The "Muscovite" 6S3P Tube Phonostage

Administrator
Joined 2004
Paid Member
For some reason I never saw a notification that you had posted to my thread, but here are a couple of thoughts.

The source impedance of the input cascode is extremely high, not making it IMO a suitable choice for driving a transformer, and if 90K is reflected to the plate as you state this would result excessive gain and probable clipping with modest input signals. The biasing of the first stage would also be incorrect. Provided all of the above were effectively addressed you would have an amplitude insertion loss of >20dB excluding the RIAA network itself. A transformer with a 5:1 ratio might work better, but again you are on uncharted territory and because of the high source Z getting sufficient primary inductance for good LF performance would be difficult.

A number of people have built my design as laid out here, and I have had mine for several years, I recommend building as designed or starting from scratch..

Kevin K
 
Administrator
Joined 2004
Paid Member
Looks pretty noisy. That cathode bias diode looks like it'll generate quite a bit of distortion and noise without any DC degeneration. Best go with the standard resistor and capacitor in parallel.

Well you would be dead wrong. Internal noise resistance of that diode is a score of ohms. Electrolytic capacitors operating with extremely low bias voltages have demonstrated linearity problems based on experience.

The diode has approximately 10mA of current flowing through it and provides a linear voltage drop of just under 700mV, and an equivalent series resistance of perhaps 10 ohms. An equivalent resistor of 68 ohms would require at least 1000uF of capacitance for equivalent performance and as I have noted few electrolytics I have tried worked well at these extreme low voltages. Linearity and noise have not been a problem - I've used it in other phono stages without issue, and the idea is not original to me in any sense. Remember that the signal level here is just a few mV, the diodes I am using (1N4148 Fairchilds) do not generate sufficient noise that I can even measure it.
 
Administrator
Joined 2004
Paid Member
I think the complexity of the design and it's power supply scares people off to a large extent. (Perhaps the fact that I and my designs are known to few here also plays a role.)

What I can say is that since I completed the prototypes they have been in constant use and have been completely trouble free. No issue with the obscure tubes used in the design either.

Each successive cartridge upgrade has revealed that system electronics and speaker system are not the resolution limitation in analog playback. My current first tier cartridge is an (SPU) Royal N on an AT Technihard shell, the increase in resolution over the SPU Meister Silver it abruptly replaced was a bit surprising.

I have two close friends who are at various stages of completing their own having heard mine.
 
I recently visited Kevin and heard this phono stage in his system. It is a remarkable unit. Very nice detail, relaxed sound, musical and natural. I'm considering building it myself. I would without hesitation if I didn't have something decent and I weren't getting past the age where I want to take on another project. If it had come along earlier, I'd have built it. if I did want to take on a project, this would be it. I have built other equipment Kevin has designed, and I've loved all of it. If you need a phono stage, give this a shot.
 
Administrator
Joined 2004
Paid Member
Yesterday the plate transformer developed a short in the high voltage secondary, died very quietly with the primary fuse blowing. Cool to the touch.. Replacement on order.

The timing was not propitious with my current health issues, although had this happened a few months ago I'd have been completely up the creek, at least I was able to lash together a temporary and largely acceptable PSU... :D

I'm now using the development supply which I've tweaked a bit to improve its performance, but I notice the loss of performance as compared to the intended dual mono supply. (Some loss of depth, speed, detail, the bottom end is not as tight, and there are a few other things) I therefore STRONGLY recommend the stereo supply as designed for the one or two of you who might actually build one of these.. :p
 
Administrator
Joined 2004
Paid Member
They've gone up very little from what I can tell, the cheapest I paid was $1.50 each in bulk for the 6S3P-EV and several sellers are currently offering them within pennies of the price I paid.

I do not recommend the DR for this design, the samples I tried did not sound as good as the EV which also had much more consistent transconductance sample to sample. (Gain mismatch channel to channel which was not a problem with the EV.) As it happens sellers seem to charge significantly more for the DR.

I recall prices peaking sometime last year and they have come down quite a bit since then.
 
Administrator
Joined 2004
Paid Member
I sure hope not, performance will be poor. Please don't interpret the design in this way as I can guarantee it will NOT perform as intended.

I still get regular emails from people who "interpret" my designs and then cannot get them to work correctly or are otherwise disappointed. I have limited time and serious health issues - if you are not going to build the design largely as intended I recommend building something else.

The Salas SSHV2 is an acceptable alternative to the tube based regulator, but I have not evaluated it for use with this design. It will probably change the sonic recipe to some extent from what I intended, but it is quiet enough and has a low enough source impedance for the pre-amp circuit to work properly.