Schiit Saga vs Freya

[zackthedog]

Well, the stock tube I received from Schiit with the Saga, produced some crackling sounds and noise (in both channels) which generally means the tube is at the end of its life.. I replaced it with another Sovtek 6SN7GT NOS tube and the problem was solved. The Saga sounds brillant. I don't expect the Tung-Sol to sound better (that's 0% improvement), but NOS tubes appear to be less than 100% reliable. Hence, the new Tung-Sol..

 

Actually, I'd expect more reliability with NOS tubes than new ones. :-) I've compared 6SN7s for years now and I always come back the the GE 6SN7GTB. A solid, rich and balanced presentation. You can get matched pairs for about the same as a premium new 6SN7.

[zackthedog]

Well, the new Tung-sol tube certainly sounds different from the NOS Sovtek tube you get standard with the Saga. I think The Tung-sol sounds better..

Oh, I agree. The Sovteks are pretty blah.

[zackthedog]

zackthedog, did the premium new you tried include the "The Shuguang Treasure" (why do I feel guilty even typing that out? ) or the Sophia Electric?

 

I'm not that crazy about the Treasure/Psvane 6SN7. It's a nice tube, and it certainly depends on your system. A lot of people love them. They do have a big, open sound, similar to the Mullard ECC32. I guess I should try mine again. :-) But the quality control is so-so and they can be noisy and microphonic, the last thing you want in a preamp. I find that the GE 6SN7GTBs do everything very well for a reasonable price. Production was very consistent, they'll tolerate high voltages, microphonics are extremely low and they're generally tough as nails. NOS or very strong matched pairs go for $60 or less.

[zackthedog]

Hi Zackthedog. Do you mean the early 1960s GE 6SN7GTB: GE 6SN7GTB - matched pair? What about this one: GE 6SN7GTb: Vacuum Tubes & NOS Tubes | Tubes Unlimited? And are you familiar with the GE 6SN7WGTA (1968-1970): 6SN7WGTA General Electric - Tubes: Odd Ball Tubeampdoctor Store? Thanks.

 

I mean the first pair, yes. Better to avoid the later, wafer-base version, though they would probably be okay in a pinch. The 6SN7GTA is a different animal. They look exactly the same but believe it or not they sound a bit different, with a more prominent midrange and little less open on the high end. They also have a lower maximum voltage rating. That's a nice mil-spec model there, never seen those. But apart from the brown base I doubt it's very different from the stock versions, and I wouldn't pay that sort of premium for them.

[zackthedog]

Thanks! I ordered an early 1960s G.E. 6SN7GTB for the Saga. Should be here this week.

I hope you like it. :-) Let us know what you think.

[zackthedog]

Saw these online- what does "full base" mean? Still fit Freya?

 

I'm not sure which tube you're talking about, but the Treasure/Psvane 6SN7s tend to have a larger base than NOS tubes. I see that the Freya has its tube sockets recessed below the cover, and the tube base must fit into the hole in the cover. This could be a problem with the Treasure/Psvane tubes, but I couldn't say for sure. I would ask the folks at Schiit for their advice before you buy any. Any NOS or standard new-manufacture 6SN7 should fit, since the bases conform to standard specs.

 

"Full-base" means the tubes have the standard base that wraps around the bottom of the glass envelope. "Wafer-base" or "Coin-base" means that the full base has been eliminated and replaced with a thin phenolic wafer that holds the pins and seals the bottom of the glass envelope. This was done in the late days of vacuum tubes to save money, I assume. Is that what you were asking?

[zackthedog]

The Psvane CV181-T mkII/6SN7 does fit in the Saga (just). The glass bottle almost touches the cover.

 

That's good to know, thank you.

[zackthedog]

Oh, and the new Psvane has a very pleasant sound; well-balanced/coherent, open and dynamical. Bass, highs, texture and details are great. The Tung-sol 6SN7GTB (new/reissue) is also quite open sounding but sounds relatively harsh. I think the Psvane is in a different league soundwise..

 

Yes, they can be quite nice depending on your system. The only other new-issue 6SN7 I have is the Electro-Harmonics, which is quite coarse-sounding by comparison.

[zackthedog]

Two tuby newby questons:

 

1) Is the Psvane your referring to the same product as a "Shuguang Treasure CV181-Z"?

 

2) #1 (rather the same or two different products) is a different product altogether than a "Sophia Electric 6SN7" correct?

 

I received my Saga yesterday and with the factory tube my first impressions fall into these sorts of thoughts:

 

"who stole my subwoofer?"

"what happened to my tweeters?"

"why do most recordings sound metallic, brassy, distant, etc.?"

"why do some (but not all) voices sound a hair more "natural", and why do brass instruments (trumpet, etc.) also seem to benefit sometimes from this metallic coloration?"

 

I am sure none of this surprises you guys...

 

Yes, the Chinese premium 6SN7 is the CV181. I don't know that there's a huge difference between the Treasure and the Psvane, except possibly for selection, but I could be wrong. I *think* the Sophia is different, since Richard commissions his own tubes, but again I'm not certain. I have a feeling the Sophia might be better. A friend of mine has these and likes them very much. I understand they really do come close to the Mullard ECC32.

 

As for your Saga, I'd imagine some break-in time is in order. :-) But those Russian 6SN7s are not very good-sounding. I even have some of the supposedly "better" NOS ones, from a certain year and factory, but they're still unlistenable to me.

[zackthedog]

Yes, I also found the new Tung-sol sounding better than the factory NOS Sovtek tube. And the 6SN7 is a double triode tube. I guess T1 stands for the first triode and T2 for the second. In the Saga the 6SN7 isn't used for amplification/gain so the two triodes being precisely 'matched' is not that important. Anyway, good to hear your problem is solved.

 

Well, er, I wouldn't say that. ;-) If you want the output impedance of both channels to match precisely, I think matched sections would matter.

[zackthedog]

Hi Zackthedog. As you know, the electrical characteristics of a vacuum tube are commonly described by three variables: voltage gain (Mu), transconductance (Gm) and plate resistance (Rp). In the case of triodes these are related: Mu = Gm x Rp. Mu and Rp are most important, where Rp directly affects output impedance. However, I understand that Mu and Rp are very difficult to measure and most/all vendors measure only Gm. With two triodes matched only for Gm (Mu/Rp), there is no assurance whatsoever that Mu and Rp will match. It seems to me this practice of matching tubes (triodes) is questionable and largely a marketing ploy..

 

I think you have a point, though I don't agree that matching sections is a "ploy". In a phase-splitter, a poorly-matched dual triode is going to affect the signal balance. The very least you can do is match the halves for Gm, even if further testing would involve more time and effort. In a cathode-follower, which I assume is being used here, it's probably less critical, but I'd still want to know that the two sections were performing somewhat equally.

 

ETA: Here's a good overview of the "matching" issue that sums up the issue pretty well. It addresses your concerns about the limits of tube matching.

 

http://tubesound.com/2010/10/27/tube-matching-with-a-tube-tester/

[zackthedog]

Thanks for the link. Do you think matching Gm of the two sections of a *single* dual triode tube is more likely to indicate matched Mu and Rp, than would matching Gm of a *pair* of tubes?

 

BTW, As of yet I didn't receive the G.E. 6SN7GTB (1960s vintage). Should be here by Monday. I did try a new Shuguang Treasure CV181-Z in the Saga and it sounds different from its cousin, the Psvane CV181-T mkII. A bit more lush perhaps. I like it.

Well, that's an interesting question. The original Williamson amplfier design used four 6J5s instead of two 6SN7s. (the 6J5 is equal to one half of a 6SN7). I'm not sure of the reason for this, but one advantage would be that you could carefully select four matching 6J5s rather than search through batches of 6SN7s for well-matched halves, if you see what I mean. So, for instance, if the Saga used two 6J5s instead of one 6SN7, you could select your 6J5s for a better match. :-)

 

Which raises another interesting point: You *could* replace the single 6SN7 with a dual 6J5 adapter (you can get these on eBay) and use two 6J5s--a lovely tube, BTW. You'd probably need a socket extender (also available on eBay) because of the recessed socket. Just another tube-swapping option to drive you nuts. :-)

[zackthedog]

Well, I was thinking that because the dual triodes of (e.g.) a single 6SN7 tube are obviously of the same make/series/date, matching Gm may yield a relatively high probability that other variables (Mu and Pd) are also matched, whereas selecting two different tubes for matching Gm may have a lower probability of being matched for Mu and Pd (where Gm = Mu/Pd).. Just a thought..

 

That might well be, yes.

[zackthedog]

Sorry for the (slight) delay. I received the GE 6SN7GTB (1962 vintage) tube for the Saga. To be precise: a GE 6SN7GTB (EC33) red label, like this one:

 

.........................................[ATTACH=CONFIG]33173[/ATTACH]

 

Gm = 2400 (a dimensionless number) for each triode section, and in my current system it's the best tube I've tried; better than the Psvane and Shuguang.

 

Here's my personal (preliminary) Saga tube review:

 

- Sovtek 6SN7GT (factory 1970-80 vintage): Nice but relatively dull

- Tung-sol 6SN7GTB (new): Relatively harsh but quite open sounding

- Psvane CV181-T mkII (new): Well-balanced/coherent, open and dynamical

- Shuguang Treasure CV181-Z (new): Like the Psvane, well-balanced/coherent, open and dynamical, but a bit more lush sounding

- GE 6SN7GTB red label (1962 vintage): Best I tried so far, very dynamical, very well-balanced, very coherent, very good bass and mids and smooth highs

 

The GE tube really rocks, and I don't think this is an artifact of my system (although it does seem to tame the KEF's highs a bit). It sounds very natural and authoritative in every frequency range and genre. Also, of the tubes I tried, it sounds the most different from the Saga's 'passive mode'. Wow!

 

Told ya! ;-)

 

Seriously, I'm glad you like it. I find that they convey a lot of energy without sacrificing balance across the range. That 2400 sounds like a Hickok measurement and is a solid NOS reading. Should last you a long time.

[zackthedog]

A really amazing tube! I highly recommend it for the Schiit Saga. The Saga with this tube sounds better than anything I've tried in terms of pre-amplification (passive or solid state or other tubes) in my current (and previous) system(s) so far. There may exist better triodes, but the performance of this GE takes my current system to a level of realism in sound reproduction that I actually didn't think was possible. Thanks again!

 

On the other hand, this result makes me wonder whether solid state (pre)amplifier designs are somehow fundamentally 'flawed', and if so, and if (currently) the best audio tubes are NOS tubes, then there may be a supply problem in the (near) future!? I'm interested to hear experiences from Freya users who compared the JFETT buffer to (NOS) tubes..

 

They don't make 'em like they used to--not that there aren't some good new production tubes, but still. The GEs are pretty plentiful, but it wouldn't hurt to buy a few spares. :-) I doubt they're run very hard in the Saga, so they ought to last a few years.

[zackthedog]

GE 6SN7GTB - matched pair

 

They appear to be out of stock And the picture doesn't show the red labels I received..

 

Looks like you started a run. :-) But seriously, you can get them on eBay from well-rated dealers. I like Viva Tubes, myself. His tubes are always as advertised. Tube World also has them.

 

ETA: They were also rebranded as Westinghouse and Motorola.

[zackthedog]

the Freya is tube driven

 

Are you sure? I'm just asking, because "tube-driven" would indicate, to me, that the tube is an active gain stage, as opposed to a cathode-follower buffer stage. It would make a difference.

[zackthedog]

The GE 6SN7GTB 1960s tubes come in slightly different models. The two I have are the same as called "NOS NIB GE 6SN7GTB Gray Flat-Aligned Plate O Get" by Viva Tubes and have an O shaped (getter?) structure between the sections like visible here (right tube):

 

................................[ATTACH=CONFIG]33236[/ATTACH]

 

Do you think these differences matter soundwise? And if so, do you have a preference for a specific model?

 

I haven't found a big difference between the slight variations aomng the GEs. IMO, it's the plate structure that accounts for the biggest difference, and that is consistent across the GE models.

[zackthedog]

Freya description on product page: "Switch between passive, active JFET buffer, and tube gain modes".

 

Other Schitt quotes: "or choose the tube gain stage—a serious gain stage running on 300V rails. "; "employ highly linear 6SN7-type tubes on 200 or 300V rails in the active stages"

 

Schiit Freya Impressions and Tube Rolling Thread

 

Yes, you're right.

[zackthedog]

I agree that it would make a difference, but I do not see how "tube driven" would imply a gain stage over a buffer.

You're right. It's really the same thing. Don't know what I was thinking. :-)

[zackthedog]

I have no experience with a Tube preamp. Do preamp tubes need bias adjustments like a tube power amp?

No, a small signal tube like a 6SN7 is typically going to be used in a self-biasing circuit.

[zackthedog]

So I'm very impressed by the performance of the Saga's stepped resistor-based attenuator with active tube-driven buffer stage. Nevertheless - in my system - I need to 'lift' the protective mains ground/earth connection of the Saga (and for that matter of all my audio gear) in order to get this sound quality. Apparently, mains ground over here is rather dirty and somehow interferes quite heavily with the audio signal. Anyhow, I'm a bit worried that lifting protective ground/earth of the Saga might pose a more than average health risk, in view of the high voltage circuitry associated with tube amps? Just a personal concern..

Not to be alarmist, but that would be a concern, also the safety of your home in the case of some catastrophic event. You might want to look into a passive filter that might clean up the AC, or even an active power conditioner.

[zackthedog]

Thanks. Something in my neighborhood or even in my home (though I tried disconnecting everything) is rather heavily leaking current/voltage into the common mains ground (basically the water/heating system's copper tubing). I'm not sure whether a power conditioner will cure that, but will look into it. I'm also considering a dedicated earth connection (pin in soil) for my audio gear..

 

I hear you. I have an older house with very outdated wiring. I'm screwing up my courage (and my wallet) to have an electrician install at least one proper circuit to the AV room. No matter how often I mark the breaker board, replacing a light fixture or anything like that requires that my whole family station themselves around the house while I switch the breakers to see which outlet or ceiling fixture goes to what breaker. A single outlet in the back bedroom might connect to the kitchen, and so on. It's maddening. :-(

[zackthedog]

Thanks for the review Abtr. I have a Sophia Electric tube on the way, but I will no doubt have try one of these GE tubes as well - my Legacy Studio HD's are tipped up on the upper end too much as well...

 

I have a question: on the tube power amp side, how does Rogue Audio get 100 wpc from their quad KT120 setup (Cronos Magnum II) when everyone else seems to get 60-75 wpc? Are Rogue doing something legit to get the extra power or are they sacrificing tube life playing the numbers game? If this has already been discussed please point me to the link (I could not find a thread that addressed this)...

 

I don't know about the Rogue amp. It's certainly possible to get that from a pair of KT120s but I have to think they're driving them pretty hard. VTL used to run the daylights out of EL34s...

[zackthedog]

Well, according the manuel online at Rogue's site:

 

"Using the bias tool, slowly turn the screw on the potentiometer that is adjacent to theswitch until the bias meter reads ~ 35 miliamps (mA)."

 

Is 35 mA a 'hot' idle current setting for the 6550?

 

update: I am not trying to denigrate Rogue here - in fact I like the sound I have heard from this amp when I have demo'd it in the past and could easily see myself owning one despite the fact that I might be investing in tubes more often - I just am trying to get a grasp on what the design is really doing...

 

It's a Class AB design so your idle current is going to be pretty low. To get 100 watts out of a *pair* they're going to be running a pretty high plate voltage and your peak currents are also going to be high. But I doubt it's really a problem.