The real question here is "When is not-a-diode a diode?" Randy Yates wrote to appreciate my old audio war stories. Here's another, with background. Urm ... foreground. Thermionic diodes are quirky devices. electrons are "boiled off" a hot electrode, the tube's cathode. The energy (in electron volts) needed to remove an electron from the surface is called the /work function/. Good cathode materials have low work functions. (Materials used for tube structures that shouldn't emit despite being heated -- the anodes of transmitter tubes running at red heat are an example -- are made of high-work-function materials such as zirconium or tantalum. Tungsten is a fairly decent emitter at 4.5 ev, and while thorium is only modestly better at 3.4 ev, thoriated tungsten after forming is considerably better at 2.6 ev. Cathodes can be heated directly by running a current through them or indirectly. (The cathode of one kerosene-powered radio tube made in Russia a was heated by an external flame.) Receiving tube cathodes are usually heated indirectly; nickel tubes with heaters in them, coated with a mixture of barium and strontium oxides. The work function is so low that such a tube can function below red heat. Directly heated cathodes are called filaments; indirectly heated ones, simply cathodes. Indirectly heated cathodes have several advantages. The main one for audio is hum reduction. Even with a center-tapped filament supply, having 6V AC in the amplifier circuit is at best a bother. The designer's job is much easier with an isolated heater. There is still some capacitive coupling, but center tapping the supply helps there too and even better is a potentiometer bridging an non-center-tapped supply with it's wiper grounded. with careful design and layout, this works well enough for all applications except a magnetic-cartridge preamp and good woofers during quiet passages. The remaining hum drove us crazy. We shielded the heater wires and the tubes themselves. One maker, H.H.Scot of dynamic-noise-suppressor fame, solved by using the output stage's plate current to power the heater of the first stage. (There was a drawback. If the tube -- a 6SJ7 -- was removed, the cathode bypass capacitor would charge up and blow out the tube when it was plugged beck in. Live and learn! I bitched to Herman, gut he just said, "It's worth it until someone shows me another way.) So how does all of this relate to diodes? Remember those AC-powered heaters? They can emit electrons too. Vacuum tubes generally require a negative bias on the control grid that is typically achieved by putting a resistor (often bypasssed for AC) between the cathode and ground. That makes the cathode appropriately positive relative to the grid. The positive cathode attracts/accepts electrons from the heater. Because the heater supply is (more or less) center tapped, the current is double-frequency, easily confused with hum from a full-wave-rectified plate supply. We don't usually think of the heater-cathode configuration as a diode, but it is one. Connecting the center tap or pot wiper to the cathodes of the output stage, generally at least 30V, back-biases all the other diodes in the system. I showed that to Herman and his next amplifier was more technician friendly. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
OT: When is a diode not a diode? (Confessions of a tube jockey.)
Started by ●November 30, 2008
Reply by ●November 30, 20082008-11-30
On Nov 30, 5:03�pm, Jerry Avins <j...@ieee.org> wrote:> The real question here is "When is not-a-diode a diode?" Randy Yates > wrote to appreciate my old audio war stories. Here's another, with > background. Urm ... foreground. > > Thermionic diodes are quirky devices. electrons are "boiled off" a hot > electrode, the tube's cathode. The energy (in electron volts) needed to > remove an electron from the surface is called the /work function/. Good > cathode materials have low work functions. (Materials used for tube > structures that shouldn't emit despite being heated -- the anodes of > transmitter tubes running at red heat are an example -- are made of > high-work-function materials such as zirconium or tantalum. Tungsten is > a fairly decent emitter at 4.5 ev, and while thorium is only modestly > better at 3.4 ev, thoriated tungsten after forming is considerably > better at 2.6 ev. > > Cathodes can be heated directly by running a current through them or > indirectly. (The cathode of one kerosene-powered radio tube made in > Russia a was heated by an external flame.) Receiving tube cathodes are > usually heated indirectly; nickel tubes with heaters in them, coated > with a mixture of barium and strontium oxides. The work function is so > low that such a tube can function below red heat. Directly heated > cathodes are called filaments; indirectly heated ones, simply cathodes. > > Indirectly heated cathodes have several advantages. The main one for > audio is hum reduction. Even with a center-tapped filament supply, > having 6V AC in the amplifier circuit is at best a bother. The > designer's job is much easier with an isolated heater. There is still > some capacitive coupling, but center tapping the supply helps there too > and even better is a potentiometer bridging an non-center-tapped supply > with it's wiper grounded. with careful design and layout, this works > well enough for all applications except a magnetic-cartridge preamp and > good woofers during quiet passages. > > The remaining hum drove us crazy. We shielded the heater wires and the > tubes themselves. One maker, H.H.Scot of dynamic-noise-suppressor fame, > solved by using the output stage's plate current to power the heater of > the first stage. (There was a drawback. If the tube -- a 6SJ7 -- was > removed, the cathode bypass capacitor would charge up and blow out the > tube when it was plugged beck in. Live and learn! I bitched to Herman, > gut he just said, "It's worth it until someone shows me another way.) > > So how does all of this relate to diodes? Remember those AC-powered > heaters? They can emit electrons too. Vacuum tubes generally require a > negative bias on the control grid that is typically achieved by putting > a resistor (often bypasssed for AC) between the cathode and ground. That > makes the cathode appropriately positive relative to the grid. The > positive cathode attracts/accepts electrons from the heater. Because the > heater supply is (more or less) center tapped, the current is > double-frequency, easily confused with hum from a full-wave-rectified > plate supply. > > We don't usually think of the heater-cathode configuration as a diode, > but it is one. Connecting the center tap or pot wiper to the cathodes of > the output stage, generally at least 30V, back-biases all the other > diodes in the system. I showed that to Herman and his next amplifier was > more technician friendly. > > Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������Jerry, interesting trip down memory lane I used to repair HH Scott equipment as a high school and college job. Here's a question about tubes I never really got a good answer for. Some tubes developed what was called grid emmission. If the grid was in a high z circuit, after the tube got hot, the grid would start to go positive even though it had a resistor to ground. This would throw the bias off and cause all sorts of drift and grief in a scope amplifer for example. I had an Eico scope I think with 6U8s and they were always going leaky. Were the tubes bad or was the ciruit bad? It seemed to have something to do with heat, it never happended when the tubes were cool. The 6U8s seemed prone to it. Mark
Reply by ●November 30, 20082008-11-30
Mark wrote: ...> Here's a question about tubes I never really got a good answer for. > Some tubes developed what was called grid emmission. If the grid was > in a high z circuit, after the tube got hot, the grid would start to > go positive even though it had a resistor to ground. This would throw > the bias off and cause all sorts of drift and grief in a scope > amplifer for example. I had an Eico scope I think with 6U8s and they > were always going leaky. Were the tubes bad or was the ciruit bad? It > seemed to have something to do with heat, it never happended when the > tubes were cool. The 6U8s seemed prone to it.That happens when some of the cathode coating material gets on the grid. The coating's work function is low enough to emit significantly at temperatures that the grid can reach by radiant heating from the cathode. It's a manufacturing defect not uncommon with Sylvania (also known with Westinghouse) but I never heard of it in an RCA tube. The quick fix is lowering the value of the grid leak. The emission current is limited. "Dirty" cathodes were another manufacturing defect that rarely caused a problem, but were the source of a near disaster for McIntosh at the first Boston Audio Show around 1952. (Gordon Gow wanted me to join the company after I solved his mystery. Dumb luck.) I know someone will ask. I'll write it up some other time. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●December 1, 20082008-12-01
On Dec 1, 3:45 pm, Jerry Avins <j...@ieee.org> wrote:> Mark wrote: > > ... > > > Here's a question about tubes I never really got a good answer for. > > Some tubes developed what was called grid emmission. If the grid was > > in a high z circuit, after the tube got hot, the grid would start to > > go positive even though it had a resistor to ground. This would throw > > the bias off and cause all sorts of drift and grief in a scope > > amplifer for example. I had an Eico scope I think with 6U8s and they > > were always going leaky. Were the tubes bad or was the ciruit bad? It > > seemed to have something to do with heat, it never happended when the > > tubes were cool. The 6U8s seemed prone to it. > > That happens when some of the cathode coating material gets on the grid. > The coating's work function is low enough to emit significantly at > temperatures that the grid can reach by radiant heating from the > cathode. It's a manufacturing defect not uncommon with Sylvania (also > known with Westinghouse) but I never heard of it in an RCA tube. The > quick fix is lowering the value of the grid leak. The emission current > is limited. > > "Dirty" cathodes were another manufacturing defect that rarely caused a > problem, but were the source of a near disaster for McIntosh at the > first Boston Audio Show around 1952. (Gordon Gow wanted me to join the > company after I solved his mystery. Dumb luck.) I know someone will ask. > I'll write it up some other time. > > Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������Did Fleming not discover the diode by accident after reversing the supply on a lamp of some sort? Hardy
Reply by ●December 1, 20082008-12-01
>On Dec 1, 3:45 pm, Jerry Avins <j...@ieee.org> wrote: >> Mark wrote: >> >> ... >> >> > Here's a question about tubes I never really got a good answer for. >> > Some tubes developed what was called grid emmission. If the gridwas>> > in a high z circuit, after the tube got hot, the grid would start to >> > go positive even though it had a resistor to ground. This wouldthrow>> > the bias off and cause all sorts of drift and grief in a scope >> > amplifer for example. I had an Eico scope I think with 6U8s andthey>> > were always going leaky. Were the tubes bad or was the ciruit bad?It>> > seemed to have something to do with heat, it never happended whenthe>> > tubes were cool. The 6U8s seemed prone to it. >> >> That happens when some of the cathode coating material gets on thegrid.>> The coating's work function is low enough to emit significantly at >> temperatures that the grid can reach by radiant heating from the >> cathode. It's a manufacturing defect not uncommon with Sylvania (also >> known with Westinghouse) but I never heard of it in an RCA tube. The >> quick fix is lowering the value of the grid leak. The emission current >> is limited. >> >> "Dirty" cathodes were another manufacturing defect that rarely causeda>> problem, but were the source of a near disaster for McIntosh at the >> first Boston Audio Show around 1952. (Gordon Gow wanted me to join the >> company after I solved his mystery. Dumb luck.) I know someone willask.>> I'll write it up some other time. >> >> Jerry >> -- >> Engineering is the art of making what you want from things you canget.>>=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=>=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF= >=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF > >Did Fleming not discover the diode by accident after reversing the >supply on a lamp of some sort? > >HardyDunno about that, but I when I was at UCL I saw a couple of Fleming's prototype diodes. I remember they looking an awful lot like a light bulb with an additional electrode. I guess he had a light bulb maker produce them for him. They ran one up, using modern passive components, for an anniversary of some sort, receiving BBC Radio 1. What kind of light bulb would have had a suitable electrode to function as an anode? If Fleming didn't set out to make a diode, I guess he must have been using custom devices to experiment with electron emission when he caught on to the potential of what he'd made, and we currently benefit from that. :-\ Regards, Steve
Reply by ●December 1, 20082008-12-01
HardySpicer wrote:> On Dec 1, 3:45 pm, Jerry Avins <j...@ieee.org> wrote: >> Mark wrote: >> >> ... >> >>> Here's a question about tubes I never really got a good answer for. >>> Some tubes developed what was called grid emmission. If the grid was >>> in a high z circuit, after the tube got hot, the grid would start to >>> go positive even though it had a resistor to ground. This would throw >>> the bias off and cause all sorts of drift and grief in a scope >>> amplifer for example. I had an Eico scope I think with 6U8s and they >>> were always going leaky. Were the tubes bad or was the ciruit bad? It >>> seemed to have something to do with heat, it never happended when the >>> tubes were cool. The 6U8s seemed prone to it. >> That happens when some of the cathode coating material gets on the grid. >> The coating's work function is low enough to emit significantly at >> temperatures that the grid can reach by radiant heating from the >> cathode. It's a manufacturing defect not uncommon with Sylvania (also >> known with Westinghouse) but I never heard of it in an RCA tube. The >> quick fix is lowering the value of the grid leak. The emission current >> is limited. >> >> "Dirty" cathodes were another manufacturing defect that rarely caused a >> problem, but were the source of a near disaster for McIntosh at the >> first Boston Audio Show around 1952. (Gordon Gow wanted me to join the >> company after I solved his mystery. Dumb luck.) I know someone will ask. >> I'll write it up some other time. >> >> Jerry >> -- >> Engineering is the art of making what you want from things you can get. >> ����������������������������������������������������������������������� > > Did Fleming not discover the diode by accident after reversing the > supply on a lamp of some sort?Edison discovered the diode, but didn't realize what he had. (See "Edison effect". The Edison effect led J.J.Thompson to discover the electron.) An Englishman (not Fleming) had the patents on a diode, then Fleming got into the picture somehow. Don't forget Lee DeForest (who wrote of commercial radio, "This is DeForest's prime evil ..."). Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●December 1, 20082008-12-01
On Dec 1, 12:15�pm, Jerry Avins <j...@ieee.org> wrote:> HardySpicer wrote: > > On Dec 1, 3:45 pm, Jerry Avins <j...@ieee.org> wrote: > >> Mark wrote: > > >> � �... > > >>> Here's a question about tubes I never really got a good answer for. > >>> Some tubes developed what was called grid emmission. �If the grid was > >>> in a high z circuit, after the tube got hot, the grid would start to > >>> go positive even though it had a resistor to ground. �This would throw > >>> the bias off and cause all sorts of drift and grief in a scope > >>> amplifer for example. �I had an Eico scope I think with 6U8s and they > >>> were always going leaky. �Were the tubes bad or was the ciruit bad? It > >>> seemed to have something to do with heat, it never happended when the > >>> tubes were cool. �The 6U8s seemed prone to it. > >> That happens when some of the cathode coating material gets on the grid. > >> The coating's work function is low enough to emit significantly at > >> temperatures that the grid can reach by radiant heating from the > >> cathode. It's a manufacturing defect not uncommon with Sylvania (also > >> known with Westinghouse) but I never heard of it in an RCA tube. The > >> quick fix is lowering the value of the grid leak. The emission current > >> is limited. > > >> "Dirty" cathodes were another manufacturing defect that rarely caused a > >> problem, but were the source of a near disaster for McIntosh at the > >> first Boston Audio Show around 1952. (Gordon Gow wanted me to join the > >> company after I solved his mystery. Dumb luck.) I know someone will ask. > >> I'll write it up some other time. > > >> Jerry > >> -- > >> Engineering is the art of making what you want from things you can get. > >> ����������������������������������������������������������������������� > > > Did Fleming not discover the diode by accident after reversing the > > supply on a lamp of some sort? > > Edison discovered the diode, but didn't realize what he had. (See > "Edison effect". The Edison effect led J.J.Thompson to discover the > electron.) An Englishman (not Fleming) had the patents on a diode, then > Fleming got into the picture somehow. Don't forget Lee DeForest (who > wrote of commercial radio, "This is DeForest's prime evil ..."). > > Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������- Hide quoted text - > > - Show quoted text -Hello Jerry, It is cool that you got to "play" with vacuum tubes during this era. When I came along, tubes were very specific to their tasks and transistors were coming on strong in designs. I recall a problem with using DC for directly heated cathodes, was the nonuniform erosion of the filiment since it had a voltage drop along its length. I think this was a primary reason behind the use of indirectly heated cathodes. Of course heated metals in a vacuum like to evaporate and at high tube currents the low work function material could redeposit on cooler surfaces like the anode harming the one wayness of the diode. Clay
Reply by ●December 1, 20082008-12-01
clay@claysturner.com wrote: ...> It is cool that you got to "play" with vacuum tubes during this era. > When I came along, tubes were very specific to their tasks and > transistors were coming on strong in designs. I recall a problem with > using DC for directly heated cathodes, was the nonuniform erosion of > the filiment since it had a voltage drop along its length. I think > this was a primary reason behind the use of indirectly heated > cathodes. Of course heated metals in a vacuum like to evaporate and at > high tube currents the low work function material could redeposit on > cooler surfaces like the anode harming the one wayness of the diode.I still have an operable U.S. Army Signal Corps VT-1. According to what I had been told, ten of them were used in a field-portable voice radio set. The receiver was a two-stage TRF amplifier followed by a plate detector and headset amplifier. The transmitter was an oscillator and push-pull modulated amplifier. The modulator itself used one audio gain stage driving a push-pull modulator. One man carried the transceiver, another, the batteries. My father bought the tube as surplus after WW I for $10 -- a week's pay for unskilled labor -- to replace the diode/cat's whisker in his crystal set. He thought my using it to build a regenerative receiver was neat. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●December 1, 20082008-12-01
Jerry Avins wrote:> clay@claysturner.com wrote: > > ... > >> It is cool that you got to "play" with vacuum tubes during this era. >> When I came along, tubes were very specific to their tasks and >> transistors were coming on strong in designs. I recall a problem with >> using DC for directly heated cathodes, was the nonuniform erosion of >> the filiment since it had a voltage drop along its length. I think >> this was a primary reason behind the use of indirectly heated >> cathodes. Of course heated metals in a vacuum like to evaporate and at >> high tube currents the low work function material could redeposit on >> cooler surfaces like the anode harming the one wayness of the diode. > > > > I still have an operable U.S. Army Signal Corps VT-1. According to what > I had been told, ten of them were used in a field-portable voice radio > set. The receiver was a two-stage TRF amplifier followed by a plate > detector and headset amplifier. The transmitter was an oscillator and > push-pull modulated amplifier. The modulator itself used one audio gain > stage driving a push-pull modulator. One man carried the transceiver, > another, the batteries. > > My father bought the tube as surplus after WW I for $10 -- a week's pay > for unskilled labor -- to replace the diode/cat's whisker in his crystal > set. He thought my using it to build a regenerative receiver was neat. > > JerryMemories ... My father used to tell stories of the "good old days" ;) Made his own capacitors in a "fish tank". They got about one micro-farad/gallon. He had also operated a _legal_ land based spark gap transmitter. Those interested may wish to browse http://www.antiquewireless.org . When I visited, in late 50's, they had an *operating* permit for a spark gap transmitter. One restriction was NO ANTENNA ;}
Reply by ●December 1, 20082008-12-01
Jerry Avins wrote: ...> "Dirty" cathodes were another manufacturing defect that rarely caused a > problem, but were the source of a near disaster for McIntosh at the > first Boston Audio Show around 1952. (Gordon Gow wanted me to join the > company after I solved his mystery. Dumb luck.) I know someone will ask. > I'll write it up some other time.Vacuum tube computers used dual triodes (6SN7s or their "Special Red" equivalent, 5692) to make flip-flops[1]. Sometimes, if a flip-flop remained in one state for a long time, it wouldn't respond immediately to a set or reset. It was said to have "gone to sleep". What a way to louse up a computer run! The mystery of sleeping sickness was elucidated by the clever folks who tended Harvard's computer[2]. A paper of theirs described how impurities in the oxides coating the cathode collected as a monolayer on the surface in the absence of current, effectively putting a large resistance between the space charge and the cathode. (Think of it as a resistance between the actual cathode and the cathode pin on the tube.) This greatly lowered the tube's transconductance, and when the loop gain drops below unity, .... OK: read and filed away under irrelevant. Nothing is irrelevant forever. McIntosh had come out with a new power amplifier. Their sales manager and chief engineer, Gorgon Gow, dropped into our shop frequently and brought us (and other vendors in Boston) new samples to display. I used it with an AR-1 woofer and a Marantz electrostatic tweeter, a good match and both local products. It was small enough to fit any room and sounded really great. I took a lot of orders the first day. My the end of the day, a fuse in the tweeter had blown. Without investigating, I replaced it. (My excuse: it was hectic.) About half an hour into Sunday morning, the fuse blew again. A check with the VOM showed about 30 volts coming out of the amplifier with nothing going in. A capacitor in series with the meter lead led me to think the signal was well over 100 KHz. I replaced the fuse in the tweeter, turned off the amplifier, got the schematic out of the packing box, and sat in a corner. The feedback, as was common, went to an unbypassed cathode resistor in the triode input stage. The stage itself was unusual. There was a very high resistance in the plate, making it what was called a starved-circuit amplifier. The cathode current was microamps. In a hunch, I replaced the tube (RCA, of course, although I couldn't really explain why.) The amplifier behaved, and I reconnected everything. I had folded the schematic and put it into my back pocket. I put the remainder of the sleeve of 12AX7s in my jacket pocket and went in search of Gordon. I found him striding down a corridor and ran to fall into place beside him. I said, "Gordon, one of your amplifiers just blew one of my tweeters." He stopped dead and said "Not you too!" Then he started again and said, "That's the fourth one. I have only one spare left, and it's promised." I told him that I had fixed mine and he stopped again. "How?" I told him and he asked if I had more tubes. I handed him the sleeve with four left. He turned around and we went to the last place he's been. When he pulled the tube, I noticed that, like the one in mine, it was a Sylvania. After we left, I expressed surprise that it hadn't been an RCA tube. He hadn't noticed, but he pulled the original out of his pocket, looked at it, and mumbled something about heads rolling. I went back to my exhibit room and he went to exchange front-end tubes. About half an hour later, he came to us and asked where he could get more RCA 12AX7s on Sunday. I had a few more sleeves in my workshop loft, and he prevailed on my boss to spare me for a time. He had replaced tubes in all the oscillating amplifiers, but more were acting up as they burned in, and he wanted to replace them all. Before he dealt with the not-yet crises, he took me to lunch and asked for explanation. I realized that I was sitting on the schematic, and unfolded it. Pointing to the input tube's cathode, I asked what would happen if I inserted a couple of K /here/? after a while, he said "It would go unstable." Then I told him about sleeping sickness. A week or so later he told me that he was starting to develop an FM tuner, and offered me a job. It would have been a far better position than I had, but I turned it down because it would have put me too far from my girlfriend. It was a good trade: we married a few years later. Jerry P.S. McIntosh's "purchasing department" (AKA "she") did indeed have standing orders to buy only RCA receiving tubes, but that was overridden in the last-minute rush to get the amplifiers out the door in time for the show. IPOIO, for sure! ________________________________ 1] Eccles-Jordan monostables. 2] The computer occupied an entire building that had no heating system. Air was drawn in through vents on the roof, passed through the racks of tubes, and passed out through grates in the sidewalk. In summer, the heat was oppressive; in winter I often went out of my way to stand in 70F air and remove my earmuffs. -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
