| leadbelly |
I should explain the cryptic title: do all you die hard tube fans go tube all the way through the chain, e.g. tube RIAA or tube DAC all the way to OPT? Or is there a point where, just as many concede that silicon is better for power supplies or CCS's that the front end/low level stuff is "better" left to silicon (I know "better"is no absolute). I have read many opinions on the former issue but none on the latter.
And I ask this for practical reasons for myself. I am building myself a full audio chain and love the idea of it being balanced throughout and tube throughout, but the complexity of tackling a balanced tube RIAA preamp and a balanced tube DAC frighten me. I'm pretty sure I must choose to build either an SE tube preamp or a balanced SS preamp, and similarly for the DAC. What do you recommend or what do you have preference for? |
|
|
| Sch3mat1c |
RIAA is easy. For balanced, just make two identical networks and run diff amps fore and aft.
Hmm, if the signals are 180ø out of phase, the phase shifts in the network should just rotate around the circle and not cause any unusual common-mode cancellation, no?
As for a DAC, someone look at the digital standard (SPID/F? or whatever it is) and see how hard it would be to decode.
Depending on how many logic chips you want to use, I doubt it'll end up much worse than that all-tube clock.
Tim |
|
|
| leadbelly |
| quote: | Originally posted by Sch3mat1c
RIAA is easy. For balanced, just make two identical networks and run diff amps fore and aft.
As for a DAC, someone look at the digital standard (SPID/F? or whatever it is) and see how hard it would be to decode.
|
So you are in the "go tube AND balanced" camp I see :)
I dunno about balanced RIAA, Morgan Jones' schematic looks BIG to me.
Tube balanced DAC is easier, the KISS way is to just invert the data stream and have 1 DAC chip do the + phase and the other chip do the - phase. |
|
|
| SY |
It's difficult to deny that silicon is always going to be superior for constant voltages and costant currents. The laws of physics are on its side.
Personally, I'll use silicon for signal handling in a limited set of circumstances, mostly source followers. See the SYclotron, for example. |
|
|
| analog_sa |
| quote: | | low level stuff is "better" left to silicon |
GIven a forced choice i would rather leave the hi level 'stuff' to silicon, somehow i find the silicon signature more objectionable at the millivolt/microvolt levels. It is all, of course, very relative - i have a reasonably musical all-SS system which doesn't , surprisingly, annoy the hell out of me. Otoh, even a little silicon in my all-tube system has the ability to stick like a sore thumb. The final result is very much a question of mix and match. And not an easy mix to match right. |
|
|
| Bill*B |
| A sensible man would keep silent, - but - here goes (my opinion only). You can get perfectly good results with either vacuum or sand in audio. Sand is more difficult to design well, and the construction is different. I like tubes because a) I understand them, b) they are rugged enough not to need special care in handling, and c) the glow of their filaments transports me to a simpler time, and soothes my soul. Regards, Bill. |
|
|
| SY |
| Bill, you're a sensible man. There's no room for that in audio!;) |
|
|
| nuvistor |
| What about mV level with SS (JFET or bipolar) input and tube cascode with resistive load? :xfingers: |
|
|
| thoriated |
| My opinion is: Anything you can build/maintain yourself, go with all tubes, with the possible exception of the use of power rectifiers. With a format that's likely to be obsoleted in a few years or is CD quality or worse, making the final effort to do it in tubes may not be worthwhile. |
|
|
| Sch3mat1c |
| quote: | Originally posted by leadbelly
So you are in the "go tube AND balanced" camp I see :) |
Well I was just suggesting a solution for you... but now that you've dragged me into it, I KISS.
I don't really care, so I listen to 128k MP3's on my computer (and am proud of *enjoying* what you would be led to believe is the worst media ever!) and send that through my tube preamp and monoblocks (Frankenhouse and Hept'AU7).
Balanced specifically has no purpose in a home system, maybe for a generally large system like a club, church or PA system in general.
It does make PP easier, but then again, you have all those extra stages that *could* be rolled together into just one or two stages. And you get into trouble when you have someone who prefers SE output.....
| quote: | | I dunno about balanced RIAA, Morgan Jones' schematic looks BIG to me. |
I don't have his book, but balanced, like stereo, can easily get unweildy very quickly. Come to think of it, if you're going to do stereo RIAA, you need essentially four channels!
| quote: | | Tube balanced DAC is easier, the KISS way is to just invert the data stream and have 1 DAC chip do the + phase and the other chip do the - phase. [/B] |
If you just want balanced for transmission, the KISS way would be to use one DAC and split it at the output, but...
Tim |
|
|
| anatech |
Hi leadbelly,
I have both and enjoy both. I even have hybrid amps. I've heard terrible noises coming from all technologies. I don't think you can easily, blindly go with one or the other.
So design with what you know you can do your best with. I must say there are places where sand is king. Agreeing with SY, current sources are one of those things. I'll throw in rectifiers 'cause I use good regulators. Sand again.
There is no reason at all to go balanced. Your cable runs aren't long enough to warrant it. Your system noise goes up by 1.414 in theory because you doubled your noise sources, no they won't cancel. So here I agree with Sch3mat1c. Just build it.
-Chris |
|
|
| leadbelly |
| Wow, this thread turned out different than I expected. I was expecting more of the prose that sells magazine copies, and I don't mean that negatively. So, to recap: "tubes good everywhere, sand good sometimes"? :) |
|
|
| Sch3mat1c |
| quote: | Originally posted by leadbelly
I was expecting more of the prose that sells magazine copies |
Pffbt. If that were the turnout, you wouldn't believe how many times "****" would keep coming up in my posts.
:D
Tim |
|
|
| anatech |
Hi leadbelly,
I'd say more. Tubes good sometimes. Sand good sometimes. Depends on the circuit and signal type / level. You do have to consider how complex the circuit may get and how much heat you are willing to radiate.
-Chris |
|
|
| arnoldc |
| in my system, i'm currently contented with a solid state phono stage, tubed-preamp, and one of my flea powered SET amps. |
|
|
| leadbelly |
| quote: | Originally posted by anatech
You do have to consider how complex the circuit may get and how much heat you are willing to radiate.
-Chris |
Hi Chris.
It's true that I did have a question as to which direction I should head off in, but I was genuinely interested to see if there were any interesting technical arguments on the subject that I had missed. I don't really have a strong preference either way, the biggest reason I've started on tubes is to learn something new. I built my first chipamp 20 years ago and there's is a point where you have to move on out of boredom. As for what I will likely do, I haven't made up my mind yet. A big balanced phono pre is out of the question, I don't have that much vinyl to warrant it. I have enough parts bin parts for an SE phono pre, so that's looking interesting. I have cheap Edcor transformers, a 4 deck ALPS switch, and parts for a JFET balanced line stage, and am not sure I know how I want to put that all together. Maybe the only way is through hindsight :) |
|
|
| Brett |
My preference, based on building a lot of amps, is for tubes for the signal path, and sand for the supporting elements, most of the time. I also prefer differential tube stages over SE.
However the lower active in my cascode hybrid RIAA front ends is sand and it's superb. I'm also quite liking the newer digital amps for power; not as good as the best tube stuff ime, but good enough much of the time. |
|
|
| planet10 |
| quote: | Originally posted by leadbelly
I dunno about balanced RIAA, Morgan Jones' schematic looks BIG to me. |
Allen Wright's balanced phono stage is elegantly simple (power supply is quite a bit more complex). A phono cartridge is inherently a balanced device.... A balanced phono stage with balanced wiring in the tone-arm gives BIG advantages, and is the only way to properly select absolute phase for each track. The phono stage is probably the place where the most benefit can be had from a balanced stage...
(you can ignore the stuff after the selector switch -- most of the stuff after is to drive the cable to your power amp... just put your power amp in the same box... ie phono, diff stage, PP output stage.)
dave |
|
|
| richwalters |
Hi there........the RH side driver is not my type of balanced stage. As an ex Neve_ite, it may be more expensive but following tradition I love to put a tranny in this position to 600 ohm and avoid the clobber.
richj |
|
|
| planet10 |
| quote: | Originally posted by richwalters
........the RH side driver is not my type of balanced stage. As an ex Neve_ite, it may be more expensive but following tradition I love to put a tranny in this position to 600 ohm and avoid the clobber. |
I tend to agree... you can also just put the amps output stage there too.
dave |
|
|
| DanTana |
| Is there really any right or wrong way to audio design? It all comes down to, what sounds best to you. |
|
|
| leadbelly |
| quote: | Originally posted by Sch3mat1c
RIAA is easy. For balanced, just make two identical networks and run diff amps fore and aft. |
Is this right? I cannot for the life of me find any reference that explains whether this is the case or not. |
|
|
| planet10 |
| quote: | Originally posted by DanTana
Is there really any right or wrong way to audio design? It all comes down to, what sounds best to you. |
Definitly some wrong ways (ie when the design lets the smoke out), but the diversity of possible designs is what makes the field so rich.
dave |
|
|
| SY |
| Leadbelly, the howler is the "just" in that quote. It's much easier to connect a single network differentially. |
|
|
| planet10 |
| quote: | Originally posted by SY
It's much easier to connect a single network differentially. |
An example of that is in Allen Wright's circuit... in his Preamp Cookbook he goes to some length explaining that circuit which has all sorts of subtle tricks in it.
dave |
|
|
| leadbelly |
| quote: | Originally posted by SY
Leadbelly, the howler is the "just" in that quote. It's much easier to connect a single network differentially. |
What I didn't write but was thinking is that this might work for opamps.....I can't find any theory on doing balanced RIAA with opamps!
| quote: | Originally posted by planet10
An example of that is in Allen Wright's circuit... in his Preamp Cookbook he goes to some length explaining that circuit which has all sorts of subtle tricks in it.
dave |
It's a nice circuit, but I'm not too crazy about it. If I read it properly, it is passive RIAA after a differential gain stage, so discarding everything after the selector switch probably nets you a weak signal that needs gain somewhere later, no? |
|
|
| planet10 |
| quote: | Originally posted by leadbelly
It's a nice circuit, but I'm not too crazy about it. If I read it properly, it is passive RIAA after a differential gain stage, so discarding everything after the selector switch probably nets you a weak signal that needs gain somewhere later, no? |
Yes, but you already have that in your differential 6922 line-stage (ie if you ignore that monster cathode follower at the end your pre is pretty close to the middle stage. Replace the cathode follower with your poweramp output stage and you have a very elegant integrated amp.
dave |
|
|
| leadbelly |
| quote: | Originally posted by planet10
Yes, but you already have that in your differential 6922 line-stage (ie if you ignore that monster cathode follower at the end your pre is pretty close to the middle stage. Replace the cathode follower with your poweramp output stage and you have a very elegant integrated amp.
dave |
OK, you win, I will build that one. :) |
|
|
| planet10 |
| quote: | Originally posted by leadbelly
OK, you win, I will build that one. :) |
And it has a nice mix of silicon & glass... you can get the matched FETs from Joe Rasmussen. Keep in mind that the PSRR on the phono stage is quite poor so you need to have a really good power supply....
dave |
|
|
| leadbelly |
| quote: | Originally posted by planet10
And it has a nice mix of silicon & glass... you can get the matched FETs from Joe Rasmussen. Keep in mind that the PSRR on the phono stage is quite poor so you need to have a really good power supply.... |
Actually, I already have the FET's, one big reason I will build this is that I have the tubes, FET's, and a cheapo transformer. I just have to make the effort to match the FET's as close as possible. As for power supply, what I was planning was a voltage doubler with 1000uF caps and then Morgan Jones' LM317 regulator. I think ripple will be resonable with the bigger caps. I don't want to make it integrated with the power amp but put it in its own enclosure. |
|
|
| gingertube |
I came to tube audio late in life after 35 years in Electronics - the last 15 as a Senior Design Eng. If I had to point a finger at tubes, it would be at their noise performance (which is NOT great). I therefore have always used sand for low signal level stuff like phono preamps.
A caveat on that is that the ONLY tube phono RIAA preamp I've heard is the ECC807 based preamp in my little Rogers Cadet III.
The possible saving grace of tube RIAA preamps is that the RIAA equalisation de-emphasises high frequencies where noise is likely to be most objectionable. However it will emphasise 1/f "flicker" noise. It is also easier to control impedances (especially over an extended lifetime) and therefore obtain accurate RIAA response using sand.
And then there is cost - of course if you were truely worried about cost you would'nt be looking at this forum at all.
On balance its sand for RIAA preamps for me.
Just my opinion for what its worth, and I'm sure there will be plenty of folk to disagree with me.
Cheers,
Ian |
|
|
| planet10 |
| quote: | Originally posted by gingertube
If I had to point a finger at tubes, it would be at their noise performance (which is NOT great). I therefore have always used sand for low signal level stuff like phono preamps. |
The prime reason for the FETs on the bottom of Allan's cascode.
dave |
|
|
| EC8010 |
| quote: | Originally posted by leadbelly
As for power supply, what I was planning was a voltage doubler with 1000uF caps and then Morgan Jones' LM317 regulator. |
You'd do better using all of Allen's design by using his power supply which was optimised for his gain stages. The 317 is a jolly handy little device, but at the end of the day it's a 741 plus sluggish power transistor with a few bells and whistles to turn it into the regulator. It's possible to do better... |
|
|
| richwalters |
| quote: | Originally posted by gingertube
If I had to point a finger at tubes, it would be at their noise performance (which is NOT great). |
Hey No excuses..(SDEng)...what's up.. where's your physics ?Some tubes aren't so bad with the noise...ok some are bad offenders . Designing a line /low end preamp with -92dB noise figure. This used to be an early console requiremnt.. Naturally there's a tranny or two stuck in the armoury but it shouldn't be beyond the physics to get to that low noise figurê. The design certainly isn't space conscious as silicon is.
Don't try it with PSpice or other tube simulated program.....modern programs are hopeless at equalising the noise sources. Unfortunately there is a problem in that most parts of the circuit is patented thus cannot copy it...
Some parts aren't and sometime I will paste the relevant.
richj |
|
|
| wa2ise |
| quote: | Originally posted by leadbelly
I should explain the cryptic title: do all you die hard tube fans go tube all the way through the chain, e.g. tube RIAA or tube DAC all the way to OPT? Or is there a point where, just as many concede that silicon is better for power supplies or CCS's that the front end/low level stuff is "better" left to silicon
|
For some reason when I see the word "DAC" I think of the actual DAC chip, but here I think you mean the outboard DAC box you connect to the CD transport. Well, obviously the DAC chip has to be silicon, but at the point that you first have analog, that could be tubed. See my
web page of my various tube modified CD players I have hacked
----------------------------
My first time on this forum, though I'm a regular at AudioKarma. |
|
|
| SY |
| A tube DAC is possible (though not trivial). And it would be quite cool. Or hot. Well, you know what I mean. |
|
|
| leadbelly |
A very interesting page, the only one where I have seen the various common DAC output stages all tried. You don't really seem to compare them for quality. I am planning to build a balanced tube DAC with TDA1541A (no, really!). Based on your experience, would you think that the more promising path forward would be to derive the balanced signal in the digital domain with data inversion or to combine I/V and unbalanced/balanced all together in one step using a transformer?
| quote: | Originally posted by SY
A tube DAC is possible (though not trivial). And it would be quite cool. Or hot. Well, you know what I mean. |
I was thinking exactly the same thing just recently when I cam across an old Circuit Cellar where Steve C discussed roll-your-own DACs. But who has the energy! :) |
|
|
| Sheldon |
| quote: | Originally posted by richwalters
Unfortunately there is a problem in that most parts of the circuit is patented thus cannot copy it.
richj |
The whole point of patents is to give the inventor incentive to publish the invention. The inventor discloses the invention and in return is granted the right to prevent others from making or using the invention for commercial purposes, for a set period. Any information in a patent can be freely shared.
Sheldon |
|
|
| EC8010 |
| quote: | Originally posted by leadbelly
Based on your experience, would you think that the more promising path forward would be to derive the balanced signal in the digital domain with data inversion or to combine I/V and unbalanced/balanced all together in one step using a transformer? |
Do the phase splitting in the digital domain. It's a while since I've looked at digits, but I vaguely recall that to get a true analogue inversion of 2s complement notation you need to add or substract 1LSB. Or something. I could be wrong, but it might be worth checking.
More to the point, there are all sorts of linearity tricks that you can pull once you have both polarities of the digital signal and a handful of DAC chips. Have a look at some of the schemes used in the Burr-Brown (now Texas) data sheets.
There's a lot to be said for using a transformer. I haven't yet tried it myself, but the ground-breaking ability is important. |
|
|
| leadbelly |
| quote: | Originally posted by EC8010
More to the point, there are all sorts of linearity tricks that you can pull once you have both polarities of the digital signal and a handful of DAC chips. Have a look at some of the schemes used in the Burr-Brown (now Texas) data sheets. |
GREAT tip! Plenty of stuff mentioned in the datasheet for DAC2932, which isn't an audio DAC so I would never have gone to look. Thanks! |
|
|
| SY |
| quote: | | There's a lot to be said for using a transformer. I haven't yet tried it myself, but the ground-breaking ability is important. |
Indeed. As is the ease of getting excellent balance over the entire audio band.
Isolating the grounds at the input of my power amp was the single greatest quality difference in electronics that I've observed in my system in quite a few years; reduction of even subliminal hum seems to be a great positive for clarity. And, of course, putting the transformer at the input means that the signals are handled at a low level and that feedback may be more easily applied than is the case for interstage transformers.
(NB: This is not OT- my tube amp's input is a pair of pFETs.;) ) |
|
|
| DSP_Geek |
| quote: | Originally posted by EC8010
Do the phase splitting in the digital domain. It's a while since I've looked at digits, but I vaguely recall that to get a true analogue inversion of 2s complement notation you need to add or substract 1LSB. Or something. I could be wrong, but it might be worth checking.
|
Give the man a cigar!
You're spot on the money. Let's look at an 8 bit system, for example. With 2's complement math, which is almost universally used, 45 decimal = 00101101 binary, and -45 decimal = 11010011. Note that adding up the two numbers gets you 100000000, which is 256 decimal.
Now having to make sure that addition comes out correctly might become a real pain in the neck, but the error caused by merely inverting the data turns out to bias the output by 1 LSB in the positive direction. If you're passing 24 bit or even 20 bit data, that's literally lost in the noise. In other words go right ahead - let the math purists *****, fewer moving parts mean less stuff to break.
Francois. |
|
|
| wa2ise |
| True that not adding an LSB 1 after inverting the bits will create a slight error, but the error is small and would be a DC offset anyway. Of no significance in audio. If you're using a serial bit stream DAC chip all you'd need is a single inverter. Some DAC chips have differential outputs already. Or you could construct a pair of SRPP circuits with both bottom triode cathodes tied together and fed by a constant current or a big resistor connected to a very negative supply. One grid of the bottom tubes goes to ground, the other grid of the other bottom tube goes to the DAC output IV resistor. Both top tube cathodes will produce a differential pair of signals. |
|
|
| amperex |
| The only SS I used is a B+ shunt regulator placed after an LC feeding another LC with an oil cap on the output. |
|
|
|
