Just on a maintenance standpoint alone it would not be practical to use a 26 in a recording studio and or radio station. The 26 takes a lot of circuitry to obtain it clean clear output. Studios would not have the time to keep it running correctly because of being booked non-stop all day long. Another thing to consider is the tube availability and cost. They haven't made the 26 for probably 80yrs or so. Never worked in a studio yet that uses tube gear for mix purpose or recording purpose. You might hear a tube amplifier on a pair of monitors but that would be about it.
Leave the 26 where it belongs...in your home system.
Quote
use separate floating power supply's of 1,5 V ( a battery is absolute humfree
Tried batteries and still had hum.
That was my initial thoughts but the studios have been cranking out so much digital brickwalled/compressed garbage for years I commend this guy for trying something different. We know studios add compression, harmonic distortion, reverb, you name it via DAW so why not use a DHT to add a little compression and harmonics rather than some crappy algorithm?
The biggest issue will be 60hz noise (or 50hz in europe) and its harmonics. Even if the filament supply is perfect any choke or transformer in the circuit will pick it up like a radio receiving FM. I would use it with a high pass crossover 200hz and leave the bass to solid state. But its open-mindened creativity like this guy that the studios need to get out of the mess they are in.
If anyone likes the idea of valve-based recording, and you have not seen it - enjoy the article on Toe Rag Studios in London. It's analogue-only, & valve dominated:
Analogue productions aplenty at ToeRag - StudioNews-content | PSNE Magazine Online & In Print
.
Analogue productions aplenty at ToeRag - StudioNews-content | PSNE Magazine Online & In Print
.
I don't see maintenance as an issue or obtaining stocks of 26 - plenty about - and I don't think studios are that busy these days. I see this as a project for any part of the chain requiring maximum clarity and sound quality. Not sure where that would be exactly - I thought most recording was digital. But I do see noise/hum/chokes/transformers as needing some design solutions. All in favour of creative solutions!! That's the whole fun of designing.
andy
andy
Glow Tube Regulators - Advantage or Disadvantage?
Hi all,
I'm about to begin my 26 preamp project, and I have a question about two alternative schemes for B+. I already have a CLCLC supply to use, so that is set in stone. Here are the two scenarios:
1) CLCLC power supply (340V) to CCS-fed 0A3-0D3 in series (225V) (one set of CCS and regulators per channel) to CCS on plate of each 26 (135V or so, whatever the CCS sets it at for 6mA). First CCS set at 26mA, second at 6mA. Bypass 0A3-0D3 pair with .056uF cap to reduce noise. 9V Battery grid bias.
2) CLCLC power supply (340V) directly to CCS on plate of each 26 (6mA, ~135 V), with no shunt regulators. 9V battery grid bias.
Please only consider these two scenarios (no mosfet shunt regs, etc.). Is there a clear advantage in sound quality to adding the extra CCS and gas regulators to the circuit, or are they simply unnecessary complexity? My concern is that the regs may actually add more noise to the system than they remove. Following the KISS principle I'm leaning toward #2, but the lure of four glowing regulator tubes keeps me coming back to #1.
I'll most likely end up trying both scenarios anyway, but I'm looking to learn from the experience of others if I can. Anyone?
Hi all,
I'm about to begin my 26 preamp project, and I have a question about two alternative schemes for B+. I already have a CLCLC supply to use, so that is set in stone. Here are the two scenarios:
1) CLCLC power supply (340V) to CCS-fed 0A3-0D3 in series (225V) (one set of CCS and regulators per channel) to CCS on plate of each 26 (135V or so, whatever the CCS sets it at for 6mA). First CCS set at 26mA, second at 6mA. Bypass 0A3-0D3 pair with .056uF cap to reduce noise. 9V Battery grid bias.
2) CLCLC power supply (340V) directly to CCS on plate of each 26 (6mA, ~135 V), with no shunt regulators. 9V battery grid bias.
Please only consider these two scenarios (no mosfet shunt regs, etc.). Is there a clear advantage in sound quality to adding the extra CCS and gas regulators to the circuit, or are they simply unnecessary complexity? My concern is that the regs may actually add more noise to the system than they remove. Following the KISS principle I'm leaning toward #2, but the lure of four glowing regulator tubes keeps me coming back to #1.
I'll most likely end up trying both scenarios anyway, but I'm looking to learn from the experience of others if I can. Anyone?
After reading all 159 pages here....
It seems as if your second choice maybe the one preferred. A couple of people who had gas regulator tubes in their system eventually took them out...But surely try both if you have the parts on hand !
I read all the pages, too, and this topic did not seem to have a clear resolution; some liked them some didn't. Also, I didn't see anything about using two gas tubes in series per channel - I wonder if that might cause even more noise than a single tube.
The CCS should be able to drop that much on each channel without an issue. Since each 26 only pulls 6 mA, it comes out to about 1.2W dissipated per channel.
I'm using 340V because that's the power supply I have handy. If I find that the 26 preamp beats my current preamp (that uses the 340V power supply) I'll build another PS for the 26 with lower output voltage, probably LCLC.
I'm using 340V because that's the power supply I have handy. If I find that the 26 preamp beats my current preamp (that uses the 340V power supply) I'll build another PS for the 26 with lower output voltage, probably LCLC.
Quote:
I don't see maintenance as an issue or obtaining stocks of 26 - plenty about - and I don't think studios are that busy these days. I see this as a project for any part of the chain requiring maximum clarity and sound quality. Not sure where that would be exactly - I thought most recording was digital. But I do see noise/hum/chokes/transformers as needing some design solutions. All in favor of creative solutions!! That's the whole fun of designing.
I'm not trying to throw water on the idea by any means. My experience with studios here in the U.S. has been that they are always booked. They generally abuse equipment and or leave it on all the time. Considering the life of a #26 tube I don't see leaving it on 24/7 as a good idea. While there are some #R6's left if studios were to adopt the idea I think the price of the 26 is going to go up and the supply will go down. I don't happen to see a source of new 26 anywhere. I personally feel that the studios will continue to employ equipment that is both cost effective and maintenance free so to speak.
I don't see maintenance as an issue or obtaining stocks of 26 - plenty about - and I don't think studios are that busy these days. I see this as a project for any part of the chain requiring maximum clarity and sound quality. Not sure where that would be exactly - I thought most recording was digital. But I do see noise/hum/chokes/transformers as needing some design solutions. All in favor of creative solutions!! That's the whole fun of designing.
I'm not trying to throw water on the idea by any means. My experience with studios here in the U.S. has been that they are always booked. They generally abuse equipment and or leave it on all the time. Considering the life of a #26 tube I don't see leaving it on 24/7 as a good idea. While there are some #R6's left if studios were to adopt the idea I think the price of the 26 is going to go up and the supply will go down. I don't happen to see a source of new 26 anywhere. I personally feel that the studios will continue to employ equipment that is both cost effective and maintenance free so to speak.
I personally feel that the studios will continue to employ equipment that is both cost effective and maintenance free so to speak.>>
Yes - I'm sure this is true. Directly heated tubes will always be an offbeat choice, but in some cases sound quality may be a priority.
Does anybody know the relative lifespan of directly and indirectly heated tubes?
andy
Yes - I'm sure this is true. Directly heated tubes will always be an offbeat choice, but in some cases sound quality may be a priority.
Does anybody know the relative lifespan of directly and indirectly heated tubes?
andy
post 1538
I think I remember mentioning that I was going to use VR tubes for shunt regulation. I know that I received several comments to the fact that they are noisy. Well I have a thread about rectifier tubes sounding different and there is a post a reference to the Ammnity Raven and something else. Anyway there is a comment on VR tubes which I will quote here
A deluxe feature, which was easy to add, is the Voltage Regulator (VR) tube shunt-regulation for the driver. I wanted to drop 270 volts, so why not regulate at the same time? The noise of VR tubes is only 1mV of very smooth broadband noise, while the more common Zener diodes have 3 to 5mV of spectrally nonflat noise, occasional LF bumps and pops from "popcorn" noise, problems with temperature coefficients, and a huge amount of grossly nonlinear capacitance. Zener diodes need a lot of filtering and additional circuitry to isolate the problems. By contrast, VR tubes need no additional circuitry at all - just keep them away from capacitive loads.
The above I took from a write up on the Ammnity Raven.
I really don't know where the "Noise" comment is coming from. As it states above 1mv of broadband noise and yes that was with TWO VR tubes.
Please do enlighten me to the noise aspect. Did the others that tried the VR tubes use the SSHV2 regulator in conjunction with the VR tubes? I don't recall reading that they did. Did they use a CCS plate load and Rod's Filament supply board? My feeling is this....If they all used the above mentioned items in their design then and only then can you point the finger and say "Yes, VR tubes do contribute to the noise." they should not be considered at all. I keep hearing that the VR tubes are so bad but yet there is a mention to the effect that the VR tubes contribute only 1mv of noise and the zeniers have 3-5mv of noise. So, which is worse the zenier or the VR tubes?
So please help educate me on this because I can only see what I am reading in reviews and such and do not have an engineering degree like some of you do. I'm not trying to be a smart *** here by any means its just that I always analyze something using all the varables and I cannot see a proper answer if the same items are not used when considering the effect of VR tubes in the equation.
I think I remember mentioning that I was going to use VR tubes for shunt regulation. I know that I received several comments to the fact that they are noisy. Well I have a thread about rectifier tubes sounding different and there is a post a reference to the Ammnity Raven and something else. Anyway there is a comment on VR tubes which I will quote here
A deluxe feature, which was easy to add, is the Voltage Regulator (VR) tube shunt-regulation for the driver. I wanted to drop 270 volts, so why not regulate at the same time? The noise of VR tubes is only 1mV of very smooth broadband noise, while the more common Zener diodes have 3 to 5mV of spectrally nonflat noise, occasional LF bumps and pops from "popcorn" noise, problems with temperature coefficients, and a huge amount of grossly nonlinear capacitance. Zener diodes need a lot of filtering and additional circuitry to isolate the problems. By contrast, VR tubes need no additional circuitry at all - just keep them away from capacitive loads.
The above I took from a write up on the Ammnity Raven.
I really don't know where the "Noise" comment is coming from. As it states above 1mv of broadband noise and yes that was with TWO VR tubes.
Please do enlighten me to the noise aspect. Did the others that tried the VR tubes use the SSHV2 regulator in conjunction with the VR tubes? I don't recall reading that they did. Did they use a CCS plate load and Rod's Filament supply board? My feeling is this....If they all used the above mentioned items in their design then and only then can you point the finger and say "Yes, VR tubes do contribute to the noise." they should not be considered at all. I keep hearing that the VR tubes are so bad but yet there is a mention to the effect that the VR tubes contribute only 1mv of noise and the zeniers have 3-5mv of noise. So, which is worse the zenier or the VR tubes?
So please help educate me on this because I can only see what I am reading in reviews and such and do not have an engineering degree like some of you do. I'm not trying to be a smart *** here by any means its just that I always analyze something using all the varables and I cannot see a proper answer if the same items are not used when considering the effect of VR tubes in the equation.
Yes - I'm sure this is true. Directly heated tubes will always be an offbeat choice, but in some cases sound quality may be a priority.
Does anybody know the relative lifespan of directly and indirectly heated tubes?
andy
Andy, just for hell of it consider 5k hrs as a number and then figure 168hrs a week that it will probably run because some idiot will not turn it off. Take that 168 hrs a week X26 weeks and you have 4368 hrs figure a year and you have roughly 8760 hrs. Do you really think this is the first time someone has considered an idea like this? Please do consider the amount of tube gear that has been replace over the years. While there are thousands of us tube heads that would like to see the concept take shape I really doubt that the studios will readily adopt the tube concept again. My experience may very well differ from yours but mine shows that technology went from tube to sand to digital.
Had a recent conversation with a radio station chief engineer and mentioned tubes and I thought the guy was going to have a heart attack.
Does anybody know the relative lifespan of directly and indirectly heated tubes?
andy
Andy, just for hell of it consider 5k hrs as a number and then figure 168hrs a week that it will probably run because some idiot will not turn it off. Take that 168 hrs a week X26 weeks and you have 4368 hrs figure a year and you have roughly 8760 hrs. Do you really think this is the first time someone has considered an idea like this? Please do consider the amount of tube gear that has been replace over the years. While there are thousands of us tube heads that would like to see the concept take shape I really doubt that the studios will readily adopt the tube concept again. My experience may very well differ from yours but mine shows that technology went from tube to sand to digital.
Had a recent conversation with a radio station chief engineer and mentioned tubes and I thought the guy was going to have a heart attack.
Yes - I thought studios were digital these days. I'm a musician but haven't done any studio work for decades, so totally out of touch. My musician mates say in UK studio work is much reduced these days, but maybe a much smaller number of studios are working regularly and others have gone out of business.
I was just wondering is there were a difference in the lifetime of directly and indirectly heated tubes because of their construction? Small tubes in this case, not output tubes.
andy
I was just wondering is there were a difference in the lifetime of directly and indirectly heated tubes because of their construction? Small tubes in this case, not output tubes.
andy
Toe Rag is a real gem, one of those places that you hope will always be around.
I personally think that the issue of practicality is the reason that we lost so much quality in the recording process in the first place. It is a lot easier to use a compressor in your DAW, but will never sound as good as the real thing.
It is a really interesting time at the moment, engineers are starting to get excited about valve outboard again, which is great. I recently was at a studio in Oslo and the engineer was working on a beautiful old API desk, but using a valve summing mixer for some individual tracks - it's output stage was a 6SN7, so I don't see why(addressing the issues mentioned here) it couldnt be a 26.
Food for thought..
I visited the Oslo studio used by ECM in the 80s - Talent Studio I think - one evening for a bit of a party when I was working as a musician in Norway for 6 yrs. We listened to the master tapes of Kieth Jarrett - Jon Christensen (drs on the recording) was there and a few other guys. The sound was wonderful, but actually the quality of sound was due to the mixing which was miraculously good - the drum kit sounded so musical. The recording engineer Jan Erik Kongshaug is a really good guitarist - very musical guy. So though the sound itself was excellent, the musicianship was what did it for me! Also a gorgeous Steinway piano - probably the best I've played on.
http://franck.ernould.perso.sfr.fr/ecmvo.html Interesting article with some quotes
andy
http://franck.ernould.perso.sfr.fr/ecmvo.html Interesting article with some quotes
andy
Last edited:
Yup, that's my quandary, too. Depending upon what I read VR tubes are either horribly noisy or wonderfully benign. Gary Pimm states on his site that in conjunction with a CCS they make highly effective regulators. I suspect that in my application using a CCS both before and after the VR tubes, that noise will be minimized. Of course, trying it out will be the ultimate test. I have a scope, signal generator and RTA software for my PC, so I should be able to make some good measuements to compliment my listening. Probably won't happen for a few months, though...
Joe, I think you should go ahead and build with the VRs, if you like them, for whatever reason. The noise may not be perfect, but you could try them, without filtering, and see.
I can say this with confidence, because there are ready fixes and upgrade paths, if the VRs don't meet all your expectations.
In the first place, you can simply add a infrasonic filter (1.5M - 2.2uF/630V MKP) and then buffer the VR voltage with a N-channel FET. Should easily yield a factor of 10 or more reduction in noise, at the lowest audio frequencies. So, 5 parts on a little piece of plain FR4, with cuts in the copper for the circuit. I have made so many of these that they take only about 10 minutes to build now. I can post a photo/circuit if you should need it.
If you still have upgrade lust, you could use the VRs (filtered as above) as a reference to a shunt regulator. This would only require a few more parts.
The main thing is to leave a small amount of space in the chassis, to add some little boards - with TO220 heatsinking ability.
As a matter of taste, I would say that circuits that are good enough for Lynn Olson should be plenty good enough. I build my speakers to his design, and they have met every expectation for good sound, as well as his ability to communicate design decisions.
VR Tubes are incredibly quiet, in my experience. My setup has many CCS fed VR tubes. Preamp two VR's in series, amp input stage one VR, driver stage two VR. Broadband noise is relatively speaking non-existent. With your ear touching the tweeter you can hear a gentle 'flow of water'.
Pursuing absence of 60/120 Hz hum products is the bigger challenge, especially with DHT's. Most of my amps are AC heated, but I managed to keep hum below 1 mV.
It's funny, as the amp warms up you can hear the power supply hum and noise. As soon as the VR tubes fire, the sound instantly goes to black. Ahhhh.
Pursuing absence of 60/120 Hz hum products is the bigger challenge, especially with DHT's. Most of my amps are AC heated, but I managed to keep hum below 1 mV.
It's funny, as the amp warms up you can hear the power supply hum and noise. As soon as the VR tubes fire, the sound instantly goes to black. Ahhhh.
I am one who always rocks the boat so to speak. I have purchased Rods boards for my filament supplys. I have purchased the CCS plate load boards. I have paid for my group buy Salas SSHV2 boards and parts. I am still thinking I can use my supply with the VR tubes and the Salas Regulator after them. Should be aprox 0mv after the Salas. I am thinking that the Salas is a bigger CCS unit so my K&K Cascoded CCS boards for the plate load should work without a problem.
Attachments
Last edited: