• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

warmth in different tube types

I read an old post of Thorsten’s the other day comparing the sound of a 300B and a 2A3, to the temperature of light.

He preferred the 300B as being a little warm, while (watts aside) the 2A3 was like white light, revealing more, but could be a bit too hard on not so good recordings.

Anyone tube owners care to comment on how much *warmth of sound* different tubes have?
ALthough this is highly subjective subject.....
One statement that should be well agreed upon is that a tube is not 100% LINEAR ...
That being said.... A tube has a transfer characteristic that is not LINEAR.... This curvature is different for different tubes and bias points... The curve can be represented as a polynomial of high order.... This produces HARMONICS ...these harmonics when even ordered tend to "thicken" up the waveform when they are summed ...this is what I refer to as warmth...
A little warmth is good to my ears..but not too much....some designers are more anal and seek most LINEAR as possible..
This is a very simplistic explanantion... phase angle, intermods..ect... all play a role...
Feedback is a means of LINEARIZING this signal transfer.....but then again..strip away the harmonics and you get more sterile sound.....experience working with many tubes will become part of memory, how they sound....This is nearly impossible with equations...but not impossible...
Hi-End audio as a hobby is all about making gear that apeals to YOU.... Finding that balance between LINEARITY, warmth and "air"..... Where you find your "sweet spot" is what makes you a happy listener one day but may change over the years....

I’m not suggesting eliminate distortion, but if we know the relative distortion/ colour of different tubes, we may have a slightly better chance of tuning our system.

By comparison in the SS world, tube loving designer Hugh Dean (in brief) learnt the distortion/ colour of different components, and tuned his Aksa amp to a desired sound, which many like.
I believe the distortion introduced in the tube makes the music more exotic to our ears.

What kind of distortion do you mean? The warmer sounding tubes often have lower distortion than colder sounding tubes. The warm sound is rather the distribution of the distortion spectrum i.e. low odd order distortion...or more specifically low 3rd harmonic distortion...

IMO you simply can't say that it's the distortion that makes the tube sounds warm...it's the lack of distortion in terms of how the human ear perceives distortion.

But these are just blanket statements..no one gains anything from your or my statement...I heard a system with a solid state amp that put my SE amp in the shade in quite a few areas. One can design a good amp with anything really..it's just easier with tubes.

Back to the original question : I'd say my 6C41C is pretty neutral in terms of warmth...very hard to put it into some kind of category.

300B's can't be put into one camp either. I find the KR's colder than other 300b's
Anyone tube owners care to comment on how much *warmth of sound* different tubes have?

Here, you are asking an impossible question. There is no definitive answer. According to Max Robinson, the 6AS7G is not good at all:
The results were more than a little disappointing. I could get 12.5 watts out of it but at 6.5 percent distortion. Even when the power was lowered to 5 watts the distortion was still 4 percent. When I connected the feedback I couldn't get the distortion below 1 percent. Oh, well, these tubes were not designed for audio service.

However, These Folks refuse to use anything else, and can't say enough good things about the 6AS7G. Furthermore, they also bash mercilessly the 300B as an audio tube. Even though they have a prototype that uses 300B's, they consider that the performance is not up to their standards and refuse to put it into production despite the demand for it. You can find other web sites where the superiority of the 300B is proclaimed in no uncertain terms. Nor are we talking about fly-by-night-ers who may not know what they're talking about, or promoters who have a vested interest in selling lots of these tubes to unsuspecting customers. As for what makes a "good" tube, there is also the question of circuit topology and component quality. Add all those considerations together, and you wind up with quite a mess.

About the closest thing to a consensus I could ever find is that I have yet to see anyone saying that the 2A3 doesn't make a good final.

In the final analysis, what it comes down to is this: does the amp sound good to you, or does it not?
Gingertubes views/opinion/pure speculation on harmonic distortion.

1st any curvature of the transfer function will introduce harmonic distortion. The more sudden the curve the higher the order of the harmonics in the distortion spectrum. Thats why we all like tube amps which soft limit rather than SS Amps which clip - right!!!.

Some Music Maths (rounded to 4 decimal places):
Start with a Bottom C - 65.4064 Hz
2nd harmonic = 130.8128 Hz = C 1 octave up, large amounts of 2H are benign sonically
3rd harmonic = 196.2192 Hz, this is 0.2 Hz away from a G, 1 octave up
A C Chord is made up of C, E and G notes. a reasonable amount of 3H is just going to make the sound bigger

4th harmonic - 261.6256Hz - a C 2 octaves up
5th harmonic = 327.03196Hz , 2.5 Hz away from an E, you don't want much of 5H or it starts to sounds like a badly tuned instrument.
6th harmonic = 392.4383 Hz, 0.4 Hz away from a G, not a serious problem in small amounts
The first seriously awfull one:
7th harmonic = 457.8447 Half way between an A (440) and an A# (466.16), Musically "quint", no musical relationship to the fundamental what so ever. Tiny amounts of 7H sound awfull.
8th - another C - benign
9th - 588.6575 Hz, 1.3 Hz away from a D, not musically related to the fundamental - not as bad as the 7th harmonic BUT not good
10th - 0.5Hz away from an E - not too bad
11th - Half way between an F and an F# - Seriously awfull.

I could go on for ever BUT you get the drift - my view is that so long as each higher order harmonic distortion product is less than the preceeding one and that by the time you get to the 7th harmonic the level absolutely miniscule the sound will be lovely. Some folk say that you don't want any 5th harmonic or above and you can see from the above why this might be so. The 5H product is just a little too far from a musically related frequency BUT the 7H is not musically related at all!!

The more feedback round the amp the harder it will clip rather than soft limit and the higher the order of the harmonic distortion products produced.

Thats why 1 watt from a moderate feedback tube amp is as good as 2 or 3 Solid State Watts and why 1 tube watt from a NO Feedback tube amp is a good as 10 SS Watts.
The SS AMP has to be able to deliver the peaks without clipping.

There is no definitive answer on subjective questions, but I was hoping for a modest consensus. IMO because some have different preferences, doesn’t mean that it’s impossible to get a majority view, that could be a useful guide. If 100 people bash the 300B, they are not wrong but of opposite to I would think to the majority.

“I have yet to see anyone saying that the 2A3 doesn't make a good final” – thank you

“what it comes down to is this: does the amp sound good to you, or does it not?”

The point of the exercise was to share experiences, so that the next amp project was a little less likely to be an experiment, and a little more likely to be a sound people wanted.

Tricky subject....despite all the tube amp harmonics, I 've quite (non)impressed to the differering quality of loudspeakers.....to me that dispels the harmonic thd arguement of tube amps.
My tube amp is capable of <0.05% (thd 3rd har) at 150W 1Khz....swapping to other LS makes far more impact on sound quality....I suppose it all depends to a particular system one gets used to ?

I don't know if this has anything to do with mechanical resonance. My tests with ICW polypropylene capacitors in crossovers showed that the physically bigger and heavier the cap (i.e. larger voltage rating, and more mechanical damping), the 'colder', more recessed, less immediate but more detailed and timbrally faithful the sound. The small caps were more lively, and you could easily call this 'warmer'. One listener preferred the lively sound, one didn't.

If you look at the distortion spectra, the suggestion that increased odd order harmonics can make a 'cooler' sound is perplexing. The 12b4 or 6N30 has much lower odd order distortion than the 6SL7, but sounds typically cooler, while the 6SL7 is warm and immediate. Are octals and also UX4, UX5 based valves better damped physically? And then, where does high and low mu figure? Seems a lot of this is counter intuitive.
CLS said:


Would you share how you do it?
What tube(s) in what circuit?

Hi there.....suprisingly it's simple.......the 7199 app note (triode electronics) is a very good starting point for phasesplitter and front end gain. The 7199 tube is readily available from various vendors. Using a THD analyser one can adjust g2 volts for min output thd with good voltage swing. This is a good start for the
intermediate driver stage is the most critical and I use pair of ECL82's in triode config with 360V B+ running at 35mA (the triode section used as bias fail det). Interstage common cathode biassing offers excellent balance and eliminates any phasesplitter differences and valve differences. I see so many amps capable of better performance just by utilising a slam driver stage.

The output stage each side has 2 parallel pairs 6550B/C or KT88 in fixed bias......with combination B+ of 250/520V for lo/hi output powers 30/150W resp. The UL output transformers are Sowter types(UK) considerably overrated with slighlty higher a-a Z which is a key to low thd. Global feedback of 16dB does the rest of the trick and one is looking at -3dB points at 15Hz & 50Khz.
Power supply isn't for beginners......isolated 500W pfc provides all the stability and juice. Alot of the distortion is usually from mismatched output pairs and excessive supply noise and ripple.....the latter should get cancelled with a well matched o/p stage.
I'm using SA series ICW caps for interstage coupling.
I'll try and post a circuit diagram when finished.

-> CLS.....have enclosed an Xls graph of work done about 5 yrs ago......May be of interest.
Note: Engaging into a 150+150W amp project takes quite alot of time and heavy lifting. Not to discourage the attempt but the PFC took the longest.



  • kt88 stuff.jpg
    kt88 stuff.jpg
    96.3 KB · Views: 109