My guess(!) would be that due to the frame grid wire being so fine, they would be prone to sagging and have poor thermal conduction along the wire to the supporting frame and any heatsink. Gold plating won't help much with either with sagging. A narrower/ taller frame could have helped, like R5559. The PL802 and 6E5P used the wide anode spacing instead to reduce temperature.
Some things cannot be explained in technical terms. For example, you can have two violins that are made in the same way to the same pattern out of the same materials, but one sounding superb and the other bland. For hundreds of years experts came up with different explanations, no one good enough to make practical guidance for making a perfect instrument. But the fact is there: the superior quality of certain instruments is undisputable and the price of such instruments, the objective measure of quality, could be enormous.
Tubes are the same. A tube is a musical instrument the quality of which can be easily appreciated but cannot be adequately measured. This is why I like musician's approach to tube amplifier design that Andy practices. Try different things and find what sounds best.
To 27 mesh plates, I agree that these are very special tubes. Historically it was the first mass production indirectly heated triode, and its cathode design is fraught with all kinds of flaws. Porcelain insulation was electrically leaky and bulky, making cathode diameter large. The large size decreased the efficiency and necessitated higher temperature, causing grid emission problem. Mesh plate was used to remedy grid emission, as Rod pointed. I believe the sonic effect of mesh plate was unintended consequence. Unfortunately, too many of 27s, especially early production, are too noisy for hifi, but if you find good ones, their sound is very musical and engaging. The noisy ones can be used in old radios where it doesn't matter
Tubes are the same. A tube is a musical instrument the quality of which can be easily appreciated but cannot be adequately measured. This is why I like musician's approach to tube amplifier design that Andy practices. Try different things and find what sounds best.
To 27 mesh plates, I agree that these are very special tubes. Historically it was the first mass production indirectly heated triode, and its cathode design is fraught with all kinds of flaws. Porcelain insulation was electrically leaky and bulky, making cathode diameter large. The large size decreased the efficiency and necessitated higher temperature, causing grid emission problem. Mesh plate was used to remedy grid emission, as Rod pointed. I believe the sonic effect of mesh plate was unintended consequence. Unfortunately, too many of 27s, especially early production, are too noisy for hifi, but if you find good ones, their sound is very musical and engaging. The noisy ones can be used in old radios where it doesn't matter
Hi gupsta,
Frame grids are tensioned and have more support than a spiral grid. They were developed for shock and vibration environments. Plus with more tightly controlled mechanical alignment, the characteristics are far more uniform tube to tube.
Probably the best example of these differences would be the 6DJ8 (prototype) vs a 6922. The 6DJ8 is well known for variable performance and microphonics. They can be pretty noisy too.
Frame grids are tensioned and have more support than a spiral grid. They were developed for shock and vibration environments. Plus with more tightly controlled mechanical alignment, the characteristics are far more uniform tube to tube.
Probably the best example of these differences would be the 6DJ8 (prototype) vs a 6922. The 6DJ8 is well known for variable performance and microphonics. They can be pretty noisy too.
Hi sser2,
Actually, no. Tubes are not instruments and their sound is explainable. Ignoring defects, a tube has no "sound". It is how it reacts with the circuit that will determine a "sound". Tubes are supposed to adhere to a specification for performance for that tube type. Deviation in any direction is not good.
If a circuit is properly designed for a tube type, substituting a tube that has different specifications will usually mean higher distortion, differnt gain. Now if someone decides they like that, it is their personal preference. The circuit with that different tube does not perform as well as with the proper tube installed.
Design by ear is always inferior, both in measured and listening performance. If you decide you like a certain type of distortion, that's fine. It just isn't better. So whatever works for you is great for you. Go for it! Most people prefer lower distortion and noise long term.
Actually, no. Tubes are not instruments and their sound is explainable. Ignoring defects, a tube has no "sound". It is how it reacts with the circuit that will determine a "sound". Tubes are supposed to adhere to a specification for performance for that tube type. Deviation in any direction is not good.
If a circuit is properly designed for a tube type, substituting a tube that has different specifications will usually mean higher distortion, differnt gain. Now if someone decides they like that, it is their personal preference. The circuit with that different tube does not perform as well as with the proper tube installed.
Design by ear is always inferior, both in measured and listening performance. If you decide you like a certain type of distortion, that's fine. It just isn't better. So whatever works for you is great for you. Go for it! Most people prefer lower distortion and noise long term.
Design by ear is always superior. The audio gear is made for listening, not measurement. If I hear the kind of sound like in a concert hall, I am satisfied, and I don't care if measurements are not stellar and people who didn't listen to my system are telling me that this is distortion.
lol! No problem!
I have armies of clients who would disagree with you. Counterpoint is an excellent example of design by ear that sounded okay, but after redesigning those units to reduce distortion and noise, while keeping the sonic signature I have zero people who want the original. And the original didn't sound bad! The redesign eliminated the faults that caused them to fail as well. Go figure.
Of equipment I get in with serious design issues affecting reliability, almost all of expensive stuff was designed by ear. Same thing, reduce distortion and noise and it sounds better. This from their owners.
You're trying to dig up the subjective vs measure argument. Well, we aren't looking at meter needles, we are looking at audio (and beyond) spectrums beyond where you can hear. We listen too. Subjective opinion normally agrees with measured results if you do things right.
Why is it that "design by ear" manufacturers always regard their work as super secret? Normally one designer, not a team. Typically with no test equipment, but they are a "genius". Yup, this sounds like reality all right! Most times an ego that needs the truck entrance.
I have armies of clients who would disagree with you. Counterpoint is an excellent example of design by ear that sounded okay, but after redesigning those units to reduce distortion and noise, while keeping the sonic signature I have zero people who want the original. And the original didn't sound bad! The redesign eliminated the faults that caused them to fail as well. Go figure.
Of equipment I get in with serious design issues affecting reliability, almost all of expensive stuff was designed by ear. Same thing, reduce distortion and noise and it sounds better. This from their owners.
You're trying to dig up the subjective vs measure argument. Well, we aren't looking at meter needles, we are looking at audio (and beyond) spectrums beyond where you can hear. We listen too. Subjective opinion normally agrees with measured results if you do things right.
Why is it that "design by ear" manufacturers always regard their work as super secret? Normally one designer, not a team. Typically with no test equipment, but they are a "genius". Yup, this sounds like reality all right! Most times an ego that needs the truck entrance.
I accept your capacitance refutation, but I'm not buying your grid temperature argument because:
- Heater power is irrelevant because it's determined by the size of the cathode; an oxide-coated cathode operates at 1000K and a bigger cathode needs more power to hold it at 1000K.
- Heat transfer within the evacuated envelope is by radiation, so as you move further away from the source, you intersect a smaller proportion of total radiation, leading to the well-known inverse square law. But the difference between a grid being very close (perhaps 0.1mm) to a typical 1.5mm diameter cathode and astonishingly close (perhaps <10um to get that high gm) makes only 3% difference to the radiant heat received. Since the grid loses heat by radiation, which is proportional to the fourth power of absolute temperature, a 3% change in the amount of received heat changes grid temperature by 0.07%. So even if you (very pessimistically) assumed that the grid was also at 1000K, the change in grid spacing would cause a grid temperature change of 7K. Not a lot. Not enough to justify a radical design change.
- If the grid is too hot, why did they not use copper grid support rods coupled to fins? If the grid temperature is too high, then reducing the effect of radiation from a (much cooler) anode isn't going to make much difference; you need to cool the grid.
Cost / mass may be a consideration for a mesh plate.
Remember too, these tubes were not only used for audio.
Remember too, these tubes were not only used for audio.
Yes, frame grids allowed thinner wire to be used as the wire could be tensioned to within the limits of the materials elastic region which up to a certain temperature would be relaxed without sagging occuring, but the deflection is proportion to the power 4 by span and wire gauge whereas the area and thus mass is only to the power 2. Sagging in thinner wires would be one of the design limitations. For better thermal conduction to a radiator you want a greater a combined cross section of grid wire, which I guess would favor a few thick wires over many thin wires.
I design by ear because I'm a professional, conservatoire trained musician. Having been onstage for 20 years or so, the timbre and tonality of acoustic instruments is permanently engraved in my musical memory. So in all A-B testing my priority, of course, is to reproduce this accurate timbre and tonality. Build and listen, build and listen.
Musicians agonise over their choice of instrument to get exactly the timbre and tonality they require, and those who listen to audio should do the same if they really care about their music. So audition carefully, using only familiar tracks and familiar acoustic instruments. That way you get the most faithful timbre, which should be your goal. A Steinway sounds like a Steinway, a double bass sounds like a double bass.
So this means I have an ego that needs the truck entrance.......? I have absolutely no idea how you got there, but it seems that's the way your brain works.
Last edited:
Hi Andy,
I wasn't singling you out at all. I have met many audio designers and some of those folks have egos that would have trouble fitting into an airplane hanger. Typically single designers who know best by themselves. Some are regular people, but those tend to measure and listen.
I'm sure your instruments sound fantastic. Met a few musical instrument designers in my day. Lovely folks. You are creating a sound, a quality. You typically don't run the entire mix through the same effects unit.
BUT
Ya know what? A home music system is a reproduction system. You typically do not want to add extra coloration to what was laid down in the recording studio. The musician(s) got the sound they were looking for. You're going to layer that again? That would not be what they were looking for. Otherwise your efforts can be replaced with a clean system and effects pedals.
The best reproduction systems are clean, add or take away nothing. Why would you want everything to sound the same anyway? Or amp up the same effects the musician so carefully picked out. By the way, clean doesn't mean sterile at all.
I wasn't singling you out at all. I have met many audio designers and some of those folks have egos that would have trouble fitting into an airplane hanger. Typically single designers who know best by themselves. Some are regular people, but those tend to measure and listen.
I'm sure your instruments sound fantastic. Met a few musical instrument designers in my day. Lovely folks. You are creating a sound, a quality. You typically don't run the entire mix through the same effects unit.
BUT
Ya know what? A home music system is a reproduction system. You typically do not want to add extra coloration to what was laid down in the recording studio. The musician(s) got the sound they were looking for. You're going to layer that again? That would not be what they were looking for. Otherwise your efforts can be replaced with a clean system and effects pedals.
The best reproduction systems are clean, add or take away nothing. Why would you want everything to sound the same anyway? Or amp up the same effects the musician so carefully picked out. By the way, clean doesn't mean sterile at all.
Of course a music system is a reproduction system, nobody is questioning that. But exactly the same musical criteria apply. The sound of a Steinway still has to be the same sound of a Steinway as when it's played live. The timbre and tonality have to be as close as possible. You are not adding any coloration if the timbre is correct.
So it's really quite simple. Listen to the reproduction and ask yourself "Does this sound as close to a Steinway as possible?" Not a Bechstein, or a Yamaha but a Steinway concert grand. Timbre and tonality are unique. That's your goal. A grand piano is critical for timbre and tonality. When you get that right you'll find that other tonalities are pretty close. This is all assuming that you seriously listen to classical music, opera, jazz and other acoustic music. If all you listen to is rock and roll you may be easier to please.
So it's really quite simple. Listen to the reproduction and ask yourself "Does this sound as close to a Steinway as possible?" Not a Bechstein, or a Yamaha but a Steinway concert grand. Timbre and tonality are unique. That's your goal. A grand piano is critical for timbre and tonality. When you get that right you'll find that other tonalities are pretty close. This is all assuming that you seriously listen to classical music, opera, jazz and other acoustic music. If all you listen to is rock and roll you may be easier to please.
Last edited:
Hi Andy,
Absolutely! I hear enough live sound to know what you mean, and in the recording studio. While I am not a musician and can't identify make / model of instruments, I can sense when it sounds right. Even a Yamaha DX-7 can be identified easily. I just heard those a lot.
As long as your equipment can render the input accurately without adding "stuff", you're good. That goes for the entire chain once you have the media. So speakers and room come into it also. I'm sorry, but a single ended tube setup can't do that, it has a strong voice. Note I have not criticized the quality of sound as far as pleasant or not. That's up to the listener.
The Klipsch Jubilee speaker system we installed is the best I have ever heard, and maintains realism from very low SPL up to silly levels. We're still playing with the room, but man - is it ever close to being there.
Rock is related to classical. Good clean rock or some pop is demanding.
All music, whether you like it or not, is identifiable. I don't care for hip hop or rap at all, but I'm sure someone who likes it can tell between accurate or not. I like most music, have my favorites. You have to respect that most people who like music can tell the difference. I bet some musicians can tell who made a steel drum.
Absolutely! I hear enough live sound to know what you mean, and in the recording studio. While I am not a musician and can't identify make / model of instruments, I can sense when it sounds right. Even a Yamaha DX-7 can be identified easily. I just heard those a lot.
As long as your equipment can render the input accurately without adding "stuff", you're good. That goes for the entire chain once you have the media. So speakers and room come into it also. I'm sorry, but a single ended tube setup can't do that, it has a strong voice. Note I have not criticized the quality of sound as far as pleasant or not. That's up to the listener.
The Klipsch Jubilee speaker system we installed is the best I have ever heard, and maintains realism from very low SPL up to silly levels. We're still playing with the room, but man - is it ever close to being there.
Rock is related to classical. Good clean rock or some pop is demanding.
All music, whether you like it or not, is identifiable. I don't care for hip hop or rap at all, but I'm sure someone who likes it can tell between accurate or not. I like most music, have my favorites. You have to respect that most people who like music can tell the difference. I bet some musicians can tell who made a steel drum.
Sounds like the mesh tube needs real science.seed one to the haldron collider so they can do deep science for a definitive answere. If i was a betting man rod seems to be on target with his analysis.
Maybe down in the engineering lab, those guys want to believe this, but the board members/marketing/finance folks are the ones who actually make these decisions. The early 2A3 tube was 100% changed to the parallel twin triode version for production costs reasons and likely in part due to the fact the poor fidelity radios being made at the time didn't highlight the performance loss and additional capacitance this downgrade created. Possibly you have an army of people who will disagree with this article too?
https://audiopub.co.kr/wp-content/uploads/2023/12/2A3-The-Mother-of-High-Fiedelity.pdf