The "minimalist" proponents of rectifier tubes instead of SS say that no delay circuits are necessary, because the heating of the filaments is done progressively, but to my understanding, the initial current blow on all the filaments of the valves exists , and then drops as resistance increases as the filament heats up. I'm right ?
Did you determine the actual cause of the failure?
When operated within design specifications, a KT88 shouldn't need a HT delay.
What is the g2 voltage?
What is the idle dissipation?
What is the g1 grid leak resistor?
Was the KT88 at the end of it's usefull life?
Perhaps an unreliable brand?
Is there a fuse in the HT line and/or KT88 cathode?
What is the g2 voltage?
What is the idle dissipation?
What is the grid leak / g1 bias circuit resistance? The original UK KT88 datasheet states 220k (<35W Pa) and 100k (>35W Pa). 6550 max 50k, KT120 max 51k.
Point is: when you put a larger type in a potentially explosive circuit, you are waiting for a bigger bang.
Probably varies greatly based on tube design. In 30+ tubes, I have never experienced a 6c33c filament failure. It's built like a toaster oven. I leave the filaments fired up for hours as a standby. Nice little space heaters 🙂
On the DAC amp, I use NOS Marconi 5u4gb rectifier tubes. They have been going strong now for 5 years. Certainly provide a nice soft start, but not needed for the 6n2p tubes they power.
I would not worry about it. A vacuum tube is such an inanimate object, it is just not going to care any more about being "blasted" with 500v on the plate, any more than if you do that to an automotive plug........
-g
But it would still follow common physics rules, right?
Just one issue from the article in AudioXpress:
There are several mechanisms at work here. In normal operation, the heated cathode emits a cloud of electrons that bunch together at the cathode structure and his cloud is known as the space charge. With anode voltage applied, electrons are attracted to the anode and leave the space charge, which is supplied with new electrons from the cathode. But when the cathode is cold, the space charge is minimal and is rapidly depleted when the anode voltage is present. Normally some electrons collide with stray gas molecules, producing ions that are attracted to the cathode. Ions are much more massive than electrons and when they strike the cathode, they can damage the emissive structure and cause small craters, further decreasing the emissive capacity. In normal operation, the space charge will repel these ions but as mentioned, with a cold cathode the space charge is minimal and has much less capacity to limit the ion bombardment on the cathode.
Jan
Just one issue from the article in AudioXpress:
There are several mechanisms at work here. In normal operation, the heated cathode emits a cloud of electrons that bunch together at the cathode structure and his cloud is known as the space charge. With anode voltage applied, electrons are attracted to the anode and leave the space charge, which is supplied with new electrons from the cathode. But when the cathode is cold, the space charge is minimal and is rapidly depleted when the anode voltage is present. Normally some electrons collide with stray gas molecules, producing ions that are attracted to the cathode. Ions are much more massive than electrons and when they strike the cathode, they can damage the emissive structure and cause small craters, further decreasing the emissive capacity. In normal operation, the space charge will repel these ions but as mentioned, with a cold cathode the space charge is minimal and has much less capacity to limit the ion bombardment on the cathode.
Jan
Common physics: ions can be positive or negative (loosing or gaining an electron from the impact with an electron)
If the problem is positive cations, they are actually attracted to the electron cloud, so unless the protective mechanism works by absorbing multiple electrons on its way through the cloud, gaining negative charge and subsequently being expelled from the cloud, this is not the problem.
So we are talking about negative ions.
Why are those anions going to the cathode at all? The anode and g2 are nice and positive places. There is only one place where anions would prefer to go the other way: if they are located between cathode and negative g1. This a very small part of the total volume between cathode and anode where there are electrons flying around. This reduces the already low statistical chance on electron-gas atom collision even further.
Surely, over thousands of hours in operation some ions will form there and find their way to the cathode, causing some damage.
But are those minutes of warming up with HT applied making a significant addition to the total number of cathode impacts??
If you worry about that, just keep the cloud contained with more negative bias during warm-up. No need to hot switch high voltage, just low power negative.
If the problem is positive cations, they are actually attracted to the electron cloud, so unless the protective mechanism works by absorbing multiple electrons on its way through the cloud, gaining negative charge and subsequently being expelled from the cloud, this is not the problem.
So we are talking about negative ions.
Why are those anions going to the cathode at all? The anode and g2 are nice and positive places. There is only one place where anions would prefer to go the other way: if they are located between cathode and negative g1. This a very small part of the total volume between cathode and anode where there are electrons flying around. This reduces the already low statistical chance on electron-gas atom collision even further.
Surely, over thousands of hours in operation some ions will form there and find their way to the cathode, causing some damage.
But are those minutes of warming up with HT applied making a significant addition to the total number of cathode impacts??
If you worry about that, just keep the cloud contained with more negative bias during warm-up. No need to hot switch high voltage, just low power negative.
The effects I posted are from discussions with a tube designer from the Philips company. They were very much into reliability. Not that the tube destructs immediately; we know that isn't happening. But halving the life time of a $ 600 tube is of interest I think.
Some additional quotes from the interview:
Another phenomenon with a cathode not yet at operating temperature concerns unequal cathode temperature. As the cathode heats up, the temperature will not be uniform and the hotter areas will allow more current than the cooler areas. The current that flows will tend to concentrate in small hot areas, and that will increase the current density to higher values than will be the case at normal operation. That higher current density can locally damage the cathode creating miniature craters in the structure, which overall will have a negative effect on life expectancy.
All these factors suggest that the cathode should be well at operating temperature before the anode voltage is applied. This is especially important when the equipment uses solid state rectifiers in which case the anode voltage will be present almost immediately at turn-on. Although there are many amplifiers out there with solid state rectification that work reliably for many years, it is ultimately a factor in tube reliability deterioration
Edit: this is a good source too: Getting the most out of vacuum tubes, Robert B. Tomer, Howard W. Sams & Co, 1960; reprinted by KCK Media Corp. 2019, available at AudioXpress.com. It is all about tube reliability and circuit tricks to extend tube life and maintain long-term reliability as much as possible; it is NOT about designing amplifiers!
Jan
Some additional quotes from the interview:
Another phenomenon with a cathode not yet at operating temperature concerns unequal cathode temperature. As the cathode heats up, the temperature will not be uniform and the hotter areas will allow more current than the cooler areas. The current that flows will tend to concentrate in small hot areas, and that will increase the current density to higher values than will be the case at normal operation. That higher current density can locally damage the cathode creating miniature craters in the structure, which overall will have a negative effect on life expectancy.
All these factors suggest that the cathode should be well at operating temperature before the anode voltage is applied. This is especially important when the equipment uses solid state rectifiers in which case the anode voltage will be present almost immediately at turn-on. Although there are many amplifiers out there with solid state rectification that work reliably for many years, it is ultimately a factor in tube reliability deterioration
Edit: this is a good source too: Getting the most out of vacuum tubes, Robert B. Tomer, Howard W. Sams & Co, 1960; reprinted by KCK Media Corp. 2019, available at AudioXpress.com. It is all about tube reliability and circuit tricks to extend tube life and maintain long-term reliability as much as possible; it is NOT about designing amplifiers!
Jan
Last edited:
Interesting book, thanks!
I didn't find the ion bombardment yet, but the (localized) overcurrent is interesting.
Will study it more at a later time.
I didn't find the ion bombardment yet, but the (localized) overcurrent is interesting.
Will study it more at a later time.
Don't know what kind of fool wrote this article... Ions more massive...god, what a line of crap....... no such thing as "Ions striking the cathode." If this author has any kind of electrical engineering degree, he/she should take said "degree" back to that college ; and ask for their money back.......
-g
No tube is completely gas free. Gasses form ions. In moderately gassy tubes there is a distinct glow due to the ions. In and extremely gassy tube there will be a catastrophic cathode stripping event.
1. it's a hard vacuum. Would not work otherwise.
2. In my youth, I've worked on US Navy vacuum tube equipment(s). Only tube failure I ever saw was the filaments go weak and die, after years and years of service... never saw any so called cathode stripping.....
-g
2. In my youth, I've worked on US Navy vacuum tube equipment(s). Only tube failure I ever saw was the filaments go weak and die, after years and years of service... never saw any so called cathode stripping.....
-g
Jan, a little off topic but perhaps you can answer a question that has eluded me. Tubes will degrade over time based on their use eventually requiring replacement. No doubt this process is based on hours of use for a given tube. My question, is a tube 'wearing out' if it is powered but not playing music? IE if I turn on my amp for an hour to warm up, and then listen for an hour does that equal 2 hours of use wrt to rate of wear, or just the hour playing music? Sorry it that is a dumb question but I'd happily leave my amps on all afternoon in case I want to listen but hate to wear out the tubes doing it.
May i chime in ,
Degradation of emission is a purely hour-on issue, music or no music dont matter.
The "wear" process seems to be depletion of barium atoms in the outer cathode surface. These atoms evaporate into the vacuum due to heat and to a lesser part to damage caused by spallation and ion bombardment. Some barium diffuses from deeper layer but will eventually dry out causing a reducing of emission.
This is a slow process, typically several 1000 of hours is needed for any degradation to occur. Other processes such as heated without any current drawn may "poison" the cathode due to building up an isolating layers within the cathode ( this is why "stand-by" switches are bad if used )
So yes 2 hours of power on is 2 hours of degradation. On the other hand each power on is stressing the amp and tubes thus there is a trade off between keeping powered on or turning off between listening sessions.
I would not worry too much about power on time, keep it on !
Don't put too much faith in Tomer's book. It gets passed around the forums every few months but is really the work of an enthusiastic amateur rather than a scholarly work. It is simplistic and full of things that Tomer has (frankly) made up out of thin air, such as the graph in post 28. (Yes of course higher heater voltage increases the risk of heater burnout, but don't invent a graph with no axes or source data and pretend it means anything more than 'higher heater voltage increases the risk of heater burnout', which is both ambiguous and obvious to anyone who has ever used a light bulb).
If you want the straight dope you need to go to the academic journals. The memory of one old 'tube designer from the Philips company' is no match for hundreds of Philips own Technical Review papers, produced over half a century, which explain all this stuff in microscopic detail, and did not feel the need to recommend standby switches for receiving tubes.
If you want the straight dope you need to go to the academic journals. The memory of one old 'tube designer from the Philips company' is no match for hundreds of Philips own Technical Review papers, produced over half a century, which explain all this stuff in microscopic detail, and did not feel the need to recommend standby switches for receiving tubes.
Last edited:
Yes it is. It is wearing out whenever the cathode is heated, even if the anode current is completely switched off (e.g. standby).
Tube 'wear'
Thank you Petertub, Merlinb and Jan for clearing that up. It is as I suspected, but I have a clear understanding now. Much appreciated.
Walter
Thank you Petertub, Merlinb and Jan for clearing that up. It is as I suspected, but I have a clear understanding now. Much appreciated.
Walter
Actually, Philips trusted him with QC on tube production so they (Philips) assumed he knew his chops.
That said, I do agree that tube life and reliability data is a) hard to come by and b) often contradicts itself. I tried to separate the chaff from the corn.
Jan