Ahh well, it seems there is a degradation mechanism for directly heated cathodes.
813 SE triode amps
http://www.diyaudio.com/forums/tubes-valves/32368-filament-source-connection-dht.html
wireless.fcc.gov/commoperators/element6.doc
AC filament supply?
813 SE triode amps
http://www.diyaudio.com/forums/tubes-valves/32368-filament-source-connection-dht.html
wireless.fcc.gov/commoperators/element6.doc
AC filament supply?
There is a long, hard road to travel until you reach Maxwell's equations, once there, it's like video games, when you reach the last level and you are "immortal".
I assure you that the version posted on the "forum chit-chat" are quite similar to those published by Maxwell, in fact, have not changed since 1873.
However, in modern textbooks, confused the fields E and D, just write E, also confused the fields B and H, just write B.
Forget the system of units, I assure you that at the time of Maxwell, also used the cgs, only then called Gaussian.
I already showed, in post # 36, the results for the field B, which is responsible for actions at distance, and for the problem at hand would be enough, however I'm struggling to find an expression for the field H in the case of an AC current, so far I failed.
Almost everything you do with amps, and valves, with the exception of thermionic effect, and semiconductors, comes from Classical Electrodynamics, Maxwell's equations are his core; survived to the Theory of Relativity, since they are covariant at Lorentz transformations, they say stupid things about them, not just me, I ask please do not defames.
BTW, where you suggest that published, American Journal of Physics?
Not true, so far we are getting interesting conclusions, you do not understand is another thing.
Best regards
Johann
Last edited:
No probs my end - I was just suggesting that propositions and technical debate that starts from Maxwells equations is a little too surreal for this forum.
Even though I haven't used a DH tube, I now appreciate where the concern on DC supply polarity and life has originated. I had a quick look through reliability papers on DH tubes, which are quite numerous and intensive in the 1930-60's and focussed on transmitter applications, but all references viewed were related to AC heaters. Some references that identified moving to DC heaters to reduce residual performance spec issues didn't comment on life and weren't for 'normal' tubes.
I even saw a commercial preamp handbook that talked about swapping tubes to alternate the DC polarity - but the tubes were all indirectly heated cathode types - so I'm guessing that was just snake oil hype.
Maybe George wants to comment on the OP's tubelab.com reference.
Even though I haven't used a DH tube, I now appreciate where the concern on DC supply polarity and life has originated. I had a quick look through reliability papers on DH tubes, which are quite numerous and intensive in the 1930-60's and focussed on transmitter applications, but all references viewed were related to AC heaters. Some references that identified moving to DC heaters to reduce residual performance spec issues didn't comment on life and weren't for 'normal' tubes.
I even saw a commercial preamp handbook that talked about swapping tubes to alternate the DC polarity - but the tubes were all indirectly heated cathode types - so I'm guessing that was just snake oil hype.
Maybe George wants to comment on the OP's tubelab.com reference.
I haven't read any from those years, but I like to think, however I do it slowly.
In post # 40, I pulled an idea, although brief, In a DH cathode, no matter the form, Suppose that it has two terminals (1) and (2), the terminal (1) is at a potential V(1) = Vk, and we apply a DC voltage, V (DC), then the terminal (2) is at a potential
V(2) = Vk +/- V(DC)
Then with + sign we have
V(2) = Vk + V(DC) > Vk = V(1)
With - sign
V(2) = Vk - V(DC) < Vk = V(1)
The lower is the potential from ground, the greater the emission, from Maxwell's equations, but if you do not like, I see every day in CRTs.
Remember that I'm just a TV repairman.
In the first case, terminal (1) emits more than terminal (2)
Reversing the polarity in the second case is the other way round, terminal (2) emits more than terminal (1)
Note that between the terminals there is a certain distribution of emission depends on the geometry of the valve.
Ironically, from Maxwell's equations we derive Poisson's equation, from here is derived the Child-Langmuir law, which describes the behavior of the valves.
Poisson's equation can be solved in different coordinate systems, but in cylindrical coordinates is easier, theory and practice, due to the cylindrical symmetry we can almost ensure that our DH cathode has almost a linear distribution of emission, then a half emits more than the other.Reversing the polarity is the other way round.
Ugh ... I got tired, I owe the rest for another day.
Best regards
Johann
Last edited:
Weather you choose to use AC or DC to heat your tubes makes a completely insignificant difference on the life of the tube. Stop and think about the voltages involved for a minute or for our TV repair man 2 minutes. I mean come on:
At what percent of the emission capacity of the cathode are your tubes operating at using your chosen operating point?
How much does the difference in voltage from one end of the filament to the other effect those emissions?
So if DC heating lessens the life of your tube by 5% I would by very very surprised.
I am extremely disappointed in our TV repair man despite giving him days to come up with this conclusion.
Oh and as far as plating the cathode think about that too.
The original post was talking about PLATING the cathode. (ie. some elements that should not be there are there smothering the electron emissions) This has zero to do with maxwell at any level. It could be due to electrons being given to the cathode OR it could be due to electrons being taken away from the cathode causing the elements that are inside the tube to form bonds and somehow plate the freekin' cathode.
MelB's Theorem: The more knowledge you spew the less intelligent you are.
At what percent of the emission capacity of the cathode are your tubes operating at using your chosen operating point?
How much does the difference in voltage from one end of the filament to the other effect those emissions?
So if DC heating lessens the life of your tube by 5% I would by very very surprised.
I am extremely disappointed in our TV repair man despite giving him days to come up with this conclusion.
Oh and as far as plating the cathode think about that too.
The original post was talking about PLATING the cathode. (ie. some elements that should not be there are there smothering the electron emissions) This has zero to do with maxwell at any level. It could be due to electrons being given to the cathode OR it could be due to electrons being taken away from the cathode causing the elements that are inside the tube to form bonds and somehow plate the freekin' cathode.
MelB's Theorem: The more knowledge you spew the less intelligent you are.
Well I would initially think that the key voltage differential is from cathode to grid. So if my grid-cathode voltage was circa 5V more at one end of the cathode than say a mid-point 0V (based on the 10V 10A type transmitter valves that seem to have been used for diy), then I'd reckon that was pretty substantial, and would have a pretty substantial influence on the emission level at that end compared to the total average emission.
Thoriated tungsten filaments definitely wear out pro-rata to a first degree with AC current level. Transmitter reliability data and assessments definitely shows that lifetime character.
Thoriated tungsten filaments definitely wear out pro-rata to a first degree with AC current level. Transmitter reliability data and assessments definitely shows that lifetime character.
I answer only because, addition to being a TV repairman I am a gentleman.
In a CRT there are 25KV between cathode and anode, if you move cathode voltage even a few Volt, the change in emission is instantaneous and very noticeable, the screen turns the corresponding color.
Just in case, 25KV = 25000V, V = Volt, screen is not the screen grid.
CRTs come in two flavors: AC heater and DC heater.
A CRT is that big glass thing coming on old TV, Oh, BTW, it is also a valve.
Last edited:
Yes, and this produces an easily observable modulation in a filamentary (no separate cathode and heater) valve with AC heating supply. Here in the States with 60Hz power, plate current is modulated at 120Hz. Some - - I still run my type 845's on AC and live with the modulation.
Thanks,
Chris
I remind you that the potential of the grid also can refers to ground.
I always think of fixed bias, it is easier to think
The same with the potential of anode, so your reasoning is consistent with what I say.
Last edited:
The open sesame of valve aging is emission.
I've been thinking for some time how to explain in a simple way.
I took a lot more than a minute or two, sorry MelB, I think slowly.
In metals, at normal temperature, the conduction band is essentially filled of electrons only up to the Fermi energy EF, to extract an electron from the metal is therefore necessary to give a starting energy ei, but at high temperatures the occupation of electronic states extends above EF. If the temperature is high enough, some electrons reach energies greater than EF + ei, and escape from the metal.
The work function of the cathode can be made in the form
Wk= ( E - EF ) / e
where E is the total energy of electron, EF is the Fermi energy, e is the charge of electron.
We can increase the total energy of electron applying an electric field, and guess what, yes, Maxwell's equations.
So far I've only explained emission, but, how it affects aging?
Well, it happens that not only electrons are emitted, also ions, can also be demonstrated that there is a contact potential between cathode and anode, which are responsible for all those strange electrochemical effects mentioned by you.
There are other causes of aging, such as loss of vacuum, inlet hydrogen atoms (they are very small, and sometimes pass through the glass) and a long etc. but that's not the reason for this thread.
Moral: The mother of valve aging is emission.
You may not like my explanation, but what else can I do?
I'm just a TV repairman.
Best regards
Johann
I've been thinking for some time how to explain in a simple way.
I took a lot more than a minute or two, sorry MelB, I think slowly.
In metals, at normal temperature, the conduction band is essentially filled of electrons only up to the Fermi energy EF, to extract an electron from the metal is therefore necessary to give a starting energy ei, but at high temperatures the occupation of electronic states extends above EF. If the temperature is high enough, some electrons reach energies greater than EF + ei, and escape from the metal.
The work function of the cathode can be made in the form
Wk= ( E - EF ) / e
where E is the total energy of electron, EF is the Fermi energy, e is the charge of electron.
We can increase the total energy of electron applying an electric field, and guess what, yes, Maxwell's equations.
So far I've only explained emission, but, how it affects aging?
Well, it happens that not only electrons are emitted, also ions, can also be demonstrated that there is a contact potential between cathode and anode, which are responsible for all those strange electrochemical effects mentioned by you.
There are other causes of aging, such as loss of vacuum, inlet hydrogen atoms (they are very small, and sometimes pass through the glass) and a long etc. but that's not the reason for this thread.
Moral: The mother of valve aging is emission.
You may not like my explanation, but what else can I do?
I'm just a TV repairman.
Best regards
Johann
Last edited:
Reviews of transmitter valve life in the 1950's show that thoriated tungsten filaments have an emission aging process that is impacted by loss of carbon and thorium - carbon loss has been identified as the first to cause loss of emission and require valve replacement within measureable lifetimes. Oxide coated filaments had an immeasureably long lifetime, and valve failure was dominated by other processes such as loss of pressure.
But getting back to DC powered filaments, it amazes me that anybody would be stupid enough as to use DC powering, as only a small portion of the negative end of the filament would be doing anything useful
Ciao, Tim
But getting back to DC powered filaments, it amazes me that anybody would be stupid enough as to use DC powering, as only a small portion of the negative end of the filament would be doing anything useful
Ciao, Tim
Are you saying that emission has nothing to do with valve aging?
Thorium is a natural radioactive element, with a half-life in the order of 10^10 years.Tungsten is a stable element.
It may be that the loss of carbon occurs inside the box in a new valve, or escaping through cables.
All valves that I know have some degree of vacuum, must be why they are called vacuum tubes.
There are others that have some gas inside, but at higher pressure than atmospheric...no, I do not know.
But getting back to DC powered filaments, it amazes me that anybody would be stupid enough as to use DC powering, as only a small portion of the negative end of the filament would be doing anything useful
Ciao, Tim
Ah, now I see, you're kidding me.
It took time and effort to make the post # 64 at high school level.
The issue of the terminals was part of the reasoning.What I said was
"we can almost ensure that our DH cathode has almost a linear distribution of emission"
This means that the emission is not homogeneous, and I mean ALL THE CATHODE, not only a bit.
Best regards
Johann
No. Degradation of emission (by loss of carbon) was the service life endpoint. There is a fair bit of high-temperature chemistry happening between tungsten, carbon and thorium in the filaments - interesting stuff! The outcome is the need for +/-5% voltage tolerance on the filament to keep the temperature in the right range for the chemistry to work.Are you saying that emission has nothing to do with valve aging?
Okay, one more to the list.But it has nothing to do with AC / DC filament supply.
No doubt your information, but forgive me to be wary that emission has nothing to do with the electrochemical reaction, in addition there is high voltage in such monsters, pure chemical reaction is hard to believe with such fields.
But it is very surprising, most transmission triodes have thoriated tungsten filaments, even the most popular 211/VT4C and 845.
Best regards
Johann
many tube audiophiles said that the AC filament has more musicality then DC filament. for me, is the same. little hum or noise is not big problem with 90-92dB speakers because you cannot hear them at listening chair, maybe 2 meters nor more away from speakers.
Do any one help me to explain the claim above ?
Hi tcqanh
Ugh, I always turn to me, you'll think I'm a narcissist.
But in honor of the truth, I'll give an answer I think is correct.
I can not talk about musicality, because now I suffer tinnitus.
But I can talk about hum.
In the attachment in post # 50 are Maxwell's equations.
Equation (3) tells us that, to a time-varying electric field E(t) corresponds to a time-varying magnetic field B(t), and vice versa.
If you put an AC voltage=>AC electric field E(AC) on the filament, generate an AC magnetic field B(AC), which induces an AC electric field E(AC)=>AC voltage in the grid, you have hum.
There are many tricks to reduce it, but is always present.
Equation (4) tells us that, to a DC current i(DC) corresponds a DC magnetic field H(DC), as the medium (vacuum inside the valve) is homogeneous and isotropic, using the constitutive relationship (6), we obtain a DC magnetic field B(DC).
Now, using again equation (3) the resulting electric field E(DC)=0 , no hum.
Simple, right?
Best regards
Johann
Well if viewers have got this far then they may enjoy some background reliability review docs on transmitting valves - but nothing on DC filament operation, and definitely nothing on Maxwell's equations.
http://dalmura.com.au/projects/SOME FACTORS AFFECTING TRANSMITTING VALVE LIFE.pdf
http://dalmura.com.au/projects/HIGH-POWER TRANSMITTING VALVES WITH THORIATED FILAMENTS.pdf
http://dalmura.com.au/projects/SOME FACTORS AFFECTING TRANSMITTING VALVE LIFE.pdf
http://dalmura.com.au/projects/HIGH-POWER TRANSMITTING VALVES WITH THORIATED FILAMENTS.pdf
Ah, now I can see.Very interesting and enlightening.
In the temperature range of 2000°K-2500ºK, the Fermi energy level is widely exceeded, electrons escape from the material, also ions (evaporation)
I remind you that this phenomenon is called thermionic emission.
If you look at the spectrum of black body radiation, you'll see that even electromagnetic radiation is produced, through the infrared to the visible.Enlightening.
BTW, electromagnetic radiation was predicted by Maxwell's equations.
Best regards
Johann
Last edited:
Thoughts? Yeah, it looks like nonsense. For my first hollow state design, I opted for DC for heater power. That was in 2005, and I've been running the same prototype nearly every day for several hours, and I'm still waiting for my "common beam tetrode/pentodes" (807s) to go "a bit thin and weedy". Ain't happened yet, and there doesn't seem to be any measurable effect on bias (haven't had to adjust it since the 807s settled down) gain, distortion. Still sounds the same.
I started out as a solid state guy, and heard that heaters were a big source of hum. It just ain't so, and a subsequent project used AC for heating, and that from a floating 6.3Vrms winding to take care of a potential overvolting situation, and that project is dead quiet. Of course, I twisted the heater wire, and kept these as far from signal lines as possible. Used a semi-star grounding design with a ground lifter between the DC neutral and the grounded chassis. Good construction practices do more good for eliminating hum than anything else. (Also applies to solid state designs as well.)
No need for DC heating unless it's a design that handles very low level signals and is noise sensitive, so DC might be useful for small signal, high gain pents or cascodes. Once you get to substantial signal levels, not so useful.
Big difference there. For indirectly heated types, the filaments are specifically wound to cancel magnetic interference. Not possible for DH types where you need to distribute the filament across the area enclosed by the control grid. Otherwise, it won't work.
Hi Miles Prower
The purpose for my reasonings is not to offend anyone, but to help others understand a little more about one of the most common questions when designing valve equipment, and why not learn a little more myself.
In the attachment in post # 50 Maxwell's equations are general validity, the subsequent reasoning, was made for a perfect loop, but with a more sophisticated mathematics, can be extended to more general case in all regions of space.
However, for simple matters as those discussed in this thread, are perfectly valid.
In the case of using Ac for filament supply, for indirectly heated types, the filaments are specifically wound to minimize magnetic interference, however, Maxwell's equations are implacable, imagine a filament winding with N turns, from equation (21), (AC) magnetic field is divided by a factor N, but not canceled.
I mention this only to illustrate my point, I did not design valves.
In the grid, there is always an AC component, and we can not do anything to cancel, only attenuate it, using countless tricks developed since the invention of the first triode until today.
Is sometimes audible, sometimes not, sometimes it is measurable, others not, but is always present.
Best regards
Johann
The purpose for my reasonings is not to offend anyone, but to help others understand a little more about one of the most common questions when designing valve equipment, and why not learn a little more myself.
In the attachment in post # 50 Maxwell's equations are general validity, the subsequent reasoning, was made for a perfect loop, but with a more sophisticated mathematics, can be extended to more general case in all regions of space.
However, for simple matters as those discussed in this thread, are perfectly valid.
In the case of using Ac for filament supply, for indirectly heated types, the filaments are specifically wound to minimize magnetic interference, however, Maxwell's equations are implacable, imagine a filament winding with N turns, from equation (21), (AC) magnetic field is divided by a factor N, but not canceled.
I mention this only to illustrate my point, I did not design valves.
In the grid, there is always an AC component, and we can not do anything to cancel, only attenuate it, using countless tricks developed since the invention of the first triode until today.
Is sometimes audible, sometimes not, sometimes it is measurable, others not, but is always present.
Best regards
Johann
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.