Yes, the ST shape allowed the upper mica spacer to align with the glass. The bottom of those were aligned only by the wires in the base. It also provided a vibration path to the elements of the tube since the top glass is now coupled to the inner structure. What it did do was force vertical alignment to stabilize the structure's position in relation to all it's parts. Some of the globe and balloon tubes I have had over the years were not vertically aligned as closely as the ST and modern bottle shapes. The modern bottle allows mica spacers to align bottom and top of the envelope. Much better alignment.
True, all elements can contribute to microphonics. But look at the sensitivity for each element. By far the most sensitive space is the grid to cathode (or filament) distance. A "shoot from the hip" guess would be by the mu compared to the cathode (filament) to plate distance. Then look at the magnitude of the displacement from vibration. Compare that to the distance between the element to the cathode (filament). Once again, displacement is a far higher percentage between the grid (G1) and cathode (filament).
So when worrying about the effect of vibration (displacement from position at rest), you've got to consider how much that the effect is.
I might be an idiot, but it seems to me that the plate, be it mesh or plate metal, isn't really a large player in microphonics. I could be wrong, and I'm more than willing to find information that would point in some direction. By extension, whether the plate is mesh or a sheet of metal probably is not a concern for microphonic behavior. The use of a mesh might actually be something as simple as cost in low power applications. I don't know.
True, all elements can contribute to microphonics. But look at the sensitivity for each element. By far the most sensitive space is the grid to cathode (or filament) distance. A "shoot from the hip" guess would be by the mu compared to the cathode (filament) to plate distance. Then look at the magnitude of the displacement from vibration. Compare that to the distance between the element to the cathode (filament). Once again, displacement is a far higher percentage between the grid (G1) and cathode (filament).
So when worrying about the effect of vibration (displacement from position at rest), you've got to consider how much that the effect is.
I might be an idiot, but it seems to me that the plate, be it mesh or plate metal, isn't really a large player in microphonics. I could be wrong, and I'm more than willing to find information that would point in some direction. By extension, whether the plate is mesh or a sheet of metal probably is not a concern for microphonic behavior. The use of a mesh might actually be something as simple as cost in low power applications. I don't know.
You literally said "The major contributor to microphonics is the movement of the grid wrt the cathode or filament. Not the plate, mesh or otherwise." But it now looks like this is just your educated guess?
Hi stephe,
Do I have controlled tests? No, I don't and do not know of any. You don't either. Play nice.
Your position is clear. Cast doubt without offering anything that can be thought about or considered. It's real easy to throw stones without a position beyond "could be", "might be". All that is missing is "we don't know everything".
I'm not going to waste time with someone who is merely casting doubt in every direction.
Now, what I actually said was:
Now if anyone can extend my education and correct me on what I said, please do. I'll be grateful, but don't waste people's time by clouding the issue.
Do I have controlled tests? No, I don't and do not know of any. You don't either. Play nice.
Your position is clear. Cast doubt without offering anything that can be thought about or considered. It's real easy to throw stones without a position beyond "could be", "might be". All that is missing is "we don't know everything".
I'm not going to waste time with someone who is merely casting doubt in every direction.
Now, what I actually said was:
Your edit takes things a bit out of context, which isn't honest.
Now if anyone can extend my education and correct me on what I said, please do. I'll be grateful, but don't waste people's time by clouding the issue.
One unknown fact to me is from what type of metal is made those mesh anode plates ? , did that mesh metal surface is coated with some type of getter material ?
,and yes I also think that most of the unwanted microphonics become from interaction of hot cathode and grid , since the grid is relative close spaced to cathode as electrons emitter, compared to anode-cathode distance.
,and yes I also think that most of the unwanted microphonics become from interaction of hot cathode and grid , since the grid is relative close spaced to cathode as electrons emitter, compared to anode-cathode distance.
Hi stephe,
From long experience and training from other techs (mentors), I state what has been seen and proved over the years as factual. They are not personal opinion. If anyone has reasonable evidence to the contrary, I am willing to learn. I have never stated anything different. I have no interest in misleading anyone.
I have zero issue with intelligent questioning. This is not what you are doing. I have also stated the reasons for the views I share to give others an opportunity to think about it, and to check. A response should be similar. To question and try to open doubt on everything without any solid reasoning is not helpful to anyone.
If you have a problem with me, that's cool. But do not use that as a reason to muddy issues, everyone loses.
From long experience and training from other techs (mentors), I state what has been seen and proved over the years as factual. They are not personal opinion. If anyone has reasonable evidence to the contrary, I am willing to learn. I have never stated anything different. I have no interest in misleading anyone.
I have zero issue with intelligent questioning. This is not what you are doing. I have also stated the reasons for the views I share to give others an opportunity to think about it, and to check. A response should be similar. To question and try to open doubt on everything without any solid reasoning is not helpful to anyone.
If you have a problem with me, that's cool. But do not use that as a reason to muddy issues, everyone loses.
Hi banat,
The metals used in tube construction are inert and have been treated to remove gasses. The cathode is coated with materials that emit electronics at lower temperatures, transmitting tubes generally use pure metals run at much higher temperatures.
Plates are never coated with emissive materials. That would create something called secondary emission. In beam power tubes and true pentodes there is an element used to reduced this called the suppressor grid, in beam power tubes it is a space charge and not a physical grid.
So no matter if the plate is sheet metal, or a mesh, the metals used are not supposed to emit anything but radiant heat.
The getters hold a material that is heated and flashed onto the glass (silver deposits). That is designed to absorb gases from the metals in the tubes to maintain the vacuum and avoid ions that would damage the cathode surface. If you see that material becoming brown-ish, the effectiveness is probably reduced and the tube may become electrically leaky or "gassy". If that material is white, the tube seal is broken, so use care when handling it so you don't get cut from broken glass. They may break as you remove them from the socket if they aren't already.
The metals used in tube construction are inert and have been treated to remove gasses. The cathode is coated with materials that emit electronics at lower temperatures, transmitting tubes generally use pure metals run at much higher temperatures.
Plates are never coated with emissive materials. That would create something called secondary emission. In beam power tubes and true pentodes there is an element used to reduced this called the suppressor grid, in beam power tubes it is a space charge and not a physical grid.
So no matter if the plate is sheet metal, or a mesh, the metals used are not supposed to emit anything but radiant heat.
The getters hold a material that is heated and flashed onto the glass (silver deposits). That is designed to absorb gases from the metals in the tubes to maintain the vacuum and avoid ions that would damage the cathode surface. If you see that material becoming brown-ish, the effectiveness is probably reduced and the tube may become electrically leaky or "gassy". If that material is white, the tube seal is broken, so use care when handling it so you don't get cut from broken glass. They may break as you remove them from the socket if they aren't already.
anatech
of course that anode plates are never coated with some emissive materials , but maybe those mesh-type anode plates are made from some antimagnetic material ? , which in the same time can shield grid and other internal elements structure from some external magnetic field influence ?
of course that anode plates are never coated with some emissive materials , but maybe those mesh-type anode plates are made from some antimagnetic material ? , which in the same time can shield grid and other internal elements structure from some external magnetic field influence ?
Hi banat,
For shielding, you would normally use an external tube shield. Interesting concept.
Good thought on non-ferrous metal. I'm not sure how important that is, I would think electrostatic shielding is more important - coupled with good mechanical layout. I can check some tubes with a strong magnet. As a much younger fella, I do recall broken tube plates were ferrous. That doesn't mean they all are, just some did stick to a magnet (I was lazy cleaning up a mess I made!). Hardly scientific, just an observation.
I have a bulk eraser that can be used to demagnetize any tubes that might be attracted to the magnet.
For shielding, you would normally use an external tube shield. Interesting concept.
Good thought on non-ferrous metal. I'm not sure how important that is, I would think electrostatic shielding is more important - coupled with good mechanical layout. I can check some tubes with a strong magnet. As a much younger fella, I do recall broken tube plates were ferrous. That doesn't mean they all are, just some did stick to a magnet (I was lazy cleaning up a mess I made!). Hardly scientific, just an observation.
I have a bulk eraser that can be used to demagnetize any tubes that might be attracted to the magnet.
I think that even if a plate material was constructed from a ferrous alloy, the magnetic property wouldn't withstand the heat of normal operation. Heat destroys magnetism in metals as the atoms become energized and lose alignment. Plus, any ripple or signal current would make a pretty good degausser.
Last edited:
Microphony can be caused by any of the elements of the electrode assembly. Even the Getter!
This should not be a surprise, since lightweight metallic parts formed of thin material are liable to have sharp (like high Q) resonances. Even where these do not directly generate an electrical output, a high-Q mechanical oscillation can readily couple to grid.1, or any other feature than can.
And yes, it has been measured. Here is an example where modifications to the getter-attach (1 point vs 2-point) are compared in the response to vibration-table stimulus.
It should be carefully noted that these measurements are for miniature all-glass construction.
Large structures in the kind of valve that have mesh-anodes as an alternative (power Valves) will probably have more resonances - with higher "Q" for stiff material like Nickel sheet, less so for Mesh structures.
Reference: MULLARD RADIO VALVE Co, Ltd "Mullard Technical Communications № 61" LONDON: 1962 November.
This should not be a surprise, since lightweight metallic parts formed of thin material are liable to have sharp (like high Q) resonances. Even where these do not directly generate an electrical output, a high-Q mechanical oscillation can readily couple to grid.1, or any other feature than can.
And yes, it has been measured. Here is an example where modifications to the getter-attach (1 point vs 2-point) are compared in the response to vibration-table stimulus.
It should be carefully noted that these measurements are for miniature all-glass construction.
Large structures in the kind of valve that have mesh-anodes as an alternative (power Valves) will probably have more resonances - with higher "Q" for stiff material like Nickel sheet, less so for Mesh structures.
Reference: MULLARD RADIO VALVE Co, Ltd "Mullard Technical Communications № 61" LONDON: 1962 November.
Good work, thank you!
With the 1920s valves, the Pa was not stated, because is is assumed that you run only Class-A, and use the maximim anode voltage (180V) and ½ the maximum anode current (=25mA) = 4,5W
For capacitances, we can use the values of the Mazda AC/HL.
AEI Mazda bought Met-Vick (COSMOS) in 1928, and the AC/HL was the newer version of the COSMOS AC/ valve
Cag = 3,5pF
Ca-> all: 7pF (anode to all earth-referred electrodes)
Cg-> all 7pF
Check out the history of the AC/HL here: Link-AC/HL
Read about the famous 250 000 hour BBC Mazda AC/HL! Worth a read.
Last edited:
Hi Rod,
Yes! And any resonating element will transfer that energy into the entire structure. That may well excite the more sensitive elements. I guess the one question would be, by how much? This will probably vary between production runs and certainly tube type.
Plate structures are usually circular, or box shaped with stiffening ribs. They are probably not prone to vibrating nearly as much. You would also be talking about a higher frequency, very strong vibration. In a home you'd have more pressing concerns. In an aircraft or missile/rocket, it would be cause for concern.
Yes! And any resonating element will transfer that energy into the entire structure. That may well excite the more sensitive elements. I guess the one question would be, by how much? This will probably vary between production runs and certainly tube type.
Plate structures are usually circular, or box shaped with stiffening ribs. They are probably not prone to vibrating nearly as much. You would also be talking about a higher frequency, very strong vibration. In a home you'd have more pressing concerns. In an aircraft or missile/rocket, it would be cause for concern.
Anode assemblies are the element with the largest unsupported areas of thin metallic material in the electrode space. They are often assembled by methods designed for speed of manufacture, not rigidity. Actually, the cross section often looks like two bells facing each other. Ding Dong!
If you take the trouble to read the rest of the cited article, you'll see and example of a typical two-part anode vibrating most vigorously, due to a lack of overall rigidity.
I have another article assessing the microphony of a valve expressly designed to minimise it - not wholly a success.
The article only treated all-glass valves; others are likely to be worse, and sometimes much worse in this regard.
All of this points to the anode being a potential weak spot, in power valves even more so.
There are exceptions - EML has done a fine job of suppressing microphony in power DHTs by careful design touches. But it was a conscious effort - an example of design for purpose. They are aimed at modern designers of top rank audio, in the way that Marconi-Osram, and Western Electric did with theirs ninety odd years ago.
Hi Rod,
I was going to read the article after work today (thank you by the way).
I have taken so many tubes apart! The plate structure is light, the entire tube structure is designed to be light (all elements). Plate structures are tough and rigid. They are also in a vacuum. So vibration from the glass would transmit vibration through the mica spacers to the entire structure inside. You would have to hit the resonant frequency of any one structure to get it moving, a shock (pulse) represents all frequencies so it depends on damping too. Mice has motional losses against the edge of the fibers, it isn't like a metal. Grids are the least supported structures in a tube. They are more free to move on their own, and at lower frequencies. Maybe some early tubes had sloppy plate structures, or really poorly manufactured junk tubes, but we wouldn't use those for anything critical, would we?
Looking at the source of excitation, air has high losses, so not a lot of energy there as far as tube elements are concerned. Not compared to direct mechanical shock, such as tapping the glass envelope. That isn't a common situation, so we would be looking at mechanical vibration transmitted through the tube pins and base as being the worst case for causing tube elements to vibrate. Does this sound right? Therefore for high gain applications, like a mic or phono preamp, we would shock mount the socket, and may be the sub-chassis it is built on. If air vibrations are a concern, enclose the entire circuit in a box.
I'm just trying to focus on the real world implications of tube microphonics. I'm not trying to prove plate structure are or are not the cause. Common sense tells me that the grid to cathode relationship is the most critical for microphonics as well. I have said any information is good. I have to ask though. If the plate microphonic contribution is 1/100 of the grid to cathode, where should we be looking? Maybe it's 1/20 or even 1/10 with low mu tubes, it still isn't the major contributor. We typically don't use low mu tubes in input circuits anyway, and that is where all noise is amplified by the rest of the circuit. The most critical location for noise problems.
I was going to read the article after work today (thank you by the way).
I have taken so many tubes apart! The plate structure is light, the entire tube structure is designed to be light (all elements). Plate structures are tough and rigid. They are also in a vacuum. So vibration from the glass would transmit vibration through the mica spacers to the entire structure inside. You would have to hit the resonant frequency of any one structure to get it moving, a shock (pulse) represents all frequencies so it depends on damping too. Mice has motional losses against the edge of the fibers, it isn't like a metal. Grids are the least supported structures in a tube. They are more free to move on their own, and at lower frequencies. Maybe some early tubes had sloppy plate structures, or really poorly manufactured junk tubes, but we wouldn't use those for anything critical, would we?
Looking at the source of excitation, air has high losses, so not a lot of energy there as far as tube elements are concerned. Not compared to direct mechanical shock, such as tapping the glass envelope. That isn't a common situation, so we would be looking at mechanical vibration transmitted through the tube pins and base as being the worst case for causing tube elements to vibrate. Does this sound right? Therefore for high gain applications, like a mic or phono preamp, we would shock mount the socket, and may be the sub-chassis it is built on. If air vibrations are a concern, enclose the entire circuit in a box.
I'm just trying to focus on the real world implications of tube microphonics. I'm not trying to prove plate structure are or are not the cause. Common sense tells me that the grid to cathode relationship is the most critical for microphonics as well. I have said any information is good. I have to ask though. If the plate microphonic contribution is 1/100 of the grid to cathode, where should we be looking? Maybe it's 1/20 or even 1/10 with low mu tubes, it still isn't the major contributor. We typically don't use low mu tubes in input circuits anyway, and that is where all noise is amplified by the rest of the circuit. The most critical location for noise problems.
Common sense would dictate a large thin, flat metal plate would be much more susceptible to picking up and resonating vibrations than anything else inside a tube. Going from a solid to a mesh plate would radically reduce this. First you stated this as fact, then said it was an opinion, now are again saying to is proved as factual after saying there has been no testing that you know of about this.
I have no "issue" with you personally, I have no idea who you even are. What I have an issue with is someone flip flopping about what is fact, what isn't, claiming anything they post is factual but then having no proof or evidence.
The grid is also very "wiggly" and low in mass compared to the plate, my educated guess is that wiggly things will wiggle.