Cryogenic Treatment of Tubes: An Engineer’s Perspective

A re-post from Phil Taylor of Effectrode.com

I’m not a metallurgist nor a physicist but as an audio electronics engineer I’ve a vested interest in anything that might help improve the sound quality and reliability of vacuum tubes. There is varability in microphony or electrical noise in batches of tubes and I’ve often wondered if there might be some kind of a fix for this as it seems a shame to reject a tube – which is an expensive component and completely functional – just because it’s subjectively slightly noisier or a little more microphonic than another. It briefly crossed my mind that cryogenic treatment (cryo-treatment) might perhaps reduce the inherent white and fluctuation noise that tubes generate. After all, there are many tube vendors that offer this service with claims that it “causes metallurgical molecular changes in the metals within the tube which enhance tone and increase overall life”. Does enhancing tone mean lowering noise and microphony? And can simply cooling a tube down low temperature improve its characteristics? Here are a few thoughts on the matter.

What is Cryogenic Treatment?

Broken gear tooth
Cryogenic treatment is used to improve the durability of moving steel parts.

 

Cryogenic treatment is a process of cooling steel alloy down to low very temperatures (−190 °C) to increase its surface hardness to improve its resistance to wear, the practical application being to extend the life of cutting tools, gear teeth, moving engine parts, that kind of thing. Cryogenic treatment alters the crystal structure of steel by completing the conversion of austenite to martensite making it harder. Now, I recall making a screwdriver at school in my early teens. The tip of the screwdriver was hardened by heating it up with a blowtorch and then cooling it rapidly (‘quenching’) by plunging it into a bucket of water. It’s my understanding that quenching the steel in this way and not allowing it to cool down slowly prevents austenite from forming and makes the steel much harder and more brittle. The process isn’t 100% perfect though as steel still contains some austenite crystals. Apparently cryogenic treatment completes the conversion to further harden it.

Effect on Vacuum Tubes?

Effect of underwear on tone.
Two different crystal structures that occur in steel which have absolutely no measurable effect on tone. (Picture by Tom Duerig)

 

So cryogenic treatment works to improve the hardness of ferrous metals such as steel, but what about a vacuum tube – a tube is not a lump of steel – it’s a delicate and complex component made up of many different materials. The metal electrodes are mainly nickel for the plates and heater cover, titanium for the heater wire, molybdenum for the grid wire and copper support posts. The heater is coated with strontium/barium oxides. The electrodes are supported by mica washers within a glass envelope. One can only guess as to what effect cooling a tube down to so such low temperatures will do to it as there’s been no serious scientific research investigation into this. Additionally, tube vendors have done little in the way of publishing noise measurement figures and life tests comparing treated and untreated tubes. Typically all you’ll find are references to surface hardening, maybe nice magnified images of the surface of cryogenically treated steel and even rhetoric about NASA and Einstein on vendor websites. All this is about as relevant to your guitar tone as the colour of your guitar cable or your underwear. Could it be that cryo-treatment is pants?

I did find several scientific papers indicating that cryogenic treatment hardens aluminium. Perhaps it hardens copper, nickel and the other metals used in the construction of tubes too. Although there are no moving parts in a tube and no mechanical wear, hardening the electrodes would theoretically make them stiffer. Stiff electrode structure is good thing in a tube as it should reduce microphnony, which is caused by the electrodes moving relative to one another. However let’s keep this in context. Cryogenic treatment is just a finishing process used to complete the conversion of austenite to martensite in steel as described earlier. The metal components inside a tube aren’t quenched to make them super hard in the first place so there is no conversion to complete. At this level cryogenic treatment achieves nothing and I’m inclined to think that it could potentially damage the tube. Cooling a tube to cryogenic temperatures would put undue thermal stress on the glass envelope and delicate internal parts. The manufacturers never designed or specified tubes to be stored at such low temperatures. The act of putting one of these delicate devices through cryogenic treatment will cause contraction and expansion of the different materials and increase the chance of loosening some of the internal parts. The mica may no longer stay in secure contact with the glass envelope causing the tube to become more microphonic.

Effect on Tone?

Cryogenically treated Mullard 'reissue' tube
Where are the comparative noise measurement tests before and after cryogenic treatment?

 

At best, exposing tubes to the stresses of cryogenic temperatures provides a test methodology to reveal potential early failures – a kind if negative null test. It does not improve performance, however it might explain why some people can hear a difference – it’s simply because they’re listening to a hand-selected tube. There is no evidence for the process actually improving noise or microphony though. Nor can I find anything of substance online that indicates how cryogenic treatment might possibly improve the ‘tone’ of tubes either. Does it change the way electrons are emitted from the cathode? Does it change the way electromagnetic fields form inside the tube? Does it improve the bonding of the oxide coatings to the cathode to make it quieter? Does cryogenic treatment affect the tube in any measurable or audible way at all? Tube vendors give us the science of how they cryogenically treat tubes, but what about the science of what it does to improve the performance of tubes, eh?

You’ll find plenty of fluff on vendor sites such as claims that our cryo-treated tubes have “tighter focus from top to bottom”, “more holographic 3D soundstage”, “more subtle inner resolution extracted from recordings”, “tighter bass”, “increased dynamic range”, “faster transient response”, etc, etc. If cryogenic treatment really does have an astounding and revolutionary on tube performance then the world needs to know about it. The skeptics need to be convinced. This is too big to remain on the fringe. Claims need to be supported with independently assessed noise and microphony measurements, preferably accredited by some external body like the National Physical Laboratory? If I were in the business of cryogenic treatment I would consider it a matter of some urgency to have this data to backup my claims – it would give me confidence that my work was of value. If this really is cutting edge stuff, why not apply for a research grant, write a thesis on the subject and make a genuine, significant contribution to the art of vacuum tube manufacture? What is there to lose?

Didn’t Mullard Know About It?

Mullard Blackburn oygen and hydrogen plant.
Oxygen and hydrogen production plant on the Blackburn site – Mullard had the motivation and resources to explore cryogenics

 

I don’t want to labour the point but it’s worth considering this, if there was any merit in the cryogenic treatment of tubes then surely manufacturing giants Mullard or Philips would have utilized the process. They were in the business of making tubes not just for guitar and hi-fi amplifiers but for mission critical military and scientific instrumentation applications. They were highly motivated to improve their production processes. They had vast pool of scientific and engineering resources at their disposal. Their 43 acre site at Blackburn, Lancashire had it’s own liquid oxygen and liquid hydrogen manufacturing plant. The site employed almost 7000 people amongst which were all kinds of specialists including physicists, chemists and metallurgists. Resources far beyond the small handful of eastern tube factories that serve the guitar industry today. If there had been any benefits to cryogenic treatment of vacuum tubes, Mullard must have known about it. I suspect that somewhere there is, buried in some university basement library or an old radio ham’s attic, a technical paper written by a Mullard engineer in the early 1960s titled ‘An Investigation into the Effect of Cryogenic Temperatures on Thermionic Emission’. I reckon it’s just a question of time before this paper shows up. What do you think the Mullard engineer’s findings would have been?

In Conclusion

There is variation in the noise and microphony of vacuum tubes because of engineering limitations or to put it another way, our ability to accurately and consistently fabricate these complex devices. Mullard in their heyday, with vast resources and a wealth of expertise at their disposal got as close as anyone to manufacturing the perfect vacuum tube. However, even their tubes were subject to variance in component tolerances which allow electrode movement and sensitivity to microphonic pickup; DC leakage paths in the mica insulation spacers that allow small currents to flow where they shouldn’t and fluctuations in thermionic emission from the cathode oxide coatings, both of which result in self-noise. It would be magical if cryogenic treatment were some kind of ‘cure-all’ that improves the insulation properties of mica spacers, reduces inter-electrode movement and improves the emission of the cathode coatings to reduce noise and microphony. But I cannot imagine how this works – ‘magical’ really is the right word to use here because cryogenic treatment of tubes is certainly not science. There are no well-considered explanations describing how the cooling process works to improve the electrical characteristics of a tube or even the most sparse comparative test results to validate claims for improved performance.

It concerns me that the vacuum tube industry today is currently on a path of re-branding tubes under the names of once great manufacturers and cheap marketing gimmicks such as tube dampers, tube coolers and now cryo-treatment – anything but real engineering it seems. I have several boxes of tubes I’ve rejected over the years because of subjective failures such as noise and microphony. I’m as motivated as anyone else to ‘fix’ them. It would be a dream to be able to somehow restore them to optimum performance. But that’s what it will always be, a dream. The bottom line is that cryogenic treatment will not make up for inferior materials or defects in tube construction. If you’re purchasing tubes, a better investment would be tubes from a reputable vendor that just tests, matches and guarantees them. Better still, seek out genuine N.O.S. tubes made by the likes of Mullard, Philips and others if you can find them.

If you’re interested in what the Mullard Blackburn factory was like in its glory days then do take a look at the following article ‘Speed, Efficiency & Perfection – Aims That Have Built a Mammoth Factory in 16 Years’ originally published in 1954 in the ‘Blackburn Times’ not long after the factory opened.

19 Responses

  1. Alberto
    Alberto at |

    We, the audiofans want to listen music better and better so the audio vendors invent what is necessary to sell as much as possible.

    Excellent article

    Reply
  2. Tim Stanley
    Tim Stanley at |

    I would like to think if it would help that someone would figure it out and say for sure because my Long plate Mullard 12AX7 sure are noisy sometimes. Lol. But they sound so good in my Marshall Jcm 900 SL-X

    Reply
  3. Dan Lilley
    Dan Lilley at |

    Thank you for this excellent article and for taking at least a basic scientific approach to the issue. The audio industry (sadly, IMHO) is rife with misinformation, unsupportable claims, conjecture and delusion. Your approach at least helps to clear the air and I applaud you for it.

    A tube aficionado and satisfied customer,

    Dan Lilley
    Raleigh, NC

    Reply
  4. Pekka J. Poutiainen
    Pekka J. Poutiainen at |

    Great article. Some tubes had a dirty weldings inside – so noisy then!

    Reply
  5. GOIK
    GOIK at |

    Interesting…..but as many stuff in this audio-field, completely subjective. What always impress me is how malleable some people can be.

    Reply
    1. Aleksandar
      Aleksandar at |

      I would not attributed great importance to this text, because the author of the engineer.
      He has no audiophile’s ear, and hear with engineering head !
      When engineers wondering about the role of power cords, he will give you ” expert” to respond to whatever the cable you use. For them there is no difference between the stock cable and Shunyata Research KC cx .
      This is just another opinion, which has no meaning for true audiophiles .
      Engineers should know (but often do not know) that there is a lot of natural phenomena that science has not yet explained. It would be good to follow the observations of some audiophiles, this may be a good hypothesis for a new research.

      Reply
      1. FreewayJam
        FreewayJam at |

        Aleksandar – Whoooosssshhhhh !

        What i do now for sure is that the “audiophile” industry will be forever grateful, and thankful,
        that ‘Aleksandar’ and those “true audiophile’s” will always keep the cash register’s ringing !
        Oh, and that an unbiased article published by a respected audio engineer that details vacuum tube construction, performance and history among other thing’s, is just an “opinion” that has no meaning among the very special and elite group that is the “true audiophile’s”.
        And i’m not sure what all the natural phenomena out there is that he speaks of ( Aleksandar that is ) but my guess is he is part of the group that ignore’s any proven science, when that proof doesn’t fit within his argument / stance. ( and then the conspiracy theory’s start ! )

        Reply
      2. John Sullivan
        John Sullivan at |

        Aleksandar that is bullshit: anything and everything the human ear can physically hear can be measured in some way, shape or form. To say an engineer doesn’t know what he’s talking about when it comes to this sort of thing is pure horseshit. And if that doesn’t satisfy you, a recent double-blind test with half engineers and half “audiophiles” yielded the same results. Not a single one of them could even tell when tubes were swapped out, let alone which ones were cryogenically treated and which ones were not. You are full of it.

        Reply
      3. Bill White
        Bill White at |

        Just go on believing horseshit mate. I could put you in double-blind tests and you would make a fool of yourself. You’re in fantasy land.

        Reply
  6. FreewayJam
    FreewayJam at |

    typo – ‘What i do know’ not now

    Reply
  7. Mike Joyner
    Mike Joyner at |

    Too Bad Alex cannot assimilate well explained background including fundamental design considerations. Emperors clothes always on sale

    I go back to the days of vidicons and plumicons and very much appreciate the history disclosed here. He is correct at the efforts and resources brought to bear to build a better tube at no expense be spared. If they could squeeze out another watt, another electron, another dB of performance, they would be all over it. When you see that circuits are now built in sub mm acreage, it is amazing what they accomplished with point to point wiring and large racks and rooms of equipment

    Reply
  8. Frederick J Diekman
    Frederick J Diekman at |

    Mr. Taylor,
    Your article on YOUR perspective on Cryo Treatment of tubes is well written but filled with suppositions that are not correct. Yes, there is a lot unknown about what Deep Cryogenic Treatment (DCT) really does, but there is more than the austenite to martensite transformation that you assume is the only result. Your statement that cryo “…is just a finishing process used to complete the conversion of austenite to martensite…” is flat out wrong. DCT reduces point defects in the crystal lattice structure. It reduces residual stresses. It helps refine the atom to atom distance in the crystal lattice structure. It refines the distribution of alloying elements. In steels, it creates tiny carbides evenly distributed within the crystal structures. It also modifies how the metal vibrates. DCT has a definite effect on welded structures. It stabilizes metals to prevent changes of shape. It increases the fatigue life in metals up to six times. As you can see, most of these changes occur in the crystal structure of the metal, not the microstructure.
    Your defining of DCT leaves a lot out that is essential. The cooling of the component being treated must be slow. It must be held at the lowest temperature for a period of hours and then slowly brought back to ambient temperature. You also say that DCT is done to “increase the surface hardness.” Please be aware that DCT affects every atom of the component being treated
    Your statement treating a tube will be harmful to the tube is also conjecture. We have treated thousands of tubes without damage due to the process. Our customers report increased life, not decreased life. There is evidence of improved microphony. Our experiments indicate that there is reduced vibration in treated parts, and the vibration that is left has resonant frequencies greatly reduced. We have found this in many metals. Yamaha and Powell Flute have both tested DCT in blind tests and use DCT because of their tests.
    It appears that you have stated reasons why DCT would not improve vacuum tubes based on false perceptions and failure to actually test treated vs. untreated tubes. One reason manufacturers don’t like it is that a major cause of tube failure is thermal cyclic fatigue and DCT reduces that and creates a huge increase in tube life. Another is that they ask the opinion of engineers and metallurgists who only know about the conversion of austenite to martensite. This leads them to the wrong answers. Most of what DCT does is done at the crystal lattice structure level, not the microstructural level. If you want to see tests, I would be very happy to work with you or tube manufacturers on valid testing to prove the concepts. I’ve studied DCT for over 20 years. I wrote the definition of DCT used in Volume 4A of the ASM Handbook. I work with the Cryogenic Society of America to help dispel the myths written on DCT.

    Reply
    1. Dave K
      Dave K at |

      You believe that it is conjecture to state that cryogenic treatment could be harmful to a tube, even though it is well established that different substances react (i.e. expand differently in response to extreme temperature changes), but I fail to see how any of your claimed benefits are anything other than conjecture as well. “It creates tiny carbides evenly distributed”, “it increases fatigue life up to six times” “it helps refine the atom to atom distance”…you use pseudo scientific words, but can you point to one reliable scientific study that backs up these claims, or points to the audible differences attributable to these changes?

      The big tell is your use of individual reports. Please tell me how a person can tell that any particular tube has “increased life”. I’ve had multiple tubes from the same manufacturer and vendor, and they all fail at different times. some within days, others last for years. The difference was not cryo treatment – in fact, without expert analysis it’s impossible to tell what led to the difference.

      Your assertions are not made more valid by don’t of length of reply or your membership is an organization specifically dedicated to advocating for the benefits of cryogenics. The blog author may have holes in his argument and rely on conjecture, but he’s not an expert. What’s your excuse?

      At least you announced your allegiance at the end.

      Reply
    2. John Sullivan
      John Sullivan at |

      Please show your “evidence” then. Make it available to the public for rigorous peer review without hiding behind the old “intellectual property” and “trade secrets” excuse for nondisclosure.

      Reply
  9. vonfiat
    vonfiat at |

    I have a 12axy preamp that I just converted to DC power, what an amazing improvement. That being said it still doesn’t make my badly recorded CD’s sound much better. Why so many will spend 10,000 on a speaker cable and not a better recording is beyond me. Buy some Jesse Cook, Chesapeake, Al Dimeola, Weather Report, Pink Floyd and enjoy the music. Find the best musicians and their best recordings, couple that to a decent audio system to get amazing sound. I use a Conrad Johnson sa 400 200 watts per channel into Thiel 3.6 with a basic 12ax7 tube preamp and an Onkyo CD player and better than generic cables; for under $3000 you cannot find a much better system. For Ten Thousand in speaker cables I can buy 500 audiophile recordings. Why buy a Lamborghini if you live in Paris with no highways. Get a Shelby Mustang and go to Montana where you will get to drive a Hundred different roads really fast. Get a decent system but get better music and lots of it…my two cents

    Reply
    1. Bill White
      Bill White at |

      Very sensible comment vonfiat.

      Reply
  10. J-75
    J-75 at |

    All the positive benefits of submitting a tube to cryo…
    My, my!
    The magic bullet. “Spooky science” (A. Einstein).

    Reply
  11. Tim
    Tim at |

    I’m amazed at how much we chase the one percent of difference and pay exhorbitant amounts for just one component in a whole signal chain. Sure, a bad ingredient can ruin a recipe but it’s like the difference between baking a cake with gmo grown wheat flour and the branded organic wheat flour. Most of the time I can’t really tell the difference. It still makes a cake.

    Reply
  12. Mike
    Mike at |

    I would have much more respect for the original author had he simply ponied up a few bucks and actually TRIED a set of cryogenic tubes to see if there was any appreciable difference.
    That said, from what I recall from my college metallurgy course, Mr. Diekman has provided more credible information than did the original author. Have you ever heard an annealed copper IC cable compared to one that was not? Try some audio note or Kondo wire in your system one day and you’ll see that annealing much like cryo relaxes and realignes the crystal structure of “worked” materials and will provide a more relaxed and natural sound.

    Reply

Leave a Reply