How does a tube amp work?

What does that even mean?
Who decides what power is needed to allow speakers to "run properly"?

The speaker manufacturer decides this. But Gordon's helpful response to this query in post #107 explains why a 25W tube amp might work as well as a 100W SS amp for any particular speaker. His points are that tube amps can run closer to their max output before clipping, and a 3dB volume difference does not sound much louder. 'So, if you have a 25 watt amp which you can "use" to 24 watts, vs. a 100w amp that you can only "use" up to 50- that's only 3dB difference. Not that much perceived volume difference at all...'
.
 
But you'll still use SS stuff up front because it still has a better variation on tonal differences and easier to manipulate without changing everything in a tube preamp.
 
The speaker manufacturer decides this. But Gordon's helpful response to this query in post #107 explains why a 25W tube amp might work as well as a 100W SS amp for any particular speaker. His points are that tube amps can run closer to their max output before clipping, and a 3dB volume difference does not sound much louder. 'So, if you have a 25 watt amp which you can "use" to 24 watts, vs. a 100w amp that you can only "use" up to 50- that's only 3dB difference. Not that much perceived volume difference at all...'
.
That begs this question. Since I only have 10 watts at most.
What % of power can I run my 10 watt tube unit at? All day all night. 24/7.
90%? as long as it sounds good?
No I could never run a 50 watt SS unit like that could I?
 

Attachments

  • P2210042.JPG
    P2210042.JPG
    47.1 KB · Views: 30
That begs this question. Since I only have 10 watts at most.
What % of power can I run my 10 watt tube unit at? All day all night. 24/7.
90%? as long as it sounds good?
No I could never run a 50 watt SS unit like that could I?
In the tube era, PA amps, radios and TVs, were all designed to run for long periods of time without difficulty. So, you can run your amp 24/7. Whether you want to run it just below clipping volume is another matter, unless you are running a dance club.

I would never leave a tube amp on unattended, even if it had been completely rebuilt, for eco and safety reasons.
 
I only read the first page so I don't know if it's been pointed out. When it comes to explaining electronics, it's often explained via electronic engineering terminology that makes it less simply explained/just as or more confusing than before the question is asked. But maybe there's not a better way to do it and so it comes down to looking up the terms.
 
+1 on Mono FM. In fact, I could not really get rid of the terrible noise in stereo FM on my Harman Kardon F-500X. It is still on my bench, I am waiting for a pair of RCA adapters to fix the broken RCA plate. But I think I am going to live with the mono FM, it is actually surprisingly clean and well-sounded. There are people upgrade the FM session for a modern FM chip. But I think it kind of defeat the purpose of keeping the vintage gear. That's just me. Yes I do listen to the radio despite the fact that I have nice set up with turntables, R2R tapes etc.
 
The operative word is "filter". they are filter caps that filter AC variations/pulsations in the dc to be sent to ground, thus filtering/smoothing the DC, but letting it to pass on without much affect..The freq they need to filter is usually 120hz. A piece of kake for some large value cap say 40- 20000 uF or more, as needed. All depends on how much storage you need and how much current you're running.

Electronics 101... DC power supplies

I like the fact we are bringing frequency of signal into the consideration here as well as the basic analogy; caps block DC and pass ac.

But with filter/power supply caps/electrolytics, I like to consider these caps to be like batteries, bringing DC power to where it needs to go, albeit temporary; only there when the power is on.

Going back to electronics 101, keep in mind voltage is measured across two points, so our Rail voltage might be something like +44 volts dc and -44 volts dc (sorry I'm using a more solid state example here). Then of course, current is measured through a point or node, yada yada.
 
Hey AKers,

The pictures in the original posts got deleted during an AK forum software upgrade that happened years ago. Recently put all the info from the original posts into a document with the pictures restored. You will find the PDF attached to this post.
 

Attachments

  • How-a-tube-amp-works.pdf
    1,009.4 KB · Views: 35
Great thread - and thank you for making the PDF. The one thing I have never been able to figure out, is how do you specify the output transformer? How do you calculate the turn ratio for the various final output impedances (like 4, 8 or 16)? It's just a step-down transformer, isn't it?

Thank you.

-Geoff
 
What about something like this headphone amp that uses just a single dual triod tube? I'm using an EH- 6922 tube with it.
How exactly does something like this work ?

With this, I like the sound with my Grado HPs better than with the HP section on my NAD amp.
BTW, that tube is just a bit bass heavy so a little cut on the bass with the amp's passive 3 band EQ takes care of that .

The pic is not my amp but a pic from the net .
 

Attachments

  • ]brovo_3 HPSxx.JPG
    ]brovo_3 HPSxx.JPG
    55.8 KB · Views: 20
Last edited:
That is a hybrid amp it looks like that uses transistors for the amp section.

What about something like this headphone amp that uses just a single dual triod tube? I'm using an EH- 6922 tube with it.
How exactly does something like this work ?

With this, I like the sound with my Grado HPs better than with the HP section on my NAD amp.
BTW, that tube is just a bit bass heavy so a little cut on the bass with the amp's passive 3 band EQ takes care of that .

The pic is not my amp but a pic from the net .
 
Suggest you post your transformer question in the main Tube forum to get more responses.

Great thread - and thank you for making the PDF. The one thing I have never been able to figure out, is how do you specify the output transformer? How do you calculate the turn ratio for the various final output impedances (like 4, 8 or 16)? It's just a step-down transformer, isn't it?

Thank you.

-Geoff
 
That is a hybrid amp it looks like that uses transistors for the amp section.

Yes, it is indeed a hybrid .....But, what what exactly is the tube for in this HP amp as it is only just a pre amp and can even be used as one ?
 
Yes, it is indeed a hybrid .....But, what what exactly is the tube for in this HP amp as it is only just a pre amp and can even be used as one ?

Before responding, I have not seen a schematic of this amplifier, so this is a generalization...

Typically, tube hybrid designs utilize a preamp tube on the front end for voltage gain and a transistor follower on the output for current gain. Functionally, the tube increases the voltage of the input signal, increasing volume, while the transistor follower increases current, thus lowering the output impedance for optimal damping. The transistor follower operates at "unity gain", meaning there is no increase in voltage / volume, it has a gain of 1. A circuit like this can usually function well as a preamp. Without the transistor follower, the output impedance of the circuit will be highly dependent on the internal resistance of the preamp tube, which can be very high depending on the type, not ideal for damping a headphone or proper impedance matching for a power amplifier if used as a preamp.
 
Slightly off topic, but there are a few ways to hybridize a tube stage. I like to active load the triodes often used in preamps. That way you can force a good sound out of the tube without it being NOS or some old esoteric brand name.
 
Slightly off topic, but there are a few ways to hybridize a tube stage. I like to active load the triodes often used in preamps. That way you can force a good sound out of the tube without it being NOS or some old esoteric brand name.

I use active loads often in my designs. Another advantage of doing this is lowering the output impedance of the stage. For example, if loading the tube with a MOSFET-based constant current source, the output can be taken from the source of the FET as opposed to the tube plate. Technically this puts the transistor "in the signal path", but often has a positive impact on the sound and measured performance. I have a tube output DAC build in process - in this example, the output impedance of the DAC was lowered from 1.2K (essentially the plate resistance of the output tube) to 200ohms by taking the output from the FET source. The bandwidth and noise floor of the DAC improved as well. I have done something similar in my phono stage.

Transistors can drastically improve the performance of tube circuits if used well!
 
Transistors can drastically improve the performance of tube circuits if used well!
This is very true. I stocked up DN2540 DMOS when they were available for this purpose. I like those on the plate. For differential circuits, I normally go with a B- of 30V or so and achieve with two BJTs what it would take hundreds of volts to do with a tube.
 
Tubes work by real, actual "magic".

The tubes direct and modulate streams of electrons. Except electrons really aren't anything. As far as physicists can tell, and they've been studying electrons for 120 years or so, electrons HAVE NO EXTENT. Not just very small. Not even microscopic. NO attempt at measurement has found an actual discernible size to the electron. And due to quantum rules, you can't really say they are here or there, you can at best come up with a rough probability of where one may be. Even worse, the detailed electron equations only work for a single electron circling, or anyway, near a single proton.

So tubes use things that have no size. In my book, that means they really don't exist. Except they do eventually hit a plate and make a click. So something goes on between cathode and plate but not anything you can see or even fully theorize or explain.

And, ummm, if that isn't weird enough, while they have no size, they have something akin to "spin". Worse yet, they only have 1/2 a unit of spin. You effectively have to spin an electron 720 degrees to get it back to its original state. Not speculation, 1/2 spin is the only amount that fits into dozens of electron equations. See:

Even worse, you can get a rainbow or interference pattern from an electron stream, so in some sense they are not particles but waves. Again, waves that come in discrete but invisible bundles. Weird enough?

Sure, you have your Spangelburg books with equations that help model tube behavior at a high level, but deep down things get really crazy.

So we are fully justified in calling it pure magic and enjoying our hot little friends, the tubes.
 
Last edited:
Back
Top Bottom