I prefer not to take a side in this discussion, since I don't think that in matters in a realistic audio amplifier, but... I will say that the above statement is true and I have observed this exact effect.
The tube in question was a 6AS7G which has two identical plates. If one is heated to red glow by excess dissipation, current will flow from it to the cold plate but not the other way. The effect is reversible by heating the other plate. If both plates are glowing red then current will flow in both directions. If the current from plate to plate is increased to the one amp region the plates will remain red and AC current will continue to flow from plate to plate even though the heaters and all other voltages have been removed from the tube. If allowed to proceed for several minutes there will be enough impurities released inside the tube to ionize. At this point we no longer have a "vacuum tube". The glass will soon melt and it will bulge outward due to a positive pressure inside!
You're right. Lower absolute cathode temp means lower emission thus lower max current. The freed electrons don't "care" at all (if they could 🙂 ) about the temperature of the anode they're flying towards. The infrared and other photons produced by the anode when it is hot have no measurable interactions with the moving electrons.
Kenneth
The anode is whatever terminal (conventional) current flows into, the cathode is where is leaves. The definition does not care how that current is established, temperature, potential difference, or other.
No, it's the potential difference across the electrodes. If you heat a filament rectifier and apply a very large reverse voltage, there will be a very small leakage current. Going from the cold to the hot electrode. How do you explain that?
No, temperature is not dependent on the work function.
Your point being?
The electrons in free space don't care about the temperature of the anode. They can't even feel it. A hot anode just produces some more secondary emission electrons, this is why the suppressor grid was invented.
This is in fact the major heat generator in the anode. Just bias your tube
very hot and see the anode glow 🙂
Yes, just a tiny bit, probably not worth the effort.
Kenneth
Last edited:
Interesting! I presume if there was a grid in between, one would have a sort of bi-directional triode.
Yes but such a device would be considerably less efficient than two separate devices connected reverse parallel. In "unidirectional" tubes there is only a small number of electrons (from secondary emission) "floating" around the anode, repelling flow from the cathode. In a bidirectional device you envisioned there would be a significant space charge cloud around each electrode which would impede current flow, making it less efficient (higher potential difference would be required to overcome the repulsion of space charge).
challenge
Wakibaki,
rather than continue arguing with you, I challenge you to produce a reference to a peer-reviewed paper or a textbook describing a mathematical formula for either cathode current or cathode emission which contains Ta, the plate temperature.
If what you say is true it should be easy to find.
Kenneth
Wakibaki,
rather than continue arguing with you, I challenge you to produce a reference to a peer-reviewed paper or a textbook describing a mathematical formula for either cathode current or cathode emission which contains Ta, the plate temperature.
If what you say is true it should be easy to find.
Kenneth
I don't post often but somehow I always manage to participate in the more controversial discussions here. Oh well. 😉
I'm not taking sides here but the best discussion on heat transfer within tubes that I've found to date is in the paper "Heat Transfer in Receiving Tubes" by O.H. Schade Jr. The paper can be found in the book "Electron Tube Design", Radio Corporation of America, Harrison, NJ, pp. 244-265, 1962.
The author of the paper is the same person associated with the "Schade Feedback" often discussed in this forum.
The entire "Electron Tube Design" book can be found on Pete Millett's excellent tubebooks.org - Vintage info from the age of vacuum tubes site. But note that the entire book is a 136MB download (Pete lists the book as an 84 MB download but the size is 136MB as it sits on my hard drive). So I hope I don’t cause any issues with Pete and the use of his web site by referencing that book from his site.
I took the liberty of extracting the paper from the Pete's pdf file of the book but the extracted file is over 6 MB and is still too large to upload here. Any suggestions on how I can post the extracted pdf? I hate using file share sites.
I'm not taking sides here but the best discussion on heat transfer within tubes that I've found to date is in the paper "Heat Transfer in Receiving Tubes" by O.H. Schade Jr. The paper can be found in the book "Electron Tube Design", Radio Corporation of America, Harrison, NJ, pp. 244-265, 1962.
The author of the paper is the same person associated with the "Schade Feedback" often discussed in this forum.
The entire "Electron Tube Design" book can be found on Pete Millett's excellent tubebooks.org - Vintage info from the age of vacuum tubes site. But note that the entire book is a 136MB download (Pete lists the book as an 84 MB download but the size is 136MB as it sits on my hard drive). So I hope I don’t cause any issues with Pete and the use of his web site by referencing that book from his site.
I took the liberty of extracting the paper from the Pete's pdf file of the book but the extracted file is over 6 MB and is still too large to upload here. Any suggestions on how I can post the extracted pdf? I hate using file share sites.
Last edited:
Point the fans at the transformers I think you will see a greater change.
As far as I know, we know this thanks to SY's uncle who had a amplifier company on the edge of the Nevada desert. As they would do listener testing late afternoon the outside temperature could change from +30c to -10c in just over one hour. And they were good enough to tell everyone of their findings (about 1955-57).
2nd electrons YES............
One thing that can be done like this is to maintain a more even heating from bottom to top of the valve this forms a greater area that the electrons can move from the cathode to the anode and therefore a greater current can flow. Although for my part I would think it is better to have the area of electron movement normal then it can move up and down in side the valve with outside temperature change and giving little change in sound output.
One more thing that may be good would be to cool the top of the valves but make the bottoms hot this will lower the area that the electrons past for the cathode to the anode.
This would be useful with followers, inversions, and maybe ultra path outputs as you will effectively have less wire length between components.
(but due to he vacuum this is hard to do).
As far as I know, we know this thanks to SY's uncle who had a amplifier company on the edge of the Nevada desert. As they would do listener testing late afternoon the outside temperature could change from +30c to -10c in just over one hour. And they were good enough to tell everyone of their findings (about 1955-57).
2nd electrons YES............
One thing that can be done like this is to maintain a more even heating from bottom to top of the valve this forms a greater area that the electrons can move from the cathode to the anode and therefore a greater current can flow. Although for my part I would think it is better to have the area of electron movement normal then it can move up and down in side the valve with outside temperature change and giving little change in sound output.
One more thing that may be good would be to cool the top of the valves but make the bottoms hot this will lower the area that the electrons past for the cathode to the anode.
This would be useful with followers, inversions, and maybe ultra path outputs as you will effectively have less wire length between components.
(but due to he vacuum this is hard to do).
You have misspelled my alias.
I think you'll find that any paper detailing thermionic emission treats the phenomenon as one affecting metals at elevated temperatures, not Anodes, Cathodes or even Odes to Joy. You persist in thinking of this in terms of electronics, which is a subset of my frame of reference, general physics.
A semiconductor device operates regardless of heat distribution in the device. It requires heatsinking, but not to preserve any temperature differentials.
A thermionic diode (the simplest kind of valve) operates because a temperature differential is maintained between the electrodes. There are many other becauses; but this is the prime because. This is where you start when explaining a valve to a child, children. Because of this, one 'electrode' (the hot one) emits (a lot) more 'electrons' than the other. Because of this, if you apply an alternating current, much more current flows in one direction than another. Because of this a diode can be treated as a one-way device. Di means two. The di-ode has 2 odes, the an-ode and the cat-ode. 2) The triode. The tri-ode is an adaption of the diode, it has 3 odes, but one is called the grid, because the word 'grode' was already in use, and nobody had heard of 'operator overloading' in those days.
All this stuff about absolute zero and the rest is just a desperate attempt to pick some tiny semantic hole in the picture that I have painted by people whose obvious intent is the obfuscation and mystification of science and technology, and whose private motivations are, IMO, more about their own self-aggrandisement (in the accomplishment of which which my humiliation would be a mere necessity) than any desire to assist or further the education of correspondents.
Look at the measured response from tubelab. If he could readily falsify my assertions I am sure he would have put me out of your misery by now.
w, m0ron
Rollover Beethoven...
You keep saying a cold anode is needed for a tube to work, yet you fail to give us any reference describing the relationship between Ia and Ta.
You incorrectly try to explain basic vacuum tube operation from your own intuition. Instead, read some of the old textbooks on basic VT operation and get a clue.
Repeating yourself over and over doesn't make what you say any more true.
It is not my or anyone else's job to falsify your 'assertions'. You are the one coming up with a new and previously unheard of claim; you need to provide the references to back it up. Your failure to do so makes it clear you are talking through your nose.
Thanks for smudging the diyAudio archives with misinformation.
Kenneth
This is the statement that has been made:-
A thermionic diode (the simplest kind of valve) operates because a temperature differential is maintained between the electrodes. This is the prime cause. Because of this (difference in temperature), one electrode (the hot one) emits (a lot) more electrons than the other. Because of this, if you apply an alternating current, much more current flows in one direction than another. Because of this a diode can be treated as a one-way device.
Is there anybody else who wishes to disagree with this statement? Either on the basis of the English in which it is couched, or the underlying concept?
Anybody?
w
Does anybody else have the gonads to stand up and say that it's fundamentally correct?
A thermionic diode (the simplest kind of valve) operates because a temperature differential is maintained between the electrodes. This is the prime cause. Because of this (difference in temperature), one electrode (the hot one) emits (a lot) more electrons than the other. Because of this, if you apply an alternating current, much more current flows in one direction than another. Because of this a diode can be treated as a one-way device.
Is there anybody else who wishes to disagree with this statement? Either on the basis of the English in which it is couched, or the underlying concept?
Anybody?
w
Does anybody else have the gonads to stand up and say that it's fundamentally correct?
Here's another statement that has also been made:
And another that you're repeating ad nauseam:
Current flows precisely because work function of [at least] one electrode has been overcome and sufficient potential difference has been applied to draw the freed electrons towards the other electrode (which includes overcoming any repulsion that might occur for whichever reason at the anode, be it secondary emission, heating up, radioactivity, etc.).
I guess only somebody signing his posts "M0ron" would do that.
Take a diode, say one of those with a top cap for easy access to the anode, blow compressed air across its top cap in order to establish temperature differentiral (which - unlike work function - seems to be what matters according to you) and let us know how it performs. If you have access to better cooling mechanisms (liquid nitrogen or at the very least dry ice) you'll undoubtedly be able to improve the performance alot further by increasing the differential 😀
Wakibaki,
unless you come up with a reference, don't bother to reply. You'd just be repeating yourself 😀
unless you come up with a reference, don't bother to reply. You'd just be repeating yourself 😀
Nah, my opinion was, and still is, that within reason (no LN2) you MIGHT be able to slightly extend tube life by cooling. I don't believe that reasonable cooling efforts will significantly change the operating parameters of a conventional (air cooled) vacuum tube. I may perform an experiment some day to find out if water cooling the glass envelope can remove enough thermal energy to change the power dissipation at which the plate starts to show color. These were the questions that the thread starter asked.
As to the arguments about work functions and temperature differentials, I don't have an opinion, and really don't care either. None of this information is important to me in my use and application of vacuum tubes.
I have worked in the same electronics plant for 37 years. When I started, there were still at least one product being manufactured there that used vacuum tubes. Now there is no more manufacturing, just engineering. I work in an advanced development group. I am surrounded by IC designers, but it is my job to design methods and fixtures for evaluating the prototype chips, and now design a prototype product using that chip. Do I need to know the detailed interaction going on among the zillions of individual circuits inside the chip? NO, we have 40 guys doing that. I just need to know how the chip interacts with the outside world. Any more knowledge would just cloud my already overloaded brain.
I am sure that the plate needs to be colder than the cathode, if for no other reason than to avoid melting the glass. There are tubes designed to operate with the plate glowing red since the gettering material is on the plate. The 833A is one of these.
I have observed that the plate in a conventional receiving tube can be operated well into the glow region (white hot) without seriously affecting the operation of the tube. The tube shown below was cranked to near arc over yet the distortion was still in the normal range. It must be noted that the materials used in ordinary receiving tubes are not meant to operate at these temperatures, so impurities will be released that will contaminate the vacuum in a short time.
Attachments
George, do you reckon that cooling a tube could be beneficial to the seals between glass and pins? Have you ever had a tube become gassy because of overheating?
BTW, I agree with you 100% that having an anode at typical cathode temperatures would instantly melt the tube, socket, and pcb. But it is not a fundamental problem. We could take one half 12AX7 and mount it in a glass envelope of, say, a GM100, heat all the insides to the same temperature as the cathode, and it will work.
Kenneth
BTW, I agree with you 100% that having an anode at typical cathode temperatures would instantly melt the tube, socket, and pcb. But it is not a fundamental problem. We could take one half 12AX7 and mount it in a glass envelope of, say, a GM100, heat all the insides to the same temperature as the cathode, and it will work.
Kenneth
The "prime cause" huh. Well that's great. This is like saying the wheels on a car are the "prime cause" that it works. I agree the car would stop going anywhere without them. Of course having an engine in the car helps too, wouldn't get anywhere without that. A transmission to connect them together is good too. Maybe throw in a steering wheel, but yeah why not just pick one part and say "This is the whole reason a car can take us anywhere. Anyone who doesn't look at it this way is crazy."
This thought experiment of yours is good:
wakibaki said:
If both elements in a two element tube become heated it ceases to be meaningful? Well that's handy. If you had a tube with two hot cathodes, at the same temperature, but with a voltage difference what do you suppose would happen? Current flow without a temperature gradient?
Sorry to the mods that my billy madison quote was considered inappropriate. Personally I'd rather be participating in a forum that could tolerate funny movie quotes but didn't tolerate trolling, but I don't make the rules and I know you guys have to do what you think is best for the forum.
Yes, to the first question. Some high power glass transmitting tubes require heat sinks on the pins and a minimum air flow across the seals. Is this necessary for conventional receiving tubes? What about situations like Fender guitar amps with the tubes mounted upside down? Fenders eat tubes, but so do most guitar amps. I don't really have the answer, but when I made guitar amps, I made them "right side up" like a Marshall. This has more to do with cap life than tube life though.
Yes, to the second question, but I don't believe that it had anything to do with the seals. When the plates light up like those shown above some of the metal itself will become dislodged and ionized, as will any imputities trapped within. These become "gas" and contaminate the vacuum. Eventually (or quickly in this case) the contamination causes grid current which raises the distortion, and will lead to tube runaway and catastrophic failure.
I purchased a NOS bulk pack of 100 Sylvania 6V6GTA's at a military surplus auction about 15 years ago. The box was still sealed and had never been opened. It was one of several boxes (different types) that I got that day. Upon opening the box I discovered that about 15 of the tubes had white getters, and others had lost their shine. I tested all the tubes in a guitar amp and found even more that showed signs of gas. All of the bad tubes were discarded and the rest went on a shelf inside my closet. To my dismay a few years later more tubes had gone bad.
About a year ago I got the box out again and yes more of the tubes had leaked. I "executed" some of the leakers in a Simple SE, and used a few more for some "20 watts in SE" type experiments. As of that day less than half of the tubes remain even though they have never been used. Some tubes just leak. Heat is not a factor here. Is this contamination in manufacturing? That is my thought. On the other hand the output tubes that are currently in my Tubelab SE were made in 1929, they still work great, so some tubes don't leak.
The "experiments" were here:
http://www.diyaudio.com/forums/tubelab/124527-photos-death-simple-se-6v6s.html
Actually, to me, it reads a lot like like saying that an internal combustion engine generates power due to the temperature differential between the hot gas in the combustion chamber and the colder gases in the exhaust manifold. After all, how much power would an engine generate if the exhaust system was heated to 3000F ;-)
Last edited:
😀 No that and everything else is a secondary reason Michael and totally unimportant according to my royal proclamation. In order to generate that temperature differential oil had to be extracted from the ground and highly refined into a fuel source and that oil wouldn't be there if massive amounts of plankton and algae didn't die millions of years ago. None of that would have lived if it weren't for the sun shining down on the earth so the main thing that allows us to drive around today is the Sun. It's the whole reason cars work and anyone who doesn't think so is a child.
- Status
- Not open for further replies.