Anyone compared BAT Rex vs. Atmasphere MP-1?

As usual, this thread deviated from the original question- comparison of BAT REX and Atma Sphere MP-1 (MK III?).
I'm another one who would be interested in this topic.
As many others here, I've had BAT REX in my system for a significant period of time, and so far, it's the best preamp I've ever heard (others include Supratek, Nagra, Doshi).
MP-1 I only heard under show conditions, so I have no opinion really. It's intriguing to me, since it's a full-function preamp.
So, maybe somebody could offer an opinion on it? At least not in direct compar. with REX?

Hi Raquel,

This is the question that does not have a simple answer. As one of my teachers used to say - when in doubt whether simple rules apply, consider the extreme cases. Would you keep your units on 24/7 if you only listened to them for two hours once a month? Probably not. We, therefore, can safely assume that for the multitude of applications and uses there also exist multitude of correct answers.

There are several kill mechanisms acting upon the tubes (and other components), and you correctly mentioned one of them - thermal cycling. However, there are others. Of them the cathode wear might be the most important. Needless to say, both the full operation, and the typical Standby, have the cathodes working, and therefore wearing out.

Furthermore, as is known from the reliability studies, temperature always accelerates the component failures. Its effect is exponential, and its degree varies depending upon the component's nature, but it is always there.

Also, while it is true that thermal shocks are, generally speaking, damaging, it is the rate of change that is the most critical aspect of that particular effect. There are special accelerated life tests that include high rates of temperature change, and they have been shown to be effective predictors of product reliability. However, the rate is the key element here.

By controlling the rate we can reduce the damaging effect of thermal cycling manifold - also exponentially. Therefore it is always important to understand how the product handles the changes of its state. In our particular case the tubes heat up very gradually, so the rate of temperature change is substantially reduced.

All this would of course be academic without hard experience, and that experience tells us that we do not see tubes die catastrophically after 10 or 15 years, as the rule. And damage due to thermal cycling tend to manifest itself as catastrophic failure.

What we normally see is slow deterioration of tubes, most likely due to the depletion of their cathodes. Meaning - long operating hours.

As there is no simple answer, every user has to develop his own way of operating the components - we just give him the tools. All I am trying to express here is the fact that no single universal truth exists in that regard.

Regards,

Victor

Again with apologies to those who are bothered by this tangent, I think we are actually saying much the same thing, Mr. Khomenko. I have or had mentioned in this thread and/or others that:

(i) operating voltage helps determine tube life, i.e., the more voltage a tube sees, the hotter it runs, and the quicker it will normally die; and

(ii) in order to avoid the sudden, deleterious application of voltage to tubes upon start up, some tube equipment - including your products, I am now learning - features a "soft-start" function that brings up power gradually.

Since we're on the topic and your Rex preamp features tube rectification, I should mention that tube rectification functions like a variac to bring up power gradually.

Not to belabor the point, but based upon the evidence I have seen, GENERALLY speaking, if the small-signal tubes in a given circuit are run at reasonable voltages (i.e., well below their maximum rating and they thus run cool) and the piece uses solid-state rectification without a soft-start or standby feature, which is the case, in my experience, with most tube preamps, 24/7 operation yields better tube life (and better sound). I again quote from the owner's manual of my VAC Renaissance tube amp:

"How long should tubes last? It has long been known in professional circles (and probably now forgotten) that a tube such as the 12AX7 will display BETTER performance characteristics after TWO YEARS of CONTINUAL operation than when it was new. In normal use it is not unusual for a low level [small-signal] tube to last 5 years or longer. Output tubes [i.e., power tubes used in tube power amps] are another story, as they are continuously providing significant amounts of current." (Emphasis original).

Incidentally, if, because of work demands, I know that I'm not going to be listening to my system during a period of a month or more (a common scenario for me at year-end), I will indeed t

Raquel, I think with older tubes the quote might be accurate. But with modern signal tubes, particularly low level miniatures, there is a phenomena of grid contamination that has to do with the inclusion of contaminates during the assembly process. An excellent example is fingerprints when workers are not given latex gloves or the like for assembly.

In such cases (if you are using tubes with this or similar issues) leaving the preamp/amp on 24/7 is not really all that helpful, as when the grid contaminates the transconductance will fall off sharply.

Its really not a good idea to design for tubes that are out of production, unless as a manufacturer you have no concerns about the owner being able to find tubes for the unit 20 years down the road. IOW its a good idea to design for tubes that are currently being made, so using such tubes has become a fact of life.

We have seen this contamination problem with 6SN7s, 12AT7s, 12AX7s, 6DJ8/6922 and a few others. It is considerably more rare with NOS types.

I'm just pointing this out because 24/7 as a generalization is not always the panacea for best performance.

As an additional FWIW, the issue of cathode stripping is far more pronounced with power tubes. It is a matter of controversy as to whether its a problem for small signal tubes.