Tube Life vs. Heater Voltage
written by Jac van de Walle
This is the short version:
- The MAXIMUM lifetime will be achieved at a heater voltage of ZERO percent deviation.
- The MINIMUM lifetime will be achieved by using the maximum allowed tolerance.
- A Defect will be occur by exceeding the maximum allowed tolerance.
By compressing it in the above form, it will hopefully be clear, how essentially wrong it is, to use a tube at the allowed maximum deviation of the heater voltage. So even when it is possible to use a the EML 300B at -4% deviation of the ideal heater voltage, that doesn't mean this 4% is a range in which everything works the same. It is NOT so, that lifetime is only impaired a little bit, when you get exceed this 4% just a little bit. It is rather so, that lifetime is impaired significantly for each percent outside the tolerance. And even so, there is quite a difference in lifetime expectancy between a tube used at 0% deviation, and one used at 4% deviation.
Then what happens at too high or too low heater voltage?
At a large deviation, too HIGH heater voltage will simply burn down the tube, and it can end it's lifetime within a few hours, while the tube works normally for most of the time. Reason is, the damage gets compensated by the greater emission caused by over heating. Until the greater emission also gets less, and then reducing the voltage to normal will show an almost dead tube.
A large deviation at much too LOW heater voltage is bad as well, but when the tube stops stop playing, it will begin to recover once the heater voltage is corrected at it's normal value.
The reason for the above is, there are so called "seeds" in the emissive coating, which will be destroyed at too high voltage, but there are still many left, if burned at too voltage. Then, if returned to normal voltage, the seeds will develop, and will restore the emission.
A small deviation at too HIGH heater voltage will also damage the emission capability, but if t hat has not been going on too long, enough seeds will be left to begin the recovery process, once the heater voltage is corrected.
Generally speaking, damage due to over heating however tends to be more permanent which can be better understood with the above explanation. The newer the tube is, the more seeds are present, and the better the chances on full recovery. Yet any used tubes, shorten their life time considerably by over heating.
These statements are only true for EML tubes. There are many ways for the heater chemistry, and other brands may do it different, but we can not speak for those here.
This is the long version:
The above "seeds" explanation is completed by the way those seeds grow. The seeds build at first emission "points", which develop into islands, and these grow, until they all begin to touch each other, to build an emissive surface, covering the complete cathode. This process was carefully started, and completed in the factory. In a second phase, the final burn in at the end user, will make the surface more uniform, which removes the initial small distortion, new tubes may have.
It can not be repeated often enough, that longest tube life results in the first place from the heater voltage being correct, Anode dissipation being not at the maximum limit, and the burn in by our recommended procedure is the finishing touch.
Though slow heat up with electronics modules have the theoretical benefit, to prevent heater breakage, in real life these cause cathode poisoning. The benefit of electronics will result from the filament voltage being exactly right. This simple requirement we find so often ignored. Yet, heater voltage precision is the easiest to fulfil and the most important of all.
Slow heating up modules.
If you want to impair the lifetime of you tubes, you must use those modules, which heat up the tube extremely slow, in a false attempt to do it "better". Really, tube technology is 100 years old, and when you read good books about vacuum tube design of those days, you will realise how much they knew about it. This is extremely impressing. At least it is for me. We all know, modern Getters are made of Barium, and that is because it has gas absorbing capabiltiy. However that capability at room temperature is low. For what we call "high vacuum", the getter must be in between 180°C and 600°C. Above 600° the getter reverses partially, it will begin to release small fractions of it's gas again. Here comes the problem with slow heat up. The heater is also Barium coated, and the heater will act as a getter, if below 600°C. Actually like a very good getter, because the slow heating "specialists" will keep the tubes for 20 or 40 seconds in that ideal range. So the cathode will absorb any gas residue. We call that cathode poisoning. So yes, it is nice when the heater removes that gas, only... the pure Barium, which we so carefully constructed at it's surface will be turned into an some other composit, like Bariumoxide, Bariumcarbonate, or whatever the gas residue was. The reverse process (above 600°C when normal orange color appears) is slow and incomplete, so as an avarage, gas will stay attached to the cathode, and damage it's surface.
I do this business since 1995, and there is a lot of ignorance with people who construct those electronic modules. Such modules can be good, but only if they are designed with knowledge about what tubes need, and this is simply FAST heat up, but avoid excessive fast heat up, and avoid slow heat up ALWAYS. The best has always been the natural current limiting, like an AC heater winding provides, like just 2...3 seconds and then already the heater voltage is close to 90%, and much above 600°C, and we are out of that danger zone as quick as we can. .
With electronic modules, it looks to me, any kind of strange concept is adopted uncritical, as long as it looks 'interesting', and when the maker say's it's good, that is regarded the absolute truth.
But when we say: 'Please try to get heater voltage of zero percent error' that is is regarded boring, until problems occur. Then, data sheets are read, when it is too late, and the crying is loud.
Electronic modules without any current limiting during switch on, may do damage to the tubes, so electronic current limiting is strictly necessary. A module should not provide larger start up current than an AC transformer winding, or it will shorten tube life rather than make it longer! Such things can happen, when over sized regulator IC's are used, and transformer windings with too low internal resistance. To have an idea of the start up current of a cold tube, simply do this: Measure the cold resistance of the tube, and then use Ohm's law. In case you already know how to measure the cold resistance, still please read this: The cold resistance is what you measure, MINUS the resistance of the test leads. The latter you can measure by shorting the test leads. It is 0.3 Ohm often.
You will be surprized, cold current can be up to 10x higher than the warm current. Any AC transformer winding however can NEVER supply 10x that current at normal rated voltage. If you try that, the voltage will break down, reducing the current as well. That is a normal electrical property of a transformer winding. The skilled transformer builders for vintage tube amplifiers, OF COURSE knew how to do this, and used heater windings with appropriate internal resistance. So the unloaded voltage is a higher, and voltage will drop significantly at over load. So current limiting was invented already, by something simple as transformer winding internal resistance.
Even so, a minimum resistance is a mandatory requirement for AC windings for tube rectification, as per RCA data sheets, they even tell us in writing, but this is off topic here, we just want to refer to it. (Read the 5U4G data sheet by RCA). Just with electronic modules, they can be real tube killers, ignoring the basis, and doing things 'better'. Frankly if there is a tube problem, it is 95% of the time caused by an under heated tube, and there is extreme resistance with the amplifier owner, getting involved into this. The attitude is, when heater voltage is not outside the tolerance, everything is fine, and this must be automatically so, or otherwise the tubes are bad. Unfortunately, this is not how amplifier problems are solved.
Direct or Indirect heated?
We need to distinguish first between Direct Heated Tubes (called DHT here), and indirect heated tubes. The last have a heater, which is a heater spiral only, and not the emissive part. This heater spiral is not under mechanical tension, and can be made of Tungsten, which is very rugged. So with indirectly heated tubes, break risk is low.
Directly Heated Tubes however, do not have a heater, but rather what we call a filament. Though the words heater and filament are often used together. The filaments of a DHT have an additional requirement, compared to heaters of Indirect heated tubes. DHT tube filaments are under mechanical tension, with a spring, and they may not break or extend length over time. The breakage will end the tube life of course, and extended length may cause a short circuit when 'sound research specialists' are tapping on working tubes. This has a devastating defect on tube lifetime.
However to deal with the fact, that a DHT filament is always under tension by the spring, the very high lifetime of EML tubes will only be reached by making the filament not hotter than needed. So any heater tolerance of more than 5% is unwise. If you exceed this 5%, tubes may seem to work normal for a long time, but maximum lifetime will not be reached. So this targeting for the lowest temperature possible, has an additional result when we under heat the tubes! In that case, the filaments will become not warm enough when the voltage is more than 5% too low. This explains in a nutshell, why tube filament voltage may have so little deviation.
Historical reality says: 5%.
Throughout history of making Barium coated electron tubes, we see always the same specifications for the heater voltage: +/- 5%. In some rare cases tubes are specified at +/-10%, but this does not mean these manufacturers have invented something which nobody else knew about. This was only for special applications where it had to be specified this way, and a huge compromise was made with lifetime and reliability. Allowing for -10% means the filament must be designed a bit hotter by default, there is no better trick. So at -10% there will be enough heat still. Then, if such a tube is used at +10%, end of life will come sooner. Meaning any +/- 10% tube has some compromise with the data, which compromise must be no disadvantage, but it exists still. It appears best lifetime and highest reliability is only possible when the filament voltage is specified +/-5%. And even so within this +5% and -5% there is the middle of 0% which is simply the best. We must take this as reality.
Tubes have a balance inside, to maintain emission.
- The emissive Barium surface layer evaporates simply by heat of the filament. This layer is just a few microns thin, and evaporation accelerates very much at higher temperature. So just 5% more filament voltage has a much larger effect on this as just 5%. Inside the filament coating is the Barium Oxide depot. This migrates slowly to the surface, and forms new metallic Barium at the surface, filling up the missing parts of the metallic, emissive layer.
- Both processes (1 + 2 ) go a lot faster at greater heat, like all chemical processes. So this simply wears out the tube much too fast, and problems are non reversible when the depot is empty.
- If the tube filament is not warm enough, there is not enough migration of new Barium to the surface. Of course there is also less evaporation, but the balance will fall into the direction of not enough migration. So the metallic Barium layer will disappear. However now the depot is not used up. So such tubes may be recovered by just normal use.
- If the tube filament is really much too cold, it may begin to suffer from cathode poisoning, as some specific unwanted chemical traces do not evaporate also, and this can develop into an emission killing surface. Such a tube may be recovered by the classical process, of over heating the filament, but with EML tubes we found this not needed. Besides a tube that was first under heated, and then over heated to 'repair', it is easy to understand it will not be like a new one any more.
- The ideal situation comes, when evaporation and regeneration are nicely in balance. You can trust, we designed the filaments exactly in such a way, that this happens at exactly the specified heater voltage. (meaning +/- 0%).
Highest possible lifetime
Admitted, there are cases where lifetime is not the first concern, for instance when companies specify an amplifier with exceptional wide range of mains voltage, and still use AC heating for the tubes. Any change in the mains voltage will also mean a change in the heater voltage. In such cases it will come down to stressing the 5% limit as much as possible. Or worse, they do not respect that this 5%, and we have a debate why they can not exceed this by just a few % extra. However when tube life is your first concern, deviation of the heater voltage should be 0%. Saying this in other words: The maximum deviation of heater voltage (+/-5%) is not a safe limit you can choose as you like. At 1% deviation a tube will last longer than at 2% deviation, etc, and 5 % is the hard limit. Positioning yourself at this limit will result in reduced tube life.
So after reading this, one may think, are EML tubes so sensitive against heater voltage deviation? The answer is: No, they are not more sensitive than other tubes. It is just every time when we get reported very high lifetime of our tubes, we find this was a result of good amplifier design. It must be said like this: At maximum deviation of the heater voltage, so just within the specs, it should not come as a surprise to you, that lifetime will be also just within the specs as well. And no more than that. In other words: The less deviation of heater voltage, the longer the tubes will last, and this is so for any DHT, any brand, new made, or NOS.