I am trying to remember the number of a regulator triode, which is essentially half of a 6528 on a smaller base - but I can't remember if it was magnovial, 9 pin miniature or what.... it's a graphite plate tube, very similar construction to the whole family of 6as7/6336/6528....
So, I have found the 7233, but I know that's not it - as what I'm looking for has a slightly higher mu, and I remember much higher gm, like twice the 7233... like the difference between 6336 and 6528. It is to 6528 what 7233 is to 6336.... I wish I would have bought some years ago whenever I was pondering them, but it seems I did not, after rifling through my shelves a couple times over....
Anybody know what I'm on about? I'm just gonna have to go through data sheets one by one, lawd have mercy... and the thing is - it might be a soviet tube... sigh...
So, I have found the 7233, but I know that's not it - as what I'm looking for has a slightly higher mu, and I remember much higher gm, like twice the 7233... like the difference between 6336 and 6528. It is to 6528 what 7233 is to 6336.... I wish I would have bought some years ago whenever I was pondering them, but it seems I did not, after rifling through my shelves a couple times over....
Anybody know what I'm on about? I'm just gonna have to go through data sheets one by one, lawd have mercy... and the thing is - it might be a soviet tube... sigh...
Last edited:
As regulator, l used luckily the dual triode compactron 6FM7 and the same returned to 9 pin, the 6FD7 if it can help.
Do you remember the brand? With carbon anode Tung Sol , Raytheon and Bendix come to mind.
I am starting to think it doesn't exist, and might have been in the back of my mind as a "well I wish there were one like that" kinda thought, and after this much time, and with no real notes, I'm just getting confused... 7233 might be as close as I get. Maybe it was a pentode... nah, it was one in this super low rp family, like below 750 ohms....
Bendix feels right, and it would also explain not buying any - cause maybe it was just so expensive I decided it wasn't worth it just for...
The motivation for such a tube comes from the current demands of the 6528 and the desire to monoblock it... but you can't exactly cut the tube in half! But a twice-iterated power supply is also acceptable... And if the tube was over $50 I can imagine reacting with "bah! absurd! not a chance!" and just moving on.... ah but also, to explain that, I'm wanting to use it in the driver position, so there is also a power output tube to deal with, current wise...
Bendix feels right, and it would also explain not buying any - cause maybe it was just so expensive I decided it wasn't worth it just for...
The motivation for such a tube comes from the current demands of the 6528 and the desire to monoblock it... but you can't exactly cut the tube in half! But a twice-iterated power supply is also acceptable... And if the tube was over $50 I can imagine reacting with "bah! absurd! not a chance!" and just moving on.... ah but also, to explain that, I'm wanting to use it in the driver position, so there is also a power output tube to deal with, current wise...
What kind of power tube needs a V reg. tube to drive it? Running class B2?
6C33C
6C33C
Last edited:
Well, there's alot less "need" and alot more "why not" involved here. The notion happened while breadboarding and getting "good sound" from driver op points that allowed for absurd swings, then just trying to maximize that variable. I am not conventionally smart, or an engineer, or anywhere close to where alot of this community is in terms of comprehension. I grabbed on to one variable and thought "how far can I run with this?"
I have also run after the low rp variable with cathode followers and mu-followers with high gm tubes, very very similar to the 300b thread going on right now... they are using d3a and el802, I used e55l. Settled on the simple cathode follower cause i could figure out how to direct couple that, whereas the mu-follower I don't know what I'm doing & am very much so just copying other's work...
With enough gain in front of the 6528, I think I could drive quite a dang bit with it - maybe some big transmitter tubes. I have the parts for srpp 6550... I'd do wacky stuff like quads or huge 833a's (not that i have the tube) if I was capable of winding my own iron, maybe I should look into that again... I just like the experimental stage, so instead of having a finalized plan and ordering that particular thing, I enjoy building where I have all the junk to "play around" with the values... and that for the time being excludes exotic types, very high voltage or current tubes. coincidentally, it also mostly excludes building in stereo, hence my frustration with the 6528 tube. There's alot of dedication involved with the build.
What do y'all think we should drive with it? Let's say with a... 7193 cascode behind each half of the 6528 (just to keep it complicated, you know - 4 tubes & 8 top caps for stereo!!)? We could probably drive a truck with that. I mean, it could be ~5 watts at reasonable op points... with, might I add, more swing than my parents in the 70's. See, most people would do something much more efficient that would be immediately obvious to them. Not me! I do not know what I'm doing, really.
I have also run after the low rp variable with cathode followers and mu-followers with high gm tubes, very very similar to the 300b thread going on right now... they are using d3a and el802, I used e55l. Settled on the simple cathode follower cause i could figure out how to direct couple that, whereas the mu-follower I don't know what I'm doing & am very much so just copying other's work...
With enough gain in front of the 6528, I think I could drive quite a dang bit with it - maybe some big transmitter tubes. I have the parts for srpp 6550... I'd do wacky stuff like quads or huge 833a's (not that i have the tube) if I was capable of winding my own iron, maybe I should look into that again... I just like the experimental stage, so instead of having a finalized plan and ordering that particular thing, I enjoy building where I have all the junk to "play around" with the values... and that for the time being excludes exotic types, very high voltage or current tubes. coincidentally, it also mostly excludes building in stereo, hence my frustration with the 6528 tube. There's alot of dedication involved with the build.
What do y'all think we should drive with it? Let's say with a... 7193 cascode behind each half of the 6528 (just to keep it complicated, you know - 4 tubes & 8 top caps for stereo!!)? We could probably drive a truck with that. I mean, it could be ~5 watts at reasonable op points... with, might I add, more swing than my parents in the 70's. See, most people would do something much more efficient that would be immediately obvious to them. Not me! I do not know what I'm doing, really.
Last edited:
Oops, I found it! It's the 6c-a10 (not the 6ac10), indeed expensive & hard to find... made by NEC in Japan, and all the suppliers seem to be in Japan. The transconductance is actually about the same as the 7233, but the mu is up at 8 closer to the 6528. Awesome! Here is the only data sheet I can find:
https://frank.pocnet.net/sheets/084/5/50C-A10.pdf
I also later saw the 6c33c recommendation and yes, that would indeed be even more extreme current and swing, but pushes past the realm of "doable stuff" with what parts and junk I have. That would indeed be a wild project, to have one in the driver position... the tubes 7241 and 7242 are similar - like a triple 6528 internally paralleled.
The motivation to use the 7233 or now the 6c-a10 is to have a managable power supply... by comparison to the 6528. But once again, I'm gonna himm & haww about spending $80 each on a tube...
I also read in a thread that mentions the tube that you can get pretty similar behavior out of a triode-wired 6550, and I guess that's pretty much true, looking closely. Something like a load resistor of 600 ohm, 300v b+ and 240v Vq (100ma)... that might actually be the move, being that I have a good stash of 6× 6550 tubes.
https://frank.pocnet.net/sheets/084/5/50C-A10.pdf
I also later saw the 6c33c recommendation and yes, that would indeed be even more extreme current and swing, but pushes past the realm of "doable stuff" with what parts and junk I have. That would indeed be a wild project, to have one in the driver position... the tubes 7241 and 7242 are similar - like a triple 6528 internally paralleled.
The motivation to use the 7233 or now the 6c-a10 is to have a managable power supply... by comparison to the 6528. But once again, I'm gonna himm & haww about spending $80 each on a tube...
I also read in a thread that mentions the tube that you can get pretty similar behavior out of a triode-wired 6550, and I guess that's pretty much true, looking closely. Something like a load resistor of 600 ohm, 300v b+ and 240v Vq (100ma)... that might actually be the move, being that I have a good stash of 6× 6550 tubes.
Last edited:
If you're into current and GM , why not just put a dozen $1 tubes in parallel? or use a single triode as the Voltage amp and have it control a FET or BJT for the current.
Well, tube matching could become an issue. I will probably mull-over the possible tube choices though - would be looking for low-mu and sub-20ma at the op point, for paralleling a number like 4 tubes... Could be 2 twin-triodes. Hmm but again - matching. Could be problematic, especially with high-gm.
Also, the 7233 is probably "good enough," and cheap ($12). It's easy - only days away in the mail, domestic shipping. It's just more similar to half a 6336 - still very much what I'm going for. Just need a little more gain in the preceeding stage. It was just driving me crazy not being able to remember the 6c-a10... so, consider the itch to have been itched.
Also, the 7233 is probably "good enough," and cheap ($12). It's easy - only days away in the mail, domestic shipping. It's just more similar to half a 6336 - still very much what I'm going for. Just need a little more gain in the preceeding stage. It was just driving me crazy not being able to remember the 6c-a10... so, consider the itch to have been itched.
6DQ5 can do the current for $10, but has a low mu factor of 3.3
21LG6 can too, for $4 mu 3.6 (6LG6 extinct )
6EX6 maybe, $5 mu 4.2
6CA7/EL34 mu 10.5 , KT77 mu 11.5
Are you looking for a follower device or a gain device?
21LG6 can too, for $4 mu 3.6 (6LG6 extinct )
6EX6 maybe, $5 mu 4.2
6CA7/EL34 mu 10.5 , KT77 mu 11.5
Are you looking for a follower device or a gain device?
Last edited:
I experimented with the 7233 and Russian 6S19P many years ago for a small cathode follower output stage. Both worked OK. I have not been able to find a Mu spec for the 6S19P. I got a few from two different Ebay sellers in the Ukraine. They were different in construction, and one might have had a graphite plate.
The much like the Gm, Rp of any tube varies with the current through it. Comparing data sheets numbers taken at different currents is not valid for different tubes. You state that you are looking for a lot of drive output. This usually means that you will need a tube capable of a high cathode to plate volage (higher than your expected peak to peak voltage swing.) This rules out most of the regulator triodes. Drive a truck with it?
A reasonably high peak voltage requirement, Mu in the 6 to 12 range, 5 plus watts of dissipation, and reasonable heater power requirements, where are we looking. Digging back into a dark corner of my brain that was populated 50 to 60 years ago, I find some class A linear circuits that were designed to deliver a few watts of power at high peak voltages with very good linearity. I often lifted the entire circuit and stuck a guitar cable on the input and a speaker on the OPT secondary to make free guitar amps. We are looking at the Vertical Sweep Output section of a TV set. Unlike the flyback switching operation of the horizontal output section, the vertical output circuit is a linear class A amp optimized for operation at a single frequency, 50 or 60 Hz depending on where you live. Bad linearity gave people stretched heads and tiny feet. The picture often crept up from the bottom as the tube was dying. I worked in a TV repair shop from 1967 through 1970.
Smaller TV sets from the late 40's and 50's used a 6SN7 or even a 12AU7. Single triodes are the 6S4 and 12B4A. Larger TVs got the 6CK4, 6AH4GT (single triode), the 6BL7 or 6BX7 (twin triode) before pentodes took over that socket. They are all possible choices for your application. The old pentode standard for small TV's was the 6W6GT and even the 6V6GT. There are also lots of specialized vertical sweep tubes with two triodes or a triode and a pentode. One is intended for the oscillator and the other is the output. A single tube can be used to make a decent 1 to 3 watt SE audio amp, or a driver capable of driving a big tube. Possible tube choices here are the 6DN7, 6EA7, 6EM7 and 6DR7. There are others but I can't remember the numbers.
The cathode follower experiments are here, schematic is in post #22:
https://www.diyaudio.com/community/threads/20w-cathode-follower-amplifier.105859/
Note:
While reading some of my old posts in the thread mentioned above, I found mention of the Bendix 6877. It may be the "mystery tube." Very expensive though at $65 each.
The much like the Gm, Rp of any tube varies with the current through it. Comparing data sheets numbers taken at different currents is not valid for different tubes. You state that you are looking for a lot of drive output. This usually means that you will need a tube capable of a high cathode to plate volage (higher than your expected peak to peak voltage swing.) This rules out most of the regulator triodes. Drive a truck with it?
A reasonably high peak voltage requirement, Mu in the 6 to 12 range, 5 plus watts of dissipation, and reasonable heater power requirements, where are we looking. Digging back into a dark corner of my brain that was populated 50 to 60 years ago, I find some class A linear circuits that were designed to deliver a few watts of power at high peak voltages with very good linearity. I often lifted the entire circuit and stuck a guitar cable on the input and a speaker on the OPT secondary to make free guitar amps. We are looking at the Vertical Sweep Output section of a TV set. Unlike the flyback switching operation of the horizontal output section, the vertical output circuit is a linear class A amp optimized for operation at a single frequency, 50 or 60 Hz depending on where you live. Bad linearity gave people stretched heads and tiny feet. The picture often crept up from the bottom as the tube was dying. I worked in a TV repair shop from 1967 through 1970.
Smaller TV sets from the late 40's and 50's used a 6SN7 or even a 12AU7. Single triodes are the 6S4 and 12B4A. Larger TVs got the 6CK4, 6AH4GT (single triode), the 6BL7 or 6BX7 (twin triode) before pentodes took over that socket. They are all possible choices for your application. The old pentode standard for small TV's was the 6W6GT and even the 6V6GT. There are also lots of specialized vertical sweep tubes with two triodes or a triode and a pentode. One is intended for the oscillator and the other is the output. A single tube can be used to make a decent 1 to 3 watt SE audio amp, or a driver capable of driving a big tube. Possible tube choices here are the 6DN7, 6EA7, 6EM7 and 6DR7. There are others but I can't remember the numbers.
The cathode follower experiments are here, schematic is in post #22:
https://www.diyaudio.com/community/threads/20w-cathode-follower-amplifier.105859/
Note:
While reading some of my old posts in the thread mentioned above, I found mention of the Bendix 6877. It may be the "mystery tube." Very expensive though at $65 each.
Thank you, um I've read hundreds of posts from you over the years and it makes me feel a little tingly that you'd respond to my nonsense... also a little bit of shame, or a feeling of foolishness.
Yes I am probably really misunderstanding some fundamental thing about what I am even doing....
I have parted out a couple TVs with shattered picture tubes, and I actually have the 6s4a, 12b4a, 6em7, and maybe even one or two others you mentioned, although those three I've played around with...
I am trying both common-cathode and follower arrangements. I know that it's easier to talk about one or the other, though... sorry to be kinda vague and nebulous, rather than offering a schematic. But that's where I'm at, sketching phase...
My whole interest in these regulator tubes came from getting "good sound" with the 6s4a, and a few other tubes, running them with near-vertical loadlines, as the drive stage - in both common cathode amplifierand cathode follower, there was a discernable improvement - to my ears, not as a measured phenomenon. For the 6s4a, something like a load resistor of only 1k ohms, b+ 250 and Vq 235. So my overly-simplistic misunderstanding of what is actually happening, led to looking for tubes that are "supposed" to run like this, where Ia is as high as possible when Vg=0. And, where I can get into the more linear sections of the curve while having a loadline and an op point similar to this. I was interpreting that, the differential between Ia when Vg=0 and Iq was what I was trying to maximize. That's "the variable" that I grabbed onto.
I also went after lowest possible rp, via high gm, high mu tubes, and got really good results with the e55l as a pentode cathode follower arranged like this (halfway down the page): https://www.tubecad.com/2007/04/blog0104.htm
On the vtadiy loadline calculator (which I am pretty dependent on), you can see that when you pull up the 6as7g, the "default" loadline is very similar to what I was trying with the 6s4a (and other tubes). Vertical, Vq very close to b+. This led me to comparing the datasheets of all such tubes, and selecting the 6528 as the "best." But the demands on the power supply, and the necessity of building in stereo, kept me from ever finishing the project.
Now I'm picking up the pieces, and trying to complete some half-baked amps, so I'm looking at single-triode tubes to reduce the build complexity, and to be able to play with it on breadboard in mono...
It's very possible I was already where I needed to be with the 6s4a (and other tubes), and I'm chasing geese. According to the data sheet 6s4a can handle that kind of operating point. Maybe just go back to that? I was curious though, to do "more cowbell" and just build the most that thing, if that makes sense. And i was curious to ask more of it - i would try driving a 211 or 845 but i don't have the tubes or parts, so i was gonna try with 6550 self-inverting push-pull.
Triode wired, two of them would take about twice the drive voltage of a 45, unless I have some fundamental misunderstanding about that, too (possible). I have orphan transformers for that... a couple different things in the ballpark to try different loads. Then aquire a second, or if impossible order a matching pair of something new (edcor or hammond) once I've decided on the load.
Oh, and please excuse my lousy attempts at humor...
Yes I am probably really misunderstanding some fundamental thing about what I am even doing....
I have parted out a couple TVs with shattered picture tubes, and I actually have the 6s4a, 12b4a, 6em7, and maybe even one or two others you mentioned, although those three I've played around with...
I am trying both common-cathode and follower arrangements. I know that it's easier to talk about one or the other, though... sorry to be kinda vague and nebulous, rather than offering a schematic. But that's where I'm at, sketching phase...
My whole interest in these regulator tubes came from getting "good sound" with the 6s4a, and a few other tubes, running them with near-vertical loadlines, as the drive stage - in both common cathode amplifierand cathode follower, there was a discernable improvement - to my ears, not as a measured phenomenon. For the 6s4a, something like a load resistor of only 1k ohms, b+ 250 and Vq 235. So my overly-simplistic misunderstanding of what is actually happening, led to looking for tubes that are "supposed" to run like this, where Ia is as high as possible when Vg=0. And, where I can get into the more linear sections of the curve while having a loadline and an op point similar to this. I was interpreting that, the differential between Ia when Vg=0 and Iq was what I was trying to maximize. That's "the variable" that I grabbed onto.
I also went after lowest possible rp, via high gm, high mu tubes, and got really good results with the e55l as a pentode cathode follower arranged like this (halfway down the page): https://www.tubecad.com/2007/04/blog0104.htm
On the vtadiy loadline calculator (which I am pretty dependent on), you can see that when you pull up the 6as7g, the "default" loadline is very similar to what I was trying with the 6s4a (and other tubes). Vertical, Vq very close to b+. This led me to comparing the datasheets of all such tubes, and selecting the 6528 as the "best." But the demands on the power supply, and the necessity of building in stereo, kept me from ever finishing the project.
Now I'm picking up the pieces, and trying to complete some half-baked amps, so I'm looking at single-triode tubes to reduce the build complexity, and to be able to play with it on breadboard in mono...
It's very possible I was already where I needed to be with the 6s4a (and other tubes), and I'm chasing geese. According to the data sheet 6s4a can handle that kind of operating point. Maybe just go back to that? I was curious though, to do "more cowbell" and just build the most that thing, if that makes sense. And i was curious to ask more of it - i would try driving a 211 or 845 but i don't have the tubes or parts, so i was gonna try with 6550 self-inverting push-pull.
Triode wired, two of them would take about twice the drive voltage of a 45, unless I have some fundamental misunderstanding about that, too (possible). I have orphan transformers for that... a couple different things in the ballpark to try different loads. Then aquire a second, or if impossible order a matching pair of something new (edcor or hammond) once I've decided on the load.
Oh, and please excuse my lousy attempts at humor...
Last edited:
That sounds like a recipe for high levels of 2nd harmonic distortion, but maybe that's what you are looking for, especially for SET builds or 2nd H cancelling with output stage.
Some economical but quality tubes that might be of interest:
25HX5 14W (see E235L for data), good in triode, 220 mA max DC, Magnoval socket
10/6JA5 large 19W Vertical sweep tube,
6GF5 160 mA max DC, strong 9W tube for driver or small output,
6LU8/6LR8 14W Vertical sweeps with driver triodes,
38HE7 230 mA, 15W pentode if damper diode powered down, and good triode, cheap, (pentode Htr pins 10 and 12 )
6197 7.5W driver, cheap, high quality sub for 12BY7
6HZ8 8W driver and front end triode, cheap, similar to 6CL6/6197/12BY7
6KV8 5W high gm driver and front end triode, pentode good in triode mode
12HL7 7W to 10W high gm driver
6JC6 3W high gm front end (Westinghouse/Japanese versions are good in triode)
Last edited:
One of my favorite unknown tubes. The 6GF5 is a 6GE5 with its wings clipped and stuffed into a tiny bottle. Both are 12 pin versions of the octal 6DQ6. Obviously the large plate version with it's higher plate dissipation ratings is needed for any class A or other continuous high power duty. The 6GF5 works well in class AB1 music applications where music peaks are 10 to 20 dB larger than the average power output. The pictures show the similarities between all three tubes, a size comparison with the 6V6GT and the first test of the UNSET concept in a real amp board that could be squeezed to 80 Watts per channel. These tubes would be idling in a driver scenario.
Attachments
True, but the 6GF5 brings 770V B+ or 160 mA capabilities in a smallish bottle that almost "looks" like a driver. Just perfect for those exotic tough to drive output cases like HV transmitter tubes or class B2 tubes. For a (strange) OTL like below, 30 12B4s might have been replaced by just 6 6GF5s. (would need to check Watts, but rail V's could be lower with the 6GF5 )
( 6GF5s were on the famous $1 list for a long while.)
Last edited:
They were on the ESRC dollar list, but not available at all from RES (vacuumtubes.net). After floundering on the ESRC dollar list for several years with ZERO sales, I grabbed a few and "tested" them. I then called Stan and bought all he had ( maybe 200) for something less than $1 each. I haven't tried the massively parallel tubes experiment with them yet, but I have made some kick a$$ guitar amps with one pair. In over 10 years of torturing them I have not yet killed a single tube except for the one that ran off the end of the bench and did a suicide dive to the concrete floor. I'm going to make some adapters to put a few into octal bases for some swapability in guitar amps. Proper dual voltage screen voltage supply is needed though.
Oh holy crap - those are some AMPS. Makes it seem really silly for me to be fussing & hand-wringing about a ~500mA build and "needing" to break it into mono, or seperate drive and power output, to make it happen....
I'm super curious about the power supply for that, and idk if you have a picture of the heater wiring, holy cow....
It seems I can find plenty of sub-200mA iron for nearly free, or maybe still cheap (never pairs, of course) but real big units just cost good money.... always looking for those craigslistings of a basement or barn after some poor hammy passed away... but I find these days, at least in California, alot of people are aware that kinda stuff is worth money... too many busy body craigslisters with those "cash paid for tubes" spam ads. Y'all see those? Hate that crap. If I could just get a surplus of cash to coincide with a really good estate sale opportunity at some point... I'd love to have kit like 2000vct 500ma just lying around... to play with, you know? Just for fun...
With tubes, I've gone from just grabbing good deals to being very selective. I literally will never need another 12ax7, deep drawer of mullards. I should start selling stuff and trading the capital for more interesting things, like high current iron, transmitter triodes, stuff I think of as "too expensive." I got alot of stuff to sell if I start going through it.... hmm. Get enough crap together and you've got gold!
I'm super curious about the power supply for that, and idk if you have a picture of the heater wiring, holy cow....
It seems I can find plenty of sub-200mA iron for nearly free, or maybe still cheap (never pairs, of course) but real big units just cost good money.... always looking for those craigslistings of a basement or barn after some poor hammy passed away... but I find these days, at least in California, alot of people are aware that kinda stuff is worth money... too many busy body craigslisters with those "cash paid for tubes" spam ads. Y'all see those? Hate that crap. If I could just get a surplus of cash to coincide with a really good estate sale opportunity at some point... I'd love to have kit like 2000vct 500ma just lying around... to play with, you know? Just for fun...
With tubes, I've gone from just grabbing good deals to being very selective. I literally will never need another 12ax7, deep drawer of mullards. I should start selling stuff and trading the capital for more interesting things, like high current iron, transmitter triodes, stuff I think of as "too expensive." I got alot of stuff to sell if I start going through it.... hmm. Get enough crap together and you've got gold!