if not well matched?

 

- Why does each model of tubes seem to have three different model numbers to it?

 

- How long does a set of driver (signal) tubes last when taken care of?

 

- I've got a Little Dot MKIII situated in a bit of “cubby” shelving which is against a wall. Basically there's not a whole lot of air flow around the amp when it's turned on. Will this affect the sound and/or longevity of the equipment (should I have it more out in the open)?

 

 

 

My musical tastes are all over the map: classical, metal, techno, rock, some jazz, even books on CD. I do appreciate Head-Fi and the candid community here; even if I rarely post. There's always good advice and as I said, I'm really enjoying the results of my purchases.   

post #2 of 14

There's a lot to cover here! I'll do my best to get things started…

- What are all the contacts (metal legs sticking out the bottom) for? Does each translate into a section of the sound spectrum? Obviously we're dealing with current here but I'm wondering how it all works.

 

We probably need to get this one out of the way first. The basics of how a tube works. A diode is a device that only allows electricity to run one way. The first tubes were diodes. Heating an element inside of a vacuum draws the electricity away from the heated element. Di- is two — your anode and your cathode, your two directions. Now, to way oversimplify this (as requested), the tubes we use for amplifying audio are generally triodes and pentodes. Let's just worry about triodes, adding one more element. This element is the control grid — changing the voltage applied to this grid changes the flow of electrons from the heater to the non-heated element (the plate). This is the very heart of amplification. So, all the pins? Let's look at the pinout of a 300B — 4 pins: two for the heating/operating voltage of the heated (anode) element, one for the plate (cathode), and then the control voltage to the grid. 

 

- What is the best layman's description for the how and why of the circuitry of tube amps? No need to get too detailed, just start with the input signal and proceed with each major component. Again, layman's.

 

This is the trickiest, most unanswerable one, simply because there are so many designs (topologies) for amplifiers, be they tubed or solid state. It's like asking 'how do I cook a meal?' So I'll try to give you the simplest one: the single-ended triode. Back to our 300B — we need a power supply. That'll heat up the tube, and give us our control voltage. Then you need an output transformer. Transformers allow us to turn a high voltage / low current into a low voltage / high current, or vice versa. In this case, we need to do the former. There's another thing going on here, and that's the removal of DC. Our speakers want AC; DC will murder them. We can complicate things slightly and use capacitors to ditch the DC, in order to use a different sort of transformer… but again, we can complicate things in a lot of ways — there are so many amplifier designs out there! Ultimately, however, your tube is doing the amplifying, and the whole package can be pretty simple.

 

- Are the power tubes that important? Or I guess a better question is do they have that big an influence other than gain and they'll mess everything up if not well matched?

 

The power tubes are doing the amplifying here, the heavy lifting, so… yes! Improper matching matters more in push-pull designs, anything where phase inverting is involved… less so in a more straightforward design where properly biasing will match channels… at least this is my understanding, less confident on this…

 

- Why does each model of tubes seem to have three different model numbers to it?

 

Just different standards (US, Europe, manufacturer-based)… Considering a lot of electronics work was for the military, etc., standardizing on a more regional level at that time was… inevitable, I suppose. Also, some tubes are actually different — different tolerances, slightly different heater voltage, designed for different frequencies, etc… — but maintain the same pinouts as other tubes, and are close enough to be compatible.

 

- How long does a set of driver (signal) tubes last when taken care of?

 

Few years? This, too, is a hard call. No matter how much you are taking care of them, it's all still dependent on how hard the amp is driving them, etc. This is why they don't last as long in guitar amps, where driving them to the point of distortion is the norm…

 

- I've got a Little Dot MKIII situated in a bit of “cubby” shelving which is against a wall. Basically there's not a whole lot of air flow around the amp when it's turned on. Will this affect the sound and/or longevity of the equipment (should I have it more out in the open)?

 

Heat isn't great for any electronics, and tubes are going to generate heat because… well… they have heaters in them! It may be fine, in fact I'll go out on a limb and say that unless it feels ungodly hot, it's probably fine, but generally speaking… more air circulation can never hurt. I would hazard it would affect longevity more than sound.

 

There are probably some inaccuracies in here — I'm tired, blah blah, but… hopefully this is a starting point.

post #3 of 14
Quote:
Originally Posted by cdelucia View Post

 

- What is the best layman's description for the how and why of the circuitry of tube amps? No need to get too detailed, just start with the input signal and proceed with each major component. Again, layman's.

Maybe a bit of details on how the amplification works. First, an amplifier doesn't really amplify a signal, it MODUALTES a bigger signal (the input and the output never "touches" each other in a no-feedback design). You apply a voltage between a cathode and an anode. This voltage is the absolute maximum of what the amp will output. This voltage comes from the amp's power supply, from your wall. However, there is no current running betweem them, because there is something blocking the way.... a gate (or control grid here)! This gate by default is blocked.

 

However, you can control the flow (like water through a pipe with a valve blocking) between the cathode and the anode with the gate by applying a voltage to it. Here, we'll apply music to it. Let say the music maximum voltage is + or - 2 V and the cathode/anode voltage is 15 V. When the music hits 2 V, the grid is all the way "open" and you'll get your full 15 V output. When the music hits back 0 V, the gates will close, and you'll get 0 V at the output. When the music will hit -2 V the other way around, you'll open the gate the other way and get -15 V at the output. When the music hits 1 V, the gate's halfway open, and you'll get 7.5V and so on. If you followed me correctly, you'll notice the output mimics the music voltage but on a bigger scale, thus effectively amplifying the input signal!


Edited by Rem0o - 11/1/14 at 5:29pm
post #4 of 14
Nice explanation Rem0o.

Do tube amps - or amps in general - suffer noticeable non-linearity in practice? Something like your valve following an exponential transfer curve? If so, I could see two obvious cases where the amp might exaggerate small differences at low levels (exponent < 1) or at high levels (exponent >1)...

I ask because since getting into tube amps I noticed what seemed more "vividness" or "life" compared to my SS amps. At first I thought something to do with subtle FR tailoring (my tube amps 'sound' neutral and I prefer and tube them that way, so any FR effects would have to be subtle).

But later it seemed to me these tube amps 'exaggerate' contrasts in level at certain ranges. For example, tiny vocal or instrumental changes at low or moderate levels are much more obvious than with the SS amps. The net result is much more expressive and nuanced - better micro dynamics I guess.

If true, I guess there's compromise elsewhere in the range - what you gain in micro dynamics might be lost in more compressed macro-dynamics? Or vice-versa?

Or this might be complete rubbish biggrin.gif
post #5 of 14
Quote:
Originally Posted by AiDee View Post

Nice explanation Rem0o.

Do tube amps - or amps in general - suffer noticeable non-linearity in practice? Something like your valve following an exponential transfer curve? If so, I could see two obvious cases where the amp might exaggerate small differences at low levels (exponent < 1) or at high levels (exponent >1)...
Not really. Class A and class AB are meant to counter that problem. With tube headphone amps, we're talking class A   Class A avoid the curvy part of the transfer curve by applying a high bias current to oscillate around the nice linear part of the curve.
I ask because since getting into tube amps I noticed what seemed more "vividness" or "life" compared to my SS amps. At first I thought something to do with subtle FR tailoring (my tube amps 'sound' neutral and I prefer and tube them that way, so any FR effects would have to be subtle).
2 things are known for tube amps to affect sound: Higher second order distortion and high output impedance. The latter can give a boost (1 dB or 2) around the purely resistive frequency of your dynamic headphones (when phase = 0 deg). Most of the time, it boosts and bloats the bass just a little bit, giving a more "analog" sound.
But later it seemed to me these tube amps 'exaggerate' contrasts in level at certain ranges. For example, tiny vocal or instrumental changes at low or moderate levels are much more obvious than with the SS amps. The net result is much more expressive and nuanced - better micro dynamics I guess.

If true, I guess there's compromise elsewhere in the range - what you gain in micro dynamics might be lost in more compressed macro-dynamics? Or vice-versa?

Or this might be complete rubbish biggrin.gif

For me yes lol
post #6 of 14
Thanks for the reply. Yep, I'm aware of higher Zout and 2nd order THD. No audible bass bloat nor do I chase "analog" sound (whatever that is!).

I wonder about the significance of (modest levels of) 2H THD with 'natural' instruments that themselves have various amounts of 2H as part of their timbral profile. That is, would the 'tubey distortion' fool the listener into hearing a different instrument altogether? Or is it more a matter of a Bösendorfer sounding like a Steinway eek.gif

Whatever, I'm sure I'm provoking you into a further attack of laughter. Out of concern for your well-being I'll stop now.
post #7 of 14
Quote:
Originally Posted by AiDee View Post

Thanks for the reply. Yep, I'm aware of higher Zout and 2nd order THD. No audible bass bloat nor do I chase "analog" sound (whatever that is!).

I wonder about the significance of (modest levels of) 2H THD with 'natural' instruments that themselves have various amounts of 2H as part of their timbral profile. That is, would the 'tubey distortion' fool the listener into hearing a different instrument altogether? Or is it more a matter of a Bösendorfer sounding like a Steinway eek.gif

Whatever, I'm sure I'm provoking you into a further attack of laughter. Out of concern for your well-being I'll stop now.

You're searching way to far mate. As you saw, I'm kind of a tech guy, objective for that matter. But I do own a tube amp, and I do listen to vinyl. Why? Because I enjoy it and it sounds good to me. That's all what matters at the end of the day.

post #8 of 14
Yep that's why I asked the questions - your tech background and lucid explanations in one or two posts here and there.

I'm "searching" because I'm curious. Physics degree way back but social scientist/statistician these days hence the curiosity!

But agreed: "sounds good" is what matters. Thanks for the further reply.
post #9 of 14

All tubes, every single one of them have a transfer curve that is called a "S" curve.

Some may be a bit flatter, some may have the curve moved up or down on the

transfer curve, but all exhibit this non-linearity. The trick is to pick the part

of the transfer curve where the tube is the most linear and run the tube over

this limited range. Not as easy as you might think. Run completely open loop

with no feedback, THD in the 1 to 3% range.

 

No different with solid state devices.

post #10 of 14
Very interesting Kevin. S-curve - that's the sort of thing I was thinking of.

So in practice do you mean one might get into "the most linear" part of the curve but have a little (audible) non-linearity at one end or the other?
post #11 of 14

here is a good explanation

 

http://www.john-a-harper.com/tubes201/

post #12 of 14
Thanks Kevin beerchug.gif
post #13 of 14

Very interessant thread !!

post #14 of 14
The link Kevin posted is very worthwhile
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Headphone Amps (full-size)
Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › The How and Why of Tubes