That said, I noticed many times and studied the influence of vacuum degree variations on the tone of guitar amps, and confirmed it by oscilloscope output power monitoring and tube tester checking. That's why I speak about it. Sorry, I do not have pics of all that...
OK, I can see a leaky tube sitting at a different bias point than one with a good vacuum, but how that translates into an EL34 being more distorted than tha 6L6? The equipment pictures, I would guess, are a way to show credibility. Heck, I worked as a tech in a college and had a bench that had same stuff but newer. I miss a Panasonic scope I used. But at the time I was not into tubes. A shame as I tossed out well over 1000 tubes on my second day there.
Hard vacuum tube example - 01A
Soft vacuum tube example - 00A
Both are identical tubes physically, except the 00A still has a small amount of argon gas remaining. It was used as a detector tube with 22 volts on the plate. Under those conditions, it was a more sensitive detector than the 01A at any plate voltage and could detect weaker RF signals. You had to "play" with it more to obtain that better performance though.
The small amount of gas remaining caused a very noisy amplifier with randomly varying gain while in use, thus rendering it unsuitable in an amplifier application. The noise and gain variations are caused when an electron collides with the occasional gas molecule, releasing an additional electron which is then collected by the plate, thus temporarily increasing the plate current (hundreds of microsecond time spans).
I have never read anything about "distortion" specs for the 00A, but if the gain is varying randomly, I don't think that would reduce distortion, rather the contrary.
Soft vacuum tube example - 00A
Both are identical tubes physically, except the 00A still has a small amount of argon gas remaining. It was used as a detector tube with 22 volts on the plate. Under those conditions, it was a more sensitive detector than the 01A at any plate voltage and could detect weaker RF signals. You had to "play" with it more to obtain that better performance though.
The small amount of gas remaining caused a very noisy amplifier with randomly varying gain while in use, thus rendering it unsuitable in an amplifier application. The noise and gain variations are caused when an electron collides with the occasional gas molecule, releasing an additional electron which is then collected by the plate, thus temporarily increasing the plate current (hundreds of microsecond time spans).
I have never read anything about "distortion" specs for the 00A, but if the gain is varying randomly, I don't think that would reduce distortion, rather the contrary.
That story about the family with that car is not that strange, we have warm spells in winter because of the Chinook winds, and people often under or overdress because of it. But of course the baby changes things. Good job on your part.
On topic, the equipment pictures don't completely destroy credibility. If he is saying he had equipment and know how to measure what he is claiming then it is background.
On topic, the equipment pictures don't completely destroy credibility. If he is saying he had equipment and know how to measure what he is claiming then it is background.
When they developed new tubes they were not thinking about someone unboxing them 70+ years later. Also they just did not know what would happen with the new seals or gettering.
I think a lot of the seal problems got fixed with the next generation of transmitting tubes like 4CX250 or larger. I did get a stern warning from my dad to avoid any control grid current for that sort of tube, The wires will evaporate quickly.
I think a lot of the seal problems got fixed with the next generation of transmitting tubes like 4CX250 or larger. I did get a stern warning from my dad to avoid any control grid current for that sort of tube, The wires will evaporate quickly.
I have no qualms about his statements that in the 30's a gassy tube might have been better for this or that. But in the realm of an audio amplifier to state that manufactures retain some gas in the tube (example of a EL34) to give them a softer knee rather than attribute it to the geometry of the guts inside, seems odd to me. Mind you that might just be a tradeoff for a shorter time pumping down the tube. But when manufacturers like Mullard and other quality focused companies produced the same tubes and cut corners resulting in the same plate curves?
Last edited:
This seems to be "company private information". Not that there is a vacuum, but how hard and how to get there at lowest price.
We do know that run-of-the-mill tubes were only sucked a few hours to an insufficient vacuum, and gettering took it down another order of magnitude and kept it there.
OTOH we know there was a back-room at Mullard where very extended pumping was done for military contracts. Yes, there are some gasses which don't getter and maybe what was tolerable for consumer and industrial users was worrisome to military types.
And yes, one of the EARLY types had actual gas in it to goose-up the gain at the price of much hiss. That's far too old for anybody here.
No, I do not think EL34 is gassy; that's not what gas does, it's just low-price suppression.
Effect of Gas upon Tube Characteristics
Printer2 – Addressing your original post where you question some of the information you found regarding “hard” and “soft” tubes affecting the linearity, thus the sound of different tubes.
I found this from Frederick Terman’s book Radio Engineering, 2nd edition, 1937 where he discusses the effect of gas upon tube characteristics on pages 157 and 158.
[special=“Very small traces of gas in vacuum tubes affect the characteristics adversely in a number of ways as a result of the positive ions produced in the tube by collisions between the gas molecules and the electrons flowing to the plate. The positive ions travel in the opposite direction from electrons and normally end their existence by falling into the cathode or the negative control grid. Those which bombard the cathode tend to destroy the emission of thoriated-tungsten and oxide-coated cathodes, as has already been discussed, and this effect is sufficiently serious to limit the usefulness of these types of emitters. The positive ions collected by the negative control grid also result in grid current that causes grid-circuit power loss and limits the resistance that may be inserted in series with the grid, as discussed below. Another serious effect of the positive ions is the irregularities that they produce in the space charge at the cathode. This causes a shot effect that makes the tube “noisy”; this will be discussed in Sec. 49.
When traces of gas are present in a tube, the resistance that may safely be placed in series with the negative control grid of the tube is limited. This is because the voltage drop that the grid current produces across such a resistance has a polarity that makes the grid less negative. This increases the total space current, thereby increasing the number of positive ions and causing additional grid current and a still greater reduction in the negative grid potential. If the resistance in the grid circuit is high enough, this process can become cumulative, resulting in the control-grid potential suddenly becoming positive and causing the destruction of the tube as a result of excessive plate current. As a consequence there is a maximum resistance that is permissible to place in series with the grid electrode, with the allowable value depending upon the tube characteristics and the conditions under which the tube is operated.
When the gas pressure is increased somewhat, the voltage and current relations become seriously affected. The positive ions in drifting toward the cathode neutralize a portion of the negative space charge around the cathode and make it possible for more electrons to be drawn to the plate than would otherwise be possible. The velocity of the positive ions is so low because their large mass that the life of the average ion is much greater than that of an electron. Hence the rate of production of positive ions need not be very great to increase the plate current appreciably. The characteristic curves of tubes containing small amounts of gas also tend to be irregular and to have sudden bends or kinks.
When the amount of gas in a tube is quite large, the ionization is sufficiently intense to produce a luminous glow, and the tube is then said to be “soft.” The amount of gas required to produce a soft tube is quite small, since a pressure of one-millionth of an atmosphere will commonly give a visible glow and make the tube inoperative as a high-vacuum device.
The amount of gas in a tube can be estimated from the amount of grid current present when the grid is negative. This current is proportional to the number of positive ions, which, with appreciable anode voltages, are proportional to the total space current and the gas pressure. This is the basis of the ionization gauge used to measure very low gas pressures.”]%[/special]
Elsewhere in the chapter Terman describes how the anode current is limited by the Space Charge within the tube and follows the 3/2 power law. Next he shows the action of the grid and how the space charge is modified according to the physical geometry of the tube. He shows that the action of the space charge is a function of the grid and anode voltages, and together forms the characteristic curves of the tube.
Per the RCA Receiving Tube Manual, the maximum grid resistance for the 6L6 is 0.1 Megohm. The maximum grid resistance for the 6CA7 is 0.7 Megohm. If the gas pressure (amount of gas) inside the tube is the only factor affecting the maximum allowable grid resistance, then the 6L6 would be the “softer” of the two tubes. Of course the amount of gas is not the only factor affecting the maximum grid resistance spec. More tightly spaced grid wires will be in a position to be struck more often by positive ions, thus increasing grid current. The amount of negative voltage on the grid will also attract positive ions to varying degrees, also affecting grid current.
Your suspicions appear to be correct.
Printer2 – Addressing your original post where you question some of the information you found regarding “hard” and “soft” tubes affecting the linearity, thus the sound of different tubes.
I found this from Frederick Terman’s book Radio Engineering, 2nd edition, 1937 where he discusses the effect of gas upon tube characteristics on pages 157 and 158.
[special=“Very small traces of gas in vacuum tubes affect the characteristics adversely in a number of ways as a result of the positive ions produced in the tube by collisions between the gas molecules and the electrons flowing to the plate. The positive ions travel in the opposite direction from electrons and normally end their existence by falling into the cathode or the negative control grid. Those which bombard the cathode tend to destroy the emission of thoriated-tungsten and oxide-coated cathodes, as has already been discussed, and this effect is sufficiently serious to limit the usefulness of these types of emitters. The positive ions collected by the negative control grid also result in grid current that causes grid-circuit power loss and limits the resistance that may be inserted in series with the grid, as discussed below. Another serious effect of the positive ions is the irregularities that they produce in the space charge at the cathode. This causes a shot effect that makes the tube “noisy”; this will be discussed in Sec. 49.
When traces of gas are present in a tube, the resistance that may safely be placed in series with the negative control grid of the tube is limited. This is because the voltage drop that the grid current produces across such a resistance has a polarity that makes the grid less negative. This increases the total space current, thereby increasing the number of positive ions and causing additional grid current and a still greater reduction in the negative grid potential. If the resistance in the grid circuit is high enough, this process can become cumulative, resulting in the control-grid potential suddenly becoming positive and causing the destruction of the tube as a result of excessive plate current. As a consequence there is a maximum resistance that is permissible to place in series with the grid electrode, with the allowable value depending upon the tube characteristics and the conditions under which the tube is operated.
When the gas pressure is increased somewhat, the voltage and current relations become seriously affected. The positive ions in drifting toward the cathode neutralize a portion of the negative space charge around the cathode and make it possible for more electrons to be drawn to the plate than would otherwise be possible. The velocity of the positive ions is so low because their large mass that the life of the average ion is much greater than that of an electron. Hence the rate of production of positive ions need not be very great to increase the plate current appreciably. The characteristic curves of tubes containing small amounts of gas also tend to be irregular and to have sudden bends or kinks.
When the amount of gas in a tube is quite large, the ionization is sufficiently intense to produce a luminous glow, and the tube is then said to be “soft.” The amount of gas required to produce a soft tube is quite small, since a pressure of one-millionth of an atmosphere will commonly give a visible glow and make the tube inoperative as a high-vacuum device.
The amount of gas in a tube can be estimated from the amount of grid current present when the grid is negative. This current is proportional to the number of positive ions, which, with appreciable anode voltages, are proportional to the total space current and the gas pressure. This is the basis of the ionization gauge used to measure very low gas pressures.”]%[/special]
Elsewhere in the chapter Terman describes how the anode current is limited by the Space Charge within the tube and follows the 3/2 power law. Next he shows the action of the grid and how the space charge is modified according to the physical geometry of the tube. He shows that the action of the space charge is a function of the grid and anode voltages, and together forms the characteristic curves of the tube.
Per the RCA Receiving Tube Manual, the maximum grid resistance for the 6L6 is 0.1 Megohm. The maximum grid resistance for the 6CA7 is 0.7 Megohm. If the gas pressure (amount of gas) inside the tube is the only factor affecting the maximum allowable grid resistance, then the 6L6 would be the “softer” of the two tubes. Of course the amount of gas is not the only factor affecting the maximum grid resistance spec. More tightly spaced grid wires will be in a position to be struck more often by positive ions, thus increasing grid current. The amount of negative voltage on the grid will also attract positive ions to varying degrees, also affecting grid current.
Your suspicions appear to be correct.
Also, in the case of many Eimac glass valves, gettering doesn't happen continuously, but requires the anodes to get up to low red temperature. I'm currently building a test jig to test grid current and getter a bunch of type 304TL we've accumulated for a project. Should know more in a coupla months - the power supply's already done.
YOS,
Chris