Also, you are not going to find matched pairs or quads of this device. It's an orphan teevee front end tube (the datasheets call it a 'VHF amplifier'), and not manufactured to the tight tolerances that audio premium tubes are. I bought 360 of them (200 Hitachi and 160 Mullard) and tested up myself. The tubes are cheap; you can get a hundred nest of Hitachi for $99, cleaning all those pins and testing up is not.
Miles, I have seen it said here that with a CCS in the tail, the stage will be balanced no matter the chars of the devices. This seems contraintuitive to me. What if the intrinsic gains of the tubes are a little different? What if the native operating points of the tubes are a little different? The CCS is common; it doesn't force OP balance. Am I being obtuse? It seems to me that when I first used the IXYS part, I looked at the gain of the two sides of the diff-amp, and they were just a little different, and I switched tubes to get the grid-driven side a little more A, which gives my circuit the correct overall output second partial distortion that is psychoacoustically correct. Am I disremembering? Was I kidding myself?
Aloha,
Poinz
AudioTropic
Miles, I have seen it said here that with a CCS in the tail, the stage will be balanced no matter the chars of the devices. This seems contraintuitive to me. What if the intrinsic gains of the tubes are a little different? What if the native operating points of the tubes are a little different? The CCS is common; it doesn't force OP balance. Am I being obtuse? It seems to me that when I first used the IXYS part, I looked at the gain of the two sides of the diff-amp, and they were just a little different, and I switched tubes to get the grid-driven side a little more A, which gives my circuit the correct overall output second partial distortion that is psychoacoustically correct. Am I disremembering? Was I kidding myself?
Aloha,
Poinz
AudioTropic
Poinz,
Fairly recently, SY posted in a thread about his experiments using (sic) dissimilar triodes in a CCSed LTP. Even then, symmetry was forced.
Think of the CCS as the fulcrum of a seesaw. If the current in 1 leg goes up, the current in the other leg must go down.
Fairly recently, SY posted in a thread about his experiments using (sic) dissimilar triodes in a CCSed LTP. Even then, symmetry was forced.
Think of the CCS as the fulcrum of a seesaw. If the current in 1 leg goes up, the current in the other leg must go down.
It may seem counter intuitive, but the fact is that all tubes (and transistors) are "a little different". Even when using dual triodes (or transistor arrays where they're all on the same die) there will be some deviations from perfection.
However, I have found that the major difference is in DC offset. This becomes a major PITA when it comes to solid state designs that are DC coupled from input to output. Since my hollow state designs include one capacitor coupling, DC offset is NBD and not worth worrying about. Measured DC voltages have always come in under less than a 5% deviation from design nominal, and AC performance likewise. With the 6SL7 LTP, I calculated a gain of 25 / phase, and that is exactly what I measured: Av= 25 / phase.
When it comes to AC characteristics, every LTP splitter I did (one using Sovtek 6SL7s, the other cascoded 6BQ7As) showed no measurable difference in gain between phases, a CMRR of 60db(v) or better (can't measure it to any more accuracy) and immeasurably low distortion. Unless one half of the LTP is completely borked, I haven't found this to be a problem, either with hollow state or solid state.
Even if you start out with perfectly matched tubes, they are highly unlikely to stay that way due to aging.
Thank you, Miles. The element that I found odd is the symmetry with unlike devices, but Eli has explained it very elegantly; if the fulcrum is immovable, a decrease in current in the directly driven side must be be exactly balanced by an increase in current in the indirectly driven side. Q E D.
The penny drops in me tiny brain. Thank you, gentlemen.
Aloha,
P
The penny drops in me tiny brain. Thank you, gentlemen.
Aloha,
P
hi poinz,
what bias voltage do yuo rcomend for the new production tungsol 6v6. ive read read somewhere on one of your post that it biiases diffrently. im still using the 1 volt bias, or should i increase the negative bias to 3 volts?
thanks
what bias voltage do yuo rcomend for the new production tungsol 6v6. ive read read somewhere on one of your post that it biiases diffrently. im still using the 1 volt bias, or should i increase the negative bias to 3 volts?
thanks
The old stock 6V6s I have used biased up at (315V, 35mA) about -19V on the grid. The EH and Tung-Sol are more like -23V. This is the actual bias DC voltage on the output tube grid.
What you are talking about is the voltage drop across the 15Ω bias-take resistor, which is in the common 'tail' of the output pair. A voltage of one DC volt at the top of this resistor corresponds to a current of 66.7mA of current, or 33.3mA per tube.
Poinz
What you are talking about is the voltage drop across the 15Ω bias-take resistor, which is in the common 'tail' of the output pair. A voltage of one DC volt at the top of this resistor corresponds to a current of 66.7mA of current, or 33.3mA per tube.
Poinz
Do you have a circuit for Leben amp ? I heard audio note kit 4 (6V6) with paraphase splitter and it was really nice sounding little critter🙂
No I don't, I just copied the phase splitter. The Leben has an ultralinear output stage with a Gnfb loop, mine is triode wired and no Gnfb. Anyway, nothing special about the Leben schematic except the phase splitter.
Who knows why I don't like the LTP. It sounded too much hifi I guess, some music sounded impressive but most of the time it bored me. If an amp doesn't get me high it's no good, no matter how accurate it is.
Audio Note says something about the euphonic nature of their amp too. Whatever, the floating paraphase splitter is rarely used, Morgan Jones says it has too much distortion but the amp sounds pretty clean to me. It is definitively cleaner sounding than my 2A3/45 amp. To me it sounds very much like a good SET but with better bandwidth, better bass and, as i said before, cleaner sounding. It's not as human as the SET, though.
This is my phase splitter, it's like the Leben but with no feedback loop:
Who knows why I don't like the LTP. It sounded too much hifi I guess, some music sounded impressive but most of the time it bored me. If an amp doesn't get me high it's no good, no matter how accurate it is.
Audio Note says something about the euphonic nature of their amp too. Whatever, the floating paraphase splitter is rarely used, Morgan Jones says it has too much distortion but the amp sounds pretty clean to me. It is definitively cleaner sounding than my 2A3/45 amp. To me it sounds very much like a good SET but with better bandwidth, better bass and, as i said before, cleaner sounding. It's not as human as the SET, though.
This is my phase splitter, it's like the Leben but with no feedback loop:
Attachments
Thank you
I have a pair of UL Dynaco Z565 (st35) transformers
(along with PT) waiting for a project. I asume it is Leben CS300 you're talking about. What is the high tension (B+) on splitter stage?
Regards, L
I have a pair of UL Dynaco Z565 (st35) transformers
(along with PT) waiting for a project. I asume it is Leben CS300 you're talking about. What is the high tension (B+) on splitter stage?
Regards, L
Don't want to threadjack so I will send you more details about the cs-300 via PM when I have the time, OK?
Alright, its been a while but I'm back and ready to get started on finishing up this project. Actually, lets be honest, I'm going to be starting on this project but that's the first step towards to finishing.
Having completed my move from Tokyo to Berlin I am now just waiting for my last shipment of stuff to arive, which hopefully will be in a couple of weeks. Of course, that's all my hifi/DIY stuff so I've been going a little crazy here waiting. Anyway...
First order of business, I just read through this entire thread again and am wondering if I should just bite the bullet and order a matching PT and choke from Silk to go with my OPTs rather than try to get more creative with other options; thoughts? Poinz, not considering $$$ do you prefer the Antek Toroids to the Silk PT? I'm assuming you're probably the only that's heard them side by side.
Also, if I do get a PT from Silk it's probably going to just be a custom job because I recall you saying you can't get the PT you normally use from them anymore. If I'm going to order a custom job, I might as well get EXACTLY what I need; would you be so kind as to offer up the specs on a "dream PT?"
Thanks in advance for everyone's help on this project; I'm sure there will be lots more questions soon.
Having completed my move from Tokyo to Berlin I am now just waiting for my last shipment of stuff to arive, which hopefully will be in a couple of weeks. Of course, that's all my hifi/DIY stuff so I've been going a little crazy here waiting. Anyway...
First order of business, I just read through this entire thread again and am wondering if I should just bite the bullet and order a matching PT and choke from Silk to go with my OPTs rather than try to get more creative with other options; thoughts? Poinz, not considering $$$ do you prefer the Antek Toroids to the Silk PT? I'm assuming you're probably the only that's heard them side by side.
Also, if I do get a PT from Silk it's probably going to just be a custom job because I recall you saying you can't get the PT you normally use from them anymore. If I'm going to order a custom job, I might as well get EXACTLY what I need; would you be so kind as to offer up the specs on a "dream PT?"
Thanks in advance for everyone's help on this project; I'm sure there will be lots more questions soon.
I haven't dealt with Kittipol in awhile, just because I ordered up last time, lots of everything. What he was changing was his vendor of the cans he puts the fancy ones in, so he couldn't provide exact cosmetic replacements for stuff I had.
The reason we had that conversation was that at that time he was impregnating all his transformers with polyester resin (they smelled like surfboards), and this caused a problem for me with mechanical humming in the power transformers. The chokes and output transformers didn't develop this problem, because they don't heat up much. I never could dissuade him at that time from doing this, so I built my own vacuum impregnation rig and treated the ones that hummed (about half of them) with varnish. Maybe you don't want to do this. Maybe he doesn't use polyester impregnation any more; lord knows I couldn't have been the only one to have this problem.
I will say that his stuff is (or was) economical, otherwise very high in quality, and the output transformers the best sounding I have heard. If you do order from him, you want to order up as much as you can, otherwise the shipping will kill the deal. It's far.
In lieu, below is the power supply I use now, because the Antek and Amveco parts are very good and economical, and I like the (non)sound of sand rectification better:
Since the current Antek donut doesn't have the 12.6v winding, I'm pulling bias off the input diff-amp 'tail' supply.
Aloha,
Poinz
AudioTropic
The reason we had that conversation was that at that time he was impregnating all his transformers with polyester resin (they smelled like surfboards), and this caused a problem for me with mechanical humming in the power transformers. The chokes and output transformers didn't develop this problem, because they don't heat up much. I never could dissuade him at that time from doing this, so I built my own vacuum impregnation rig and treated the ones that hummed (about half of them) with varnish. Maybe you don't want to do this. Maybe he doesn't use polyester impregnation any more; lord knows I couldn't have been the only one to have this problem.
I will say that his stuff is (or was) economical, otherwise very high in quality, and the output transformers the best sounding I have heard. If you do order from him, you want to order up as much as you can, otherwise the shipping will kill the deal. It's far.
In lieu, below is the power supply I use now, because the Antek and Amveco parts are very good and economical, and I like the (non)sound of sand rectification better:
Since the current Antek donut doesn't have the 12.6v winding, I'm pulling bias off the input diff-amp 'tail' supply.
Aloha,
Poinz
AudioTropic
Poinz:
As allows, your thoughts/insights are much appreciated. I think I'll be going with the Antek as this will free up some funding and allow Front Panel Express to make me something nice for a top plate.
Thanks for the updated schematic!
As allows, your thoughts/insights are much appreciated. I think I'll be going with the Antek as this will free up some funding and allow Front Panel Express to make me something nice for a top plate.
Thanks for the updated schematic!
I ended up with an old scott 222c chassis (Yes, I know the OPTs aren't ideal, but I'm trying to scrounge here, I'm a grad student). I'll be diving in on this, using the method for generating the negative rail that Eli mentioned here:
http://www.diyaudio.com/forums/tubes-valves/100962-scott-222c-mods.html#post1197272
However, I don't see the need to play 'shoddy OPT' tricks with these nice Scott OPTs, so I'm thinking this over the el cheapo. I've got an old magnavox console that might be an el cheapo project in the future.
Can I still power the ax7's for the phono and a pair 6v6s? If not, maybe I'll rethink and try with el84's. Signal shaping circuitry will be removed.
What is the suggested schematic for the current sink?
I'm going to try and draw Eli's suggestion for the PS, and then post it as a comprehension quiz.
http://www.diyaudio.com/forums/tubes-valves/100962-scott-222c-mods.html#post1197272
However, I don't see the need to play 'shoddy OPT' tricks with these nice Scott OPTs, so I'm thinking this over the el cheapo. I've got an old magnavox console that might be an el cheapo project in the future.
Can I still power the ax7's for the phono and a pair 6v6s? If not, maybe I'll rethink and try with el84's. Signal shaping circuitry will be removed.
What is the suggested schematic for the current sink?
I'm going to try and draw Eli's suggestion for the PS, and then post it as a comprehension quiz.
D, do you know the primary impedance load, as referenced to the proper output load? We can adjust the output tubes' operating point to adjust, or maybe even go to the EL84/6BQ5, if the load is too heavy. Most of those guys, even if the transformers were of high quality, gave a heavy load to the outputs to maximize output power specs, relying on the feedback to control the speaker.
Aloha,
Poinz
AudioTropic
Aloha,
Poinz
AudioTropic
D,
With the Scott "iron", you might be better off with "El Cheapo". The loading factor Poinz mentioned is why. 9% GNFB will not cause sound stage collapse, but it will help in the damping factor dept. Since O/P trafo magnetic headroom is substantial, set the I/P high pass pole F3 in the 15 to 17 Hz. range. Use your o'scope to get "perfect" phase compensation. If you don't have an o'scope, simply install a 120 pF. mica or NPO ceramic cap. from the "hot" wire of the O/P trafo to ground.
With the Scott "iron", you might be better off with "El Cheapo". The loading factor Poinz mentioned is why. 9% GNFB will not cause sound stage collapse, but it will help in the damping factor dept. Since O/P trafo magnetic headroom is substantial, set the I/P high pass pole F3 in the 15 to 17 Hz. range. Use your o'scope to get "perfect" phase compensation. If you don't have an o'scope, simply install a 120 pF. mica or NPO ceramic cap. from the "hot" wire of the O/P trafo to ground.
I've done some searching, and reread Charles Hansen's aexpress articles. The impedance is about the only thing he didn't measure. Guess I'll need to borrow a signal generator and follow the instructions...
(where the d stands for doug)
Eli, Poinz',
So this is really my first foray into tubed gear, and after spending a whopping $90 so far I've got
1. a eico tube tester
2. a pretty nice magnavox PP el84 console pull (3 chokes in the PS)
3. a decent motorola SE el84 with a PP 6v6 summed output with HP for the bass channel
4. an Eico ST40, and
5. this Scott.
Since the iron in the magnavox is considerably smaller, I was thinking of the el cheapo for that. It also might be the best place to start, but does need the additional PS trafo for the negative rail. Advice would be appreciated. I probably only have money for one project for the time being, and therefore try to maximize the cost/unit awesome.
edit: this amp will be paired with some of Paul J's wildburrospeakers in a resonant cabinet.
So this is really my first foray into tubed gear, and after spending a whopping $90 so far I've got
1. a eico tube tester
2. a pretty nice magnavox PP el84 console pull (3 chokes in the PS)
3. a decent motorola SE el84 with a PP 6v6 summed output with HP for the bass channel
4. an Eico ST40, and
5. this Scott.
Since the iron in the magnavox is considerably smaller, I was thinking of the el cheapo for that. It also might be the best place to start, but does need the additional PS trafo for the negative rail. Advice would be appreciated. I probably only have money for one project for the time being, and therefore try to maximize the cost/unit awesome.
edit: this amp will be paired with some of Paul J's wildburrospeakers in a resonant cabinet.