Best cost no object tube phono

Hi Agoners,

just upgraded everything to SS Accuphase, loving it and have no intention to go back to tube pre/power. However, I have to admit that I miss a bit of tube sweetness particularly on mid. What is the best cost no object tube phono stage?

Thanks in advance for any advices

Showing 11 responses by atmasphere

I'd probably have in mind to use EMT OFD 15, 25, 65 & JSD with FM easily take advantage of the EQ curves, as I have many 78s.
@mcmprov Do you have cartridges with 78rpm styli as well? The larger stylus makes less noise in the groove!
@mijostyn IIRC the MAT02 used in that circuit is nla. But I think you can make a MAT12 work. Most of the frame-grid triode family (includes the 6DJ8/6922) tend to have microphonic problems. So you really have to screen them.
There are a few threads on this. As Raul points out it is very difficult to get very low noise in an all tube phono stage.
Difficult but not impossible. To get low noise with tubes you do some of the same things that are done in solid state preamp- particularly those that have opamps:
1) have a tightly regulated power supply2) use differential inputs3) use effective constant current sources for those differential circuits
In addition you simply parallel tube sections as part of your means of reducing noise.

The reason differential gain stages are used in solid state (and in some tube phono sections) is that for every stage of gain that is differential, theoretically you get 6dB less noise. This adds up- if you have two stages of differential gain, that's 12dB lower noise. But to that theoretical limit the constant current source really has to work.

If you do this then its no worries getting a phono stage to work fine with a cartridge of 0.2mV with gain to spare.
I would be interested to know with paralleled tube circuits, how can you ensure the tube sections parallelled are matched, and stay matched over a reasonable time. If they are not perfectly matched, then I would assume smearing of signal occurs.
It turns out that they don't have to be matched so much as screened for low noise. The paralleled sections behave as a single tube section so there isn't any 'smearing'.

By using a cascode you can easily get the gain needed for LOMC and getting bandwidth isn't an issue.
Btw, I named my personal unit Essential 3180 because it has at least two important upgrades from the 3160. All those customers are satisfied with.

And you know what: unfortunatelly you will never have the experience to listen/own the 3180. You can't even imagine those quality high levels. Such is life.
This is a textbook example of trolling. 'No-one can get what I got and its the best'... Sheesh.

I see people that run junk shops do exactly this sort of thing.

Raul, perhaps take your trolling elsewhere. In case you read it, this thread is about tube phono sections, not some solid state unknown as you seem to always push; the one that you have that no-one else can get; that is a cheap junk store technique. Show some class.

You have no engineering expertise by your own admission; for you to declare what knowledge that someone else has whom you've not even met is gross hubris and simply absurd on several levels. You keep talking about your 3160 thingy but its obvious you didn't have anything to do with its design.

Put another way, if you can't get out there and show the rest of us how its done, then sit down and shut up.
It's not me who needs to show it but you that are an audio manufacturer and I know you can't do it because you have not the adequated knowledge levels with SS designs.
So I've been making all tube preamps for over 30 years, with two patents in that field, just got another patent a year ago in the field of class D amplifiers, which generally speaking are solid state and employ opamps :)  ... We're right now building a run of our beta production class D amps, which are not based on anyone else's design or modules. Just so you know, there's quite a bit of math that goes into making something like that work properly (google 'oscillation criteria'), far more than building a solid state phono section and you're trying to say that you know what knowledge I possess and what I don't. Sheesh.

Let us continue the discussion.
This thread is about tube phono sections, right? Not a defunct 3160 solid state thingy that no-one can get or has ever seen, right? I do high end audio because I like it, and not for any other reason- that's why I'm here. I'm fine with people not agreeing with me- you and I have not seen eye to eye in the past apparently; that sort of thing is normal on the web. Unlike Raul you don't seem to derail threads on a regular basis or make the thread all about you. But Raul makes false statements, he makes personal attacks and if you don't call him out, other people might think that what he says is real.

Do you think his misinformation should go uncorrected? Because we can do that, but this thread won't be about tube phono sections anymore...
Btw, atmasphere I don’t care about class D amps. I was very specific about: design of SS phonolinepreamps.
Of course you don't - that would not help your argument. The bigger point here is the need to be right for the sake of it, rather than being factual. While you may 'not care about class D amps' the simple fact that we can make one that is of our own design, and that we can get a patent in the field, is enough for anyone to know that if we wanted to make a solid state phono section, we could and with ease.

*****************Back on track:

One advantage tubes have over solid state in phono preamps in particular is that of overload due to the electrical resonance caused by the cartridge and tonearm cable (inductance and capacitance in parallel; please ask me if you don't know what I'm talking about here). In solid state circuits, if this resonance goes into excitation (oscillation) the resulting RFI can overload the input circuit. This can result in a tick or a pop, and is part of the reason that many people think that ticks and pops are endemic to LPs. 

But when a producer makes a recording on LP, he gets what is called a 'test pressing' which he listens to to see if its alright. A form accompanies the test pressing, and the producer has to sign off on it, approving the test, saying that the stamper is good and the LP production can proceed. One of the things producers listen for is ticks and pops in the LP. So when they sign off, its on something that shouldn't have any ticks or pops.

If your phono section is immune to the RFI generated by a LOMC cartridge and the tonearm cable, its common to hear entire LP sides with no ticks or pops at all (unless the LP clearly has a scratch). (For this reason, I frequently get asked at shows if I'm playing a CD- I just point to the turntable.)

This immunity comes from four things- preventing the RFI from getting in, making sure that each stage can't oscillate (with tubes or discrete devices, a 'stopping resistor' is used for this purpose), the circuit has enough high frequency overload margin that the RFI can't overload it (keeping in mind that the RFI is a peak of 20-30dB) and finally that the circuit does not rectify the RF energy.

With tube circuits, if a tube is at the input directly (as opposed to an SUT) the RFI won't be likely to overload it. It might send it into oscillation if a stopping resistor isn't present, but if its there, then overload won't occur.

(A fundamental difference between solid state and tube phono sections is where the overload occurs if a strong signal is applied to the input of the phono. With solid state, the overload can occur right at the input. With tubes, its likely one or two stages downstream.)

Normally a 'cartridge loading resistor' is used to detune the RF resonance of the LOMC cartridge and tonearm cable. Eliminating RFI from audio circuits can affect how they sound; this is part of why you hear differences with loading vs no loading (although it can affect tracking too). But if you have a tube preamp, its far more common that loading simply won't be needed (SUTs excepted- that's a different set of conditions). Its much harder to get a solid state phono section to behave in this regard since semiconductors have diode junctions in them, and these junctions can rectify RF energy. For this reason, solid state phono sections that don't need cartridge loading are less common than they are with tube phono sections. If the RFI is rectified, this sits in the noise floor- and because of its structure, prevents the ear from penetrating the noise floor to extract more detail (natural hiss is one of the very few exceptions to the ear's masking rule).

If your preamp has loading options built-in for LOMC cartridges, its a good bet that the designer didn't deal with the RF resonance. If he had, the performance would be better.

So while tubes might be a bit noisier than solid state overall, you can more easily have less ticks and pops which to me is preferable. And as long as the phono noise is less than that of the cartridge tracking a silent groove on the LP, you won't hear it. That's very doable with cartridges of 0.2mV, which is about as low as cartridge outputs go with very few exceptions.  
In the other side the " tale " about those " tick and pop " is just that a tale that can't be proved it happens.
Sure it can. I have to admit though the first time I encountered this it had me floored. That was a bit over 30 years ago. Since then I've seen it plenty of times.

There's plenty of sound engineering reason why this happens. The electrical resonance between the LOMC cartridge and the tonearm cable can be a peak of **30dB**. If the cartridge is making 0.5mV, this means that the peak could be half a volt! Many phono sections that can handle a LOMC cartridge will simply be overloaded by that much input. In solid state preamps in particular, the RFI can be rectified at the base of the very first stage of gain, resulting in a burst of noise for a very short period of time- a tick or a pop results from the circuit being overloaded.

This isn't hard to prove; we can inject a 0.5V RF signal at 1 or 2MHz into a preamp and see what happens. Often it isn't pretty. The idea that somehow this 'can't be proved it happens' is ridiculous on its face.
I've never heard the atmasphere phono product (I heard a set of their mono amps decades ago and liked them) but note it also uses balanced inputs. I reckon it is a fully balanced product like the Accuphase.
@ivanj It is, and has the distinction of being the first balanced line preamp made for home use back in 1989. People don't think about it as so weird now, but back then people looked at the MP-1 crosseyed because it used XLR connections for the phono input.

Btw, I provoke the wbt " banned " on purpose and I let to know it to gentlemans that through the wbt inside private emails gave me their support to follow posting and told that I don’t wanted anymore to stay where the owner is " hunting me " and few days latter I posted something to and that’s how things happened.
What actually happened is Raul violated the WTB posting rules (by making personal attacks on other posters), that after having been banned 7 years prior. His re-reinstatement lasted about 2 weeks.
@rauliruegas  Al (almarg) and Roger (ramtubes) have both passed on, RIP.