SET Asylum

Single Ended Triodes (SETs), the ultimate tube lovers dream.

Return to SET Asylum

Message Sort: Post Order or Asylum Reverse Threaded

what you find best around the 6SN7 driving 2a3 SE

110.32.200.254
Posted on September 16, 2013 at 02:43:33
lovetube
Dealer

Posts: 3008
Location: Melbourne Australia
Joined: June 8, 2003
 i have 3 circuit in mind . SRPP , 3 stage direct coupled between the 6SN7 something's like the Sun-audio circuit or paralell 6SN7 drive the 2a3 .
which would you think best for the 2a3 ?
regards
LT

 

Hide full thread outline!
    ...
RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 16, 2013 at 08:19:55
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 A single half of a 6SN7 will drive a 2A3 pretty well, but you'll want a preamp behind it! Use a choke or CCS to get every drop of gain out of the 6SN7 that you can.

The 2-stage 6SN7 directly coupled driver may be a decent choice if you're building an integrated amp.

I wouldn't personally go for the SRPP, but some people swear by it.

 

A heretical view, posted on September 16, 2013 at 13:41:52
andy evans
Audiophile

Posts: 4443
Joined: October 20, 2000
 I wouldn't use either the 6SN7 or the 2a3. I'd use 4P1L into PSE 4P1L, on condition both stages were in filament bias and coupled by a good interstage - currently using LL1660/18mA. Two directly heated stages, no coupling caps, no cathode bypass caps. But a really good DC filament supply for both tube stages.

2 stages work in the case of the 4P1L because the mu of the tube is around 10, so with sensitive speakers and 2v in, the gain is adequate. Simplifies a lot of things. Sounds better than any SE 2a3 or 300b amp I've built for all the above reasons. The 4P1L is an astonishing little tube.

 

RE: A heretical view, posted on September 16, 2013 at 14:11:03
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 You're like the antichrist of tube rollers ;)

(though I like your idea a lot, I might just use one 4P1L as a pentode at the input, either one should make for a nice little amp)

 

You need to make a jump away from conventional thinking!, posted on September 16, 2013 at 14:39:48
andy evans
Audiophile

Posts: 4443
Joined: October 20, 2000
 Ha! Antichrist now, and my grandfather was a preacher!

It took me a while to get my head around this kind of new thinking but my ears told me the way. For the output, it may well be the case that 300b and 2a3 are intrinsically better than the 4P1L, but maybe not by very much. But the tables change when you shackle either tube with a cathode bypass cap, and here the 4P1L gains because it can be used in filament bias. Furthermore, with a mu of around 3 you start to make sacrifices with the input tube and start using indirectly heated tubes like 6SN7 or even higher mu options. Gain of 20 times gain of 3 = 60 as a very rough calculation, so looks like a preamp. With the 4P1L, gain of 10 times gain of 10 = 100, and that's conservative - could be 11 or 12. So this could be enough with the right speakers and input. By the time you reason all this out, it makes sense to use 2 stages of 4P1L in filament bias. Heretical thinking maybe, but there are good reasons why it works. You could do the same with a PX25 (harder but not impossible in filament bias) with a mu of 9 or 10 but they are about 100 times the price. Makes you think.....

I haven't used 4P1L as a pentode - others report mixed results, and not as good as in triode mode. But the whole thing is I don't need the extra gain because I have 4P1Ls in the output. When you start to let go of low mu outputs like the 2a3 and 300b you're in another ball game.

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 16, 2013 at 16:04:27
gortnipper
Audiophile

Posts: 866
Location: Seattle expat in Auckland
Joined: November 28, 2003
 My Sun SV-2A3 is highly musical with great drive and extension. I love the sound it makes and I dont see myself selling it.

But, it is extremely sensitive @ 150mV, so good pre-amp matching is critical.

 

RE: You need to make a jump away from conventional thinking!, posted on September 16, 2013 at 18:04:42
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 What's the plate resistance of the 4P1L in triode?

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 16, 2013 at 19:20:14
hifipaul
Audiophile

Posts: 739
Location: NY
Joined: December 22, 2008
When I use an SRPP I put a choke between the the two tubes in place of the resistor. 2-3 henries will do it. DCR for the choke needs to be 500-1000 ohms so it biases right. Shielding is important too or you'll get cathode driven hum. Allan Kimmel wrote about similar circuits in VTV some years ago.

The sound of the circuit retains the strong bass of the SRPP, and alleviates the midrange glare some people dislike about the SRPP. It is one of my favorite circuits.

 

RE: You need to make a jump away from conventional thinking!, posted on September 17, 2013 at 01:22:11
andy evans
Audiophile

Posts: 4443
Joined: October 20, 2000
 Plate resistance of the 4P1L in triode should be between 1.5K and 2K depending on operating point. The Russian data refers to its use as a pentode, but almost everybody in audio uses it as a triode. Gain is somewhere around 12 depending on operating point.

It's a lot like a 45 but with about 3 times the gain. Max voltage around 260v, current up to 40mA, dissipation about 8 watts or 9 max. Use two of them in parallel and you're in similar territory to a slightly beefed up 2a3.

The big difference is you can use it in filament bias. The output section of my amp uses the parallel 4P1Ls at a bias of -19v. Add 2v for the filaments and you need to input 21v from the DC current regs (Rod Coleman's). Allow 6-7v for the current regs to function and you have a supply of 28v at 1.3 amps. That's not difficult. The bias for a 2a3 is more like -45v at 2.5A and the 300b more like -70v at 1.3A so clearly filament bias isn't very practical. So with them you're usually back to cathode bypass caps on the resistor which will affect the sound. In filament bias the resistor is only 15 ohms.

 

look like we all search for different goal here, posted on September 17, 2013 at 01:56:25
lovetube
Dealer

Posts: 3008
Location: Melbourne Australia
Joined: June 8, 2003
 thanks
LT

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 17, 2013 at 07:07:10
Palustris
Audiophile

Posts: 2459
Location: Cape Cod
Joined: September 12, 2008
 Why in the world would you use a preamp in front of an integrated amp with an input sensitivity of .150V? It has a volume control!

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 17, 2013 at 14:23:34
gortnipper
Audiophile

Posts: 866
Location: Seattle expat in Auckland
Joined: November 28, 2003
 My phono is integrated in the pre.

But, I even found that was the case with passive pre's wiht input selectors that careful matching was in order.

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 17, 2013 at 14:28:01
gortnipper
Audiophile

Posts: 866
Location: Seattle expat in Auckland
Joined: November 28, 2003
 >>A single half of a 6SN7 will drive a 2A3 pretty well, but you'll want a preamp behind it! Use a choke or CCS to get every drop of gain out of the 6SN7 that you can.

I have also heard this work very nicely with a 45 amp.

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 17, 2013 at 17:16:36
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 I can't tell if your Shindo VR has a tape out but you should not be using the line stage(s) with the Sun SV-2A3.

The Sun has plenty of gain and the stages in the Shindo VR (while very good I'm sure) will only add unwanted/unneeded noise and distortion.

If there is no tape out you could wire your own. Be careful about cable length, I don't think the phono pre output is buffered. You don't want to lose any highs.

The point where the phono output feeds the selector switch is the spot you want.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 17, 2013 at 19:38:42
gortnipper
Audiophile

Posts: 866
Location: Seattle expat in Auckland
Joined: November 28, 2003
 No, no tape out.

Thanks for the idea. I dont think I will ever use the "TV" RCA jacks...

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 17, 2013 at 20:45:21
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 Do you have a schematic?

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 18, 2013 at 01:20:10
gortnipper
Audiophile

Posts: 866
Location: Seattle expat in Auckland
Joined: November 28, 2003
 No, and I am sure the Mob would put a hit out on me if I posted a Shindo schematic.

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 18, 2013 at 05:56:44
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 Absolutely agree except that drive taken from the bottom tube's plate aka current sourced grounded cathode sounds better than taken from the cathode of the top tube IME. I also like to use higher inductance, which usually works out well coil resistance wise.

Best part of this circuit is huge linear voltage swing you can get out of it.

Naz

 

RE: You need to make a jump away from conventional thinking!, posted on September 18, 2013 at 08:03:24
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 You don't really need low bias voltage to use filament bias. You can directly couple the previous stage and run it from a negative voltage supply.

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 18, 2013 at 08:12:15
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 Have you tried using the cathode of the top tube while feeding the stage from a CCS feed shunt VR tube regulated supply?

Here's my point. One of the advantages of active loading is you take the power supply largely out of the signal path.

When you use the cathode of the top tube as the output you lose some of that power supply isolation.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Filament bias - designs?, posted on September 18, 2013 at 10:18:52
andy evans
Audiophile

Posts: 4443
Joined: October 20, 2000
 You don't really need low bias voltage to use filament bias. You can directly couple the previous stage and run it from a negative voltage supply.>>

How would you do that? The voltage on the cathode is determined by the filament supply going through the tube and down to ground. On a 2a3, if you put 2.5A through the tube and down to ground through the cathode resistor, the bias voltage appears on the top of the cathode resistor and you need about 10v DC more than the bias to feed the filament. Are you suggesting taking the cathode resistor down to a negative voltage or what? Could you give a quick diagram maybe? That would be most helpful.

Andy

 

RE: Filament bias - designs?, posted on September 18, 2013 at 11:01:37
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004







This is a total exaggeration of using filament bias, but just barely.

You could rebalance the voltages as you see fit.

 

anything but 6SN7, actually..., posted on September 18, 2013 at 11:09:17
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

I think it is important to define one's goals.

Tubes have certain sonic characteristics that appear to show through no matter how the tube is used. This means a specific construction (precise maker, year etc.) but also general types.

The next step is to achieve a desired outcome.

To wit, one may select tubes in ways that allow the resulting final circuit to have an even-handed set of virtues and vices with an even tonal balance.

I call that "all'round" tuning. Such amplifiers tend to suit wider ranges of partnering equipment and music.

Another option is to select tubes to emphasise a specific quality. Then you select tubes such to maximise this quality.

Let me give two examples.

Take a 300B Output Tube. It is one big fat mama of a tube. It has warmth, it has grove, it can rock'n'roll. But delicacy is not it's strength.

Couple it with a driver tube and circuit that is great on detail and possibly tends a little towards forwardness, get a very balanced sounding Amp.

As alternative use a tube and circuit that also have the same bend - now you have an amp that is so warm and syrupy, that I personally feel it dragging the music through molasses, yet with certain speakers and music the results are great, such an amp can make Lowther's enjoyable.

So, you are the Chef in that articular kitchen. You have defined the output (2A3 - single plate I hope - I always hated the chinese dual plate ones). Now define the sound attributes you want.

Some hints I would give.

1) Direct couple - direct coupled amplifiers recover instantly from overload, which becomes more likely the more the power goes down.

2) The 2A3 tends towards brightness/whiteness in sound quality. Many 6SN7 do so too. The Cascade with 6SN7 sounds sufficiently warm, but muddy. I can easily get more clarity from a 300B than from a 6SN7 cascade with 2A3, so best consider a tube that sounds slightly subdued if you want a balanced sound.

The "Baby O" Design using a 12AT7 (which is RC coupled, SRPP) and the J.C. Morrison 6SL7 SRPP Loftin White design from the Fi-Primer are examples of amplifiers that get a good balance and having heard builds of both, they make excellent amplifiers in my books. Personally I would likely favour the J.C. Morrison Amp for it's larger dynamic envelope.

If I had to use a 6SN7 like tube, I would probably go to the 5687 and use it as Cascade.

But horses for courses, or "different strokes for different blokes" as they say in the merry old land of Oz...

Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

RE: Filament bias - designs?, posted on September 18, 2013 at 14:42:48
andy evans
Audiophile

Posts: 4443
Joined: October 20, 2000
 Thanks for that!

Hmmmm.... I'm still trying to get my head round this. What's the voltage on the plate of the 2a3?

 

RE: what you find best around the 6SN7 driving 2a3 SE, posted on September 18, 2013 at 16:36:27
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 I've used shunt and series regs but not with VRs. Interestingly, I don't necessarily find that shunt is better than series, depending on topology and components used.

You make a good point and this could easily be a looong response but will try to condense my reasons.

1) You can't assume that a regulated supply of any kind will give you anything like perfect isolation. I've always been mystified by the sonic difference when changing a rectifier tube even if chokes and regulators are employed in the PSU.

2) You still get much better isolation by taking drive from the plate of the bottom tube, which must help to some extent due to point 1)

3) To my ears the CF nature is still audible the when taking drive off the cathode. This alone is enough to dominate my decision in most cases despite the higher OP impedance. If I need more drive I use a different circuit.

Naz

 

RE: Filament bias - designs?, posted on September 18, 2013 at 19:24:30
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
 If you wanted the classic operating point, 255V.

 

RE: anything but 6SN7, actually..., posted on September 19, 2013 at 20:47:13
Donald North
Manufacturer

Posts: 1296
Joined: February 8, 2001
 I agree with much of what you're saying, however I don't see how the 12AT7 is a better sounding driver than the 6SN7. To me the 12AT7 is on the bright/metallic side.

 

RE: anything but 6SN7, actually..., posted on September 19, 2013 at 23:31:06
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

No idea what 12AT7 you were using and how.

I have not found the 12AT7 bright/metallic, I am using it in some commercial designs precisely because it sounds quite warm and relaxed, but more detailed than the 12AX7...

It also does not sound like that in the Baby-O.

Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

RE: anything but 6SN7, actually..., posted on September 20, 2013 at 12:12:25
Frihed89
Audiophile

Posts: 15703
Location: Copenhagen
Joined: March 21, 2005
 Doesn't the 6SN7GT have lower distortion than either the 12AT7 or 5687?

 

RE: anything but 6SN7, actually..., posted on September 20, 2013 at 18:07:22
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 The 12at7 looks very non-linear to me. I can't see an operating point that would not cause the negative half of the voltage output to be much larger than the positive half.

Am I missing something?

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: anything but 6SN7, actually..., posted on September 20, 2013 at 21:38:28
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

> Doesn't the 6SN7GT have lower distortion than either the 12AT7 or 5687?

Well, the 5687 has around halve the distorton of the 6SN7 in my measurements. The 12AT7 is a different story.

However, THD distortion is meaningless and has zero correlation with percieved good sound. So even if the 6SN7 had the lowest THD it would mean precisely nothing whatsoever.

Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

RE: anything but 6SN7, actually..., posted on September 20, 2013 at 21:59:16
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

> The 12at7 looks very non-linear to me. I can't see an operating
> point that would not cause the negative half of the voltage output
> to be much larger than the positive half.

Well, the Baby-O uses the 12AT7 in SRPP. Have you looked at the curves for that?

Note the Baby-O is not my design, but by Gordon Rankin.

http://www.wavelengthaudio.com/SP-BabyO.PDF

As I said - the one I heard sounded great.

Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

RE: anything but 6SN7, actually..., posted on September 21, 2013 at 07:20:46
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 "Have you looked at the curves for that?"

The best case scenario for any triode, in terms of linearity, would be a horizontal load line and I have looked at that.

"the one I heard sounded great."

I only said that the 12at7 was not linear. I didn't say anything about the sound of an amplifier.



Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: anything but 6SN7, actually..., posted on September 21, 2013 at 07:46:45
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 "However, THD distortion is meaningless and has zero correlation with percieved good sound. So even if the 6SN7 had the lowest THD it would mean precisely nothing whatsoever."

It almost sounds like you're saying HD is meaningless.

The upper ordered distortion is what we find objectionable and THD doesn't tell us how much is lower vs. higher so THD is meaningless, not HD.

In other words, what we want to know is which tube has the lowest upper order HD. That should correlate with perceived good sound.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: anything but 6SN7, actually..., posted on September 24, 2013 at 04:59:40
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

> It almost sounds like you're saying HD is meaningless.

No,not HD, correctly weighted, just THD.

> In other words, what we want to know is which tube has
> the lowest upper order HD. That should correlate with
> perceived good sound.

Maybe, you do need to consider the complete amplifier...

Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

RE: anything but 6SN7, actually..., posted on September 24, 2013 at 08:32:46
Tre'
Industry Professional

Posts: 17896
Location: So. Cal.
Joined: February 9, 2002
 "you do need to consider the complete amplifier..."

I assume you are referring to distortion cancelling? (correcting the distortion of the output tube using the distortion of the driver tube)

How do we achieve a perfect match along the whole amplitude operating range?

Is there a penalty if we don't?

Tre'



Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: anything but 6SN7, actually..., posted on September 24, 2013 at 09:53:40
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

> I assume you are referring to distortion cancelling?
> (correcting the distortion of the output tube using
> the distortion of the driver tube)

Not really.

I am only pointing out that if you have two tubes in series looking at the spectrum of one tube will not tell you much about the spectrum at the output of the Amplifier. And there is where it matters.

As you note, levels make a difference and so on, so there is a certain Alchemy to making especially low/no feedback circuits "right", just looking at tube curves does not help much in this task.

Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

RE: A heretical view, posted on August 5, 2024 at 20:16:42
cin5
Audiophile

Posts: 95
Location: Long Island
Joined: August 17, 2015
 Hello Andy,

Next year I want to see about choosing a 300B SET amp. While I can't find any reviews on it in the audiophile press, I'm wondering about this one.
https://www.decwareproducts.com/sarah Any thoughts or experience hearing this amp?

Yes, many like Mark Jeffs, who used to frequent diyaudio and audiotalk forums, raved about the 4p1L tube; too bad it's long been discontinued. But how well Mark managed to damp out its high microphonics I wouldn't know. Actually, I know very little about tube amp design. And the only tube hardware I have is the Don Sachs DS2 preamp I bought early this year. http://www.dsachsconsulting.com/custom%20line%20stage.html

Can't audition it until my 93db speakers are finished. I'll be using that preamp to drive my First Watt F4 amp. https://www.firstwatt.com/product/f4/ At 25 wpc into 8 ohms the F4 will easily drive my speakers. And my understanding is that the DC coupled zero feedback F4 was intended to be act as a "chameleon"; sounding very much like whatever's driving it.

But as an alternative to using the DS2 preamp to drive the F4 amp, are you familiar with any 2A3 preamps? I thought of the 2A3 after reading that its sound has a lot in common with that of the 300B. If yes, please suggest 3 or more excellent preamps for ~ $2.5 to 3K.

Would you say that the best sounding 2A3 preamps are the ones where the 2A3 tube is driven by a 6SN7 tube?

But do most 2A3 preamps have at least a 5V output? I would need that to drive the F4 amp.

Regards.

Greg

 

Page processed in 0.040 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]