In the Conrad Johnson amp the tubes aren't biased for too high idle current.
How does J.H. van de Weijer prove that HE is correct. This cathode stripping has been described by others to be caused b/c the cathodes are hot, and not having a B+.
Mr.J.H. van de Weijer's considerations of what happens at molecular levels is no more proof than the other claims. And how can he say this when saying it applies to SOME tubes, as in not all, and WHICH tubes does he mean then?
No matter how we treat tubes, SOME of them will last longer than others, and some will die earlier...
We all agree that tubes have a finite lifespan. And minute imperfections in how we design to use them can have an effect on lifespan, but if solve for one condition, you have the other school of thought that may not satisfy longest lifespan, etc etc.
Iam willing to stake my neck out, and claim Mr. J.H. van de Weijer's hasn't proven anything, merely written down what he believes, adn thus creates yet another myth, be it true or not.
Fact remains, tubes have a finite life, B+ on cold cathodes have little effect on that.
Now, regarding the cycling on/off hot/cold cathodes matches the conclusion I got from reading 'Getting the Most Out of Vacuum Tubes'. It is nothing to do about B+, but the mechanical stress on the heater itself. Which is why I now use a soft start on the heater to limit inrush currents. (Though I think even that is overkill).
It certainly can't hurt, although interestingly the paper I referenced earlier has this to say:
"In the ENIAC where anodes and screens are operated at about half their continuous service voltage ratings, and about a quarter their maximum current ratings, experiments tended to show that the optimum heater voltage for 6.3V valves lay somewhere between 6.0V and 6.3V.
It seems likely that heater failures will be reduced if it can be arranged to avoid the heavy current surge which normally occurs when switching on. This is admittedly only conjecture, and tests on the ENIAC failed to prove that it gave any measurable advantage."
(The ENIAC was a computing machine)
I too have found heater failures to be extremely rare. It has also been pointed out by metallurgists that if the heater wire is perfectly uniform, it is impossible for it to fuse due to inrush current, since the inrush is proportional to its conductance which in turn falls with temperature. It is the minor defects in a heater which cause it to fail.
I don't know whether this allows for the mechanical stretching/fatiguing though.
You are confusing cathode STRIPPING with cathode POISONING (zwischenschichtbildung) - which latter is a risk when indirect heated tubes (with ordinary tubular sleeve cathode base, rather than the lock-seam cathode) are heated with no, or small anode current.
Well, Philips made a huge number of vacuum tubes of all sorts. Their research staff could then be expected to have far more valuable things to say about the long-term behaviour of the materials inside them, than the average mountebank.
If you have some hard facts about cathode materials, let's have them.
We are not trying to prove that all tubes will suffer, or that all tubes will be fine if no pre-heating is used. You would need a huge sample to do that, and the equipment needed to analyse them.
What is on offer here is an explanation, from the staffer of one of the longest serving and biggest names in tube technology saying WHAT HAPPENS when you don't preheat. If you think that nickel sleeves and ceramic-matrix cathode coatings expand at the same rate, then you won't believe that particles of cathode can be shed. If you don't think that isolated islands of charge may remain on a cold cathode, you won't believe that these particles can be drawn to the grid.
I think these things are certainly plausible. That doesn't mean it's proof, but sufficient evidence that for expensive valves, or my best amplifier, will have preheating.
Of course you can do what you like, but the difficulty is assessing whether you amplifier is degraded after some cycles of heater use. You might get a bit more distortion (from having an 'active' grid), you might get more noise.
The rest of your system might not be good enough to allow you to tell.
But if you want to stand up and say -
'Applying HT to cold cathodes does no harm',
first get the advice of someone who has actually designed a cathode in a vacuum device. The rest of the world is just guessing - and if you want to do that with hard-to-replace valves, then I recommend to guess safe.
You are not adding anything to the debate here.
This effect would be expected to degrade an amplifier, not cause a breakdown - except in the case of large grid leak resistor, and a great deal of heat cycling.
Have you never been asked to look at a valve amp that showed a noisy valve, or a bit too much grid leakage? How can you be certain that there were no bits of cathode stuck to the grid in such a case?
I don't claim gross failures for this effect. It's more of a precaution that - if you want the best amplifier, be aware of this risk (and if you're DIY, why would you ever want less than the best?)
I've owned a few of those. Amazing how long they keep going. But guess what? They all have Tube rectification.
Tell that to Marantz and Dynaco. I won't claim statistical significance for the hundred or so amps and preamps I've build using SS rectification and no B+ delay, but my experience is consistent with theirs.
Well designed amps using non-faulty tubes? Nope.
What is their experience exactly?
Are you suggesting they (or you) surveyed a significant number of valves used in crash-started SS-rectified amps? and found that there was no evidence of degradation due to cathode particles attached to the grid?
Of course they didn't, and a commercial company of decades past is hardly going to care if customers have to change their valves after many cycles of use. Even if they are able to tell that anything is wrong.
My point is that a DIYer, striving to achieve the best sound possible, DOES care about degradation of performance, even if it is very hard to measure, and of minor influence. If you're using expensive or hard-to-get valves, these considerations might be important.
If you're just making an amp that is used for more or less uncritical listening, you may not care. Lack of preheating may not actually break the amp. And again, if you're trying to make money from an amp design, you may not care - the customer can just go buy new valves if anything goes too far out of spec - you're not going to add cost to give a preheating facility. That consideration will apply to Dynaco, for sure.
In most of my amp stages, I use a separate mains switch to preheat the valves, and use slow ramp-up 6CJ3 damper rectifiers, and current limited DHT heating. None of this may make the slightest difference, for all I really know. But the van de Weijer explanation is plausible, and I don't want to risk having to find the reason for any slow fall-off in performance of my amps.
van de Weijer's advice carries more weight than the kind of 'I've never seen it - therefore it can't exist' approach.
If we can summon up an engineer who has actually worked on cathode materials in vacuum tubes, you might get better insight on this, and on van de Weijer's words. Until then, we all carry on building based on what we THINK. It's DIY here, after all.
If you read MJ carefully (or at all), he offers that what does kill a tube is just turning it off and on. So, if you want to save your precious tubes, never turn the amp off, or never turn it on - your call.
Or sell the tubes, and buy a bottle of fine scotch. All your worries will go away, and from what I hear, MJ would heartily approve. I personally cannot think of anything that would improve the sound of any amp than a fifth of Laphroaig.
Somehow, I imagine a Ferrari forum where some might advocate not making turns in the first 15 minutes to save the tires. Prove that wrong, Speed Racer! Weird. Tubes go in sockets for a reason... just as tires go on rims for a reason...
Several hundred thousands of amps with decades of reliable use. That certainly beats, "I read something somewhere that somebody put on teh internets." Throw Audio Research into that mix as well, if you want an example of a more contemporary tube amp maker with a reputation for reliability.
What I referenced gave an author (van de Weijer), a relevant source (Philips) and an explanation of an effect that is plausible. Compared to your own unquantified interpretation of a vague measure (reliability) of a third parties (amp makers) who have published no evidence, or data, or even claims to back it up.
That is even before you consider that reliability is not even the claim I made for the effect, but rather degraded performance of a valve over long service. Maybe noisy, maybe degraded sound quality. An end-user might just buy new valves and not feel cheated, nor that the product was 'unreliable'.
Whether this experience translates to DIY depends on your goals. If you build a new amp every other month, or tuberoll constantly, or swap valves for any other reason, you may notice nothing at all of any of this. But if you are aiming high, and want the best from expensive valves over a long life, you may wish to consider effects that do not concern commercial operations, or those that just 'build and see'.
A casual, unsupported statement, posted in a newsgroup (not in a Philips publication). Versus decades of market experience and hundreds of thousands of units in the field. If the former is what you'd care to believe, have at it.
I do note that I cannot find ANY recommendation on ANY Philips datasheet for receiving tubes. And many (most) of their datasheets call out a specification for maximum cold anode voltage. There is also, as Gimp points out, nothing in military reliability manuals about this issue.
But hey, some guy, whose name appears nowhere in any of Philips' technical documents on tubes, wrote something on the internet. Must be true.
There you go!
What I was trying to say in my last post, which according to someone has no relevance, is that tubes will die, and some will die from poor QA during manufacture, some from impurity in metals, some from poor vacuum, some from heater problems, some from many events. Many will die by a combination of many events during their use.
So that document from the guy who worked at Philips says nothing that proves that B+ on cold cathodes kills tubes. Just that SOME tubes wont like it, but as mentioned, SOME tubes will suddenly fault for some other reason as well, so who can say with 100% certainty that it was due to B+ on cold cathodes....the story goes on and on...
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.