Member
Joined 2009
Paid Member
Jut looking at plate dissipation limit on two popular pentode/beam tetrodes.
6L6 - 30W
KT66 - 30W
I read that the KT66 was designed as a competitor to the 6L6.
So, why does the KT66 heater ask for 1.3A and the 6L6 0.9A. The more I look around the more I see that the 6L6 is quite frugal on heater current given that both tubes are capable of similar plate dissipations. Wouldn't you have designed the KT66 to run of 900mA heater current too ?
The Siemens F2a is even worse. Plate dissipation of 25W but a whopping 2A of heater current.
6L6 - 30W
KT66 - 30W
I read that the KT66 was designed as a competitor to the 6L6.
So, why does the KT66 heater ask for 1.3A and the 6L6 0.9A. The more I look around the more I see that the 6L6 is quite frugal on heater current given that both tubes are capable of similar plate dissipations. Wouldn't you have designed the KT66 to run of 900mA heater current too ?
The Siemens F2a is even worse. Plate dissipation of 25W but a whopping 2A of heater current.
In 1952 the KT66 was advertised as the finest audio tube in the world.
The KT66 was essentially a 6L6 redesigned with a larger cathode and a shorter, fatter anode. The KT66 had less rapid acceleration, longer electron paths, and almost no kink. It was ace for audio hi-fi but the electrons took longer to reach the anode and were more open to influence from magnetic, etc, fields on the way. Moreover, under HF conditions the electron stream was not perfectly in phase with the grid signal and this resulted in a loss of output power if used for RF applications.
A pair run in ultra linear AB2 push pull will deliver 50 Watts with 2% distortion.
The 6V6 receiver type of beam tetrode with which it shares pin connections is a derivative. This valve was mainly used for audio amplifiers.
The 807 was based on the RCA type 6L6 beam tetrode, which had a long, fairly thin, anode and a carefully-aligned grid structure. The 6L6 employed rapidly-accelerated electrons and short electron paths. This made it efficient but its characteristic had a distinct residual kink. It was a very successful valve but its characteristic was not quite straight enough to give the ultimate in audio quality at high output levels.
A pair of 6L6's working in class AB2 push pull would deliver 47 Watts at 2% distortion. A lot of power in the 1940s and 1950s.
The metal tube hides the inner construction. Our all glass 6l6GT version has a description of the electrode construction.
The wide metal tube envelope is 30 mm in diameter and excluding the IO base pins, is 90 mm tall.
References: Datasheet and private correspondence. Type 6L6 was first introduced in 1936.
The KT66 was essentially a 6L6 redesigned with a larger cathode and a shorter, fatter anode. The KT66 had less rapid acceleration, longer electron paths, and almost no kink. It was ace for audio hi-fi but the electrons took longer to reach the anode and were more open to influence from magnetic, etc, fields on the way. Moreover, under HF conditions the electron stream was not perfectly in phase with the grid signal and this resulted in a loss of output power if used for RF applications.
A pair run in ultra linear AB2 push pull will deliver 50 Watts with 2% distortion.
The 6V6 receiver type of beam tetrode with which it shares pin connections is a derivative. This valve was mainly used for audio amplifiers.
The 807 was based on the RCA type 6L6 beam tetrode, which had a long, fairly thin, anode and a carefully-aligned grid structure. The 6L6 employed rapidly-accelerated electrons and short electron paths. This made it efficient but its characteristic had a distinct residual kink. It was a very successful valve but its characteristic was not quite straight enough to give the ultimate in audio quality at high output levels.
A pair of 6L6's working in class AB2 push pull would deliver 47 Watts at 2% distortion. A lot of power in the 1940s and 1950s.
The metal tube hides the inner construction. Our all glass 6l6GT version has a description of the electrode construction.
The wide metal tube envelope is 30 mm in diameter and excluding the IO base pins, is 90 mm tall.
References: Datasheet and private correspondence. Type 6L6 was first introduced in 1936.
This is only true if 807 is used in a conventional way, with plate and screen voltage of 300 V. However, this is a wrong way of using power pentodes in audio. Pentodes do their best at Ua>>Ug2. With plate voltage of 750-800 and screen voltage of 300 V, a pair of 807s can deliver 120 W in AB2 at less than 2% distortion, and driving power of only about 100 mW. With its unique high voltage capability, 807 outclasses almost all 6L6 family tubes as audio output pentode.
Hi Kevin,
I know I have said it before, but the current Electroharmonix 6L6EH looks internally like a 7581A and will survive the same misuse too. It is one of my favorite tubes, even if you have no plans to push it (designed abuse).
Hi sser2,
The 807 has the plate on the top contact. Its the only way to avoid internal arcing at the higher voltages. But for commercial audio amplifiers, it is an insurance nightmare having all that energy easily accessed. Most audio people like to run their equipment with cages removed, so a nice, safe tube with hidden connections would be the safe way to run things.
-Chris
I know I have said it before, but the current Electroharmonix 6L6EH looks internally like a 7581A and will survive the same misuse too. It is one of my favorite tubes, even if you have no plans to push it (designed abuse).
Hi sser2,
The 807 has the plate on the top contact. Its the only way to avoid internal arcing at the higher voltages. But for commercial audio amplifiers, it is an insurance nightmare having all that energy easily accessed. Most audio people like to run their equipment with cages removed, so a nice, safe tube with hidden connections would be the safe way to run things.
-Chris
Chris: All I can say I am glad that I am not commercial
, because 807 sounds so much better at high voltages.Actually, some tubes (although not exactly 6L6 family) without top cap are rated high voltage, like 800 V EL34 and EL37, or 600 V 6550. But no manufacturer has the guts to use these capabilities. Long live DIY!
MOV-GEC chased RCA in to commercial use with its own indirectly heated cathode KT66 in 1937. The KT66, KT77, KT88 family had progressively higher heater currents (1.3, 1.4, 1.6A) - that family also went smaller.
At that time, in the mid-1930's there was rapid advancement going on with internals, including heater coatings, and cathodes, and improving on service life - so a proper consideration as to why 0.9A versus 1.3A would probably be quite a complex tale relevant to the times. I mean KT66 development was 10 years prior to Williamson developing his version of a low distortion amplifier.
Introduced in 1949 by Philips/Mullard, the EL34 has 1.5A, which many are happy to accommodate.
At that time, in the mid-1930's there was rapid advancement going on with internals, including heater coatings, and cathodes, and improving on service life - so a proper consideration as to why 0.9A versus 1.3A would probably be quite a complex tale relevant to the times. I mean KT66 development was 10 years prior to Williamson developing his version of a low distortion amplifier.
Introduced in 1949 by Philips/Mullard, the EL34 has 1.5A, which many are happy to accommodate.
Remember that the datasheets allow this condition only for ICAS service, not for continuous operation. I guess that otherwise the tubes' life would be dramatically shortened.
CCS (continuous service) allows a plate voltage of 600 V, giving an output power of 80 watts per pair.
The issues reasulting from this kind of operation are well known, at least for EL34's: Arcing between the plate pin #3 and the adjacent heater pin #2, which often leads to damage the phenolic base and/or the tube holder. A plate-to-plate spark gap or something similar is highly recommended (by experience).
Best regards!
Heater power and cathode current
One might think that heater power is related to cathode emissivity, which in turn determines maximum steady cathode current. Although generally more cathode current requires more heater power, there are other considerations, which may override current requirement.
To wit, directly heated small signal triodes 30 and 26. These two tubes operate at similar cathode currents, but filament power of 26 is whopping 25 times more! I guess the major considerations for high filament power of 26 were robustness, long service life, and lower microphony.
Cathode geometry may be a factor determining heating power in the 6L6 family. True 6L6 types including 807, 5881, and 7581 have thin tubular cathodes, whereas KT66, EL37, and 6550 have rectangular profile cathodes with much larger cross-section. Rectangular cathodes may be advantageous for precise alignment of control grid, but, as they have more surface area, they require more power to keep the emissive surface at the optimal temperature, hence higher heater rating.
One might think that heater power is related to cathode emissivity, which in turn determines maximum steady cathode current. Although generally more cathode current requires more heater power, there are other considerations, which may override current requirement.
To wit, directly heated small signal triodes 30 and 26. These two tubes operate at similar cathode currents, but filament power of 26 is whopping 25 times more! I guess the major considerations for high filament power of 26 were robustness, long service life, and lower microphony.
Cathode geometry may be a factor determining heating power in the 6L6 family. True 6L6 types including 807, 5881, and 7581 have thin tubular cathodes, whereas KT66, EL37, and 6550 have rectangular profile cathodes with much larger cross-section. Rectangular cathodes may be advantageous for precise alignment of control grid, but, as they have more surface area, they require more power to keep the emissive surface at the optimal temperature, hence higher heater rating.
What a beautiful amplifier! I have a weakness for design aesthetics of old times.
That's a lot for a thoriated tungsten filament. This would be considered a bright filament because it runs with the filament like a light bulb its so bright. Transmitting tubes are so far away from receiving tubes consumers use.
A 125 watt plate with a 3 KV rating would explain this tube. Forced air cooling recommended. Can you imagine a pair of these for an audio power amplifier? The output transformer would be something special too.
-Chris
A 125 watt plate with a 3 KV rating would explain this tube. Forced air cooling recommended. Can you imagine a pair of these for an audio power amplifier? The output transformer would be something special too.
-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.