The "Muscovite" 6S3P Tube Phonostage

Administrator
Joined 2004
Paid Member
Second Stage Battery Bias

Referring back to the schematic where I showed how to implement battery bias on the lower triode of the choke loaded mu follower I have implemented this in my pair of pre-amps.

First I'll state that measurable differences are pretty close to nil with the exception of a very slight change in operating point.

A couple of points I will mention are is that the RIAA network is now polarized to the tune of ~ -770mVdc which is the negative grid bias.

The dynamic impedance/non-linearity of the 1N4148 has been removed from the signal path in the second stage. The gain is a few tenths of a dB higher.

I had a little trouble finding a spot for the battery holder. The required changes are to replace R11 a 562K with a 1M resistor, and add R12 also a 1M resistor.. Note this is slightly different than in the previous post referencing this mod. See attachment below.

What I can say is that it does seem to make a subtle, but positive difference. Resolution seems very slightly enhanced, the bottom end has gained a little additional definition, and the highs are subtly smoother. Very incremental changes, and perhaps not worth the bother of an additional set of batteries, but I think they will be staying.

Current drain is 0.75uA so essentially battery life should be close to shelf life as in the case of the 12V bias battery on the upper section.

Potentially a worthwhile improvement for the obsessive.. :D
 

Attachments

  • batt_bias_as_built.png
    batt_bias_as_built.png
    36.8 KB · Views: 1,019
Administrator
Joined 2004
Paid Member
Hi Ejam,
I prefer it before because it does not form an implicit voltage divider with the series resistance in the EQ network and that allows me to maximize the available gain, also one less component tolerance to be concerned about wrt to equalizer accuracy. It is also just a wee bit easier to calculate the EQ network - not that I didn't long ago write a little basic program to do most of the work for me.. :D

The one definite down side of this approach is that it does require a larger coupling capacitor which drives up cost significantly.

Sonically speaking I doubt it makes one iota of difference, except that the SNR should be marginally better when that resistor is placed before the network due to its minor noise contribution but more importantly due to the attenuation introduced if placed after the network.

Note that with certain capacitor types it is alleged that polarizing them with plate voltage makes for subjective improvements, I'm not sure. I do tend to use fixed bias in the second stage of my designs, but have not definitely been able to answer this question with any design approach. I do tend to use good quality polystyrene capacitors in the equalizer network however.

We're talking very incremental differences and valid arguments can be made for either approach.

I've been doing it this way for about 24yrs as a result of having an early epiphany and realizing that I was throwing away a couple of dB of gain in those early 12AX7A based designs. In fact with high transconductance types using low resistance values and large caps in the EQ a 1M grid resistor shunting the network is not much of a problem, very different from my early 12AX7 based designs where that series resistor was around 274K - 301K. (Can you spell noise?! :D)

Which Loesch are you referring to? I was privileged to know Arthur in his altogether too short life, and am familiar with Thorsten Loesch's LCR based design as well.
 
Hello Kevin,
Funny you should mention Loesch's LCR design. I'm in need of either separate phono and linestage preamps or an all in one unit. I've been looking at various diy options such as Loesch's LCR as well as your Muscovite design. There's no way I can make a choice other than to trust other people's subjective evaluations such as your description of your Muscovite preamp. So, please indulge me and describe your subjective merits of your preamp with LCR types and others.
Regards,
David
 
Administrator
Joined 2004
Paid Member
Hello Kevin,
Funny you should mention Loesch's LCR design. I'm in need of either separate phono and linestage preamps or an all in one unit. I've been looking at various diy options such as Loesch's LCR as well as your Muscovite design. There's no way I can make a choice other than to trust other people's subjective evaluations such as your description of your Muscovite preamp. So, please indulge me and describe your subjective merits of your preamp with LCR types and others.
Regards,
David

Hi David,
It's been so long since I had the Loesch LCR phono stage on hand that I couldn't really compare them subjectively and be fair about it. What I can tell you is it adheres closely to the RIAA standard, is quiet, is quite transparent, and has no particular quirks that call attention to it. I designed it to be as neutral and honest as I currently know how, and also to be relatively modest in terms of cost to construct. Compared to my older design (which almost no one here has heard) it is quite a bit quicker, more detailed, and presents a deeper/wider image.

Probably not the sort of information you are looking for, but I'm not too comfortable discussing the subjective merits of my designs as compared to those of others.
 
Hi David,
Compared to my older design (which almost no one here has heard) it is quite a bit quicker, more detailed, and presents a deeper/wider image.

Probably not the sort of information you are looking for, but I'm not too comfortable discussing the subjective merits of my designs as compared to those of others.

Hello,
Thanks for your thoughts. When you say "Compared to my ...it is quite a bit quicker...", the "it" refers to your current design the Muscovite, correct? If so, then I'm in. I also like the fact that it will be cheaper to build. Those LCR units are very expensive.

Actually your subjective ruminations are exactly what I am after since that is all I can ever get over the internet. As long as these thoughts ae born out of experience and are honest, I have no problem in using them to make my diy decisions.
Regards,
David
 
Administrator
Joined 2004
Paid Member
Hi David,
Yes I was referring to the Muscovite which I think is a significant improvement over my last design. I'm still learning, but I am very pleased with the both the technical and subjective performance of the Muscovite, and because I was doing this entirely for myself there was no design orthodoxy involved - I literally tried every concept I've wanted to evaluate in this design - it was all for my own amusement.. :D I've never named a phono stage design before so I guess that says something.. :p
 
Muscovite Sound

Having listened to several of Kevin's previous phono stages and now having two listening sessions with the Muscovite, I would suggest that Kevin is being a bit modest here.

I cannot in fairness compare it in detail to other phono stages, but this new design sounds very good and represents a substantial advancement over his older designs, espcially the D3A design, which design I am fairly familiar with. It is very dynamic and clear and extremely quiet while having a high gain. The design is neither terribly difficult or costly to build, so I think it would be a good build to consider for anyone interested in an extremely competent phono stage.
 
Administrator
Joined 2004
Paid Member
Evaluated the 6S3P-DR in my pre-amplifier and concluded that I like the 6S3P-EV better, found that the consistency of the EV was better on the mu/transconductance front as compared to the smaller batch of DR on hand.

The DR was somewhat less resolving and more laid back sounding to the surprise of my friend (GTHICM) and me as we listened to both types.. Not what we expected.

I have two close friends who are planning to build their own having heard mine, and a couple of others here who have expressed an interest in doing so. Any questions please feel free to ask here.

I'm really happy with the result.. :D
 
Administrator
Joined 2004
Paid Member
I'm expecting the custom Edcor power transformer and standard Edcor 5H choke any day now.

Hoping I will have enough time to build the final supply for this pre-amplifier before I schlepp it to the local summer audio-fest to show it off at the end of this month..

The final design will use 6BD5GT as the pass elements but a 6V6 or 6L6GA/5881 would be suitable as well.

The pre-amp has proven an instrument of rediscovery in that I am rediscovering my vinyl collection and hearing things not heard before - the design has sufficient resolution as well that the differences between the SPU Classic GM E II and SPU Meister Silver GM II are readily apparent..

I'm happy with the overall result, and will report the outcome of the new supply build when done.

The battery bias change in the second stage incidentally is a keeper. (This is the change that requires a single 1.5V AA in addition to the 12V A23 already used in the second stage.)
 
Administrator
Joined 2004
Paid Member
Final PSU

The power transformers arrived on Thursday night and Saturday afternoon I started to punch the chassis and mount all of the parts - I managed a little wiring before knocking off to go out to dinner. Today after I got back from ice skating I started wiring in earnest, and was able to complete and debug the power supply over the course of about 8 hrs today/tonight.

The 6BD5GT that I choose to use instead of 6V6/6L6 is a little more fussy than I am used to, and my initial reward for all of this work was relaxation oscillations and some VHF garbage. This was resolved in a few hours along with a bad buzz originating in the reference for the right channel supply. (Don't reference filaments directly to a zener used as a reference - apparently this can result in sufficient leakage current to make the reference noisy)

The design has been running for about 1 hour without problems so I would not consider this fully vetted. Build at own risk and expect some issues to be resolved if you follow this route.

Note that I built two supplies, one for each channel sharing a common raw, but rather well filtered supply. The output voltage is approximately 302V, and output noise and ripple is approximately 700uVrms - the pre-amp is now extremely quiet.

Performance overall seems slightly improved if anything, as I mentioned it is quieter overall, and the tonal balance seems if anything even more neutral than before.

The filament supply is not yet documented, but is conventional LT1084 implementation. (It is fully floating.)

EDIT: Power supply misbehavior root cause was pinned down tonight, I also determined that the filament supply is OK, and spent several frustrating hours before I discovered one of the two 6BD5GT pass tubes was gassy. (This caused popping and all sorts of other bad behavior and was visible on the raw plate supply. The strange violet glow at the base of the tube was the clue as was the occasional muted pop when the B+ came up. Replacing it solved all odd behavior.. I did change R6 to 4.7K which further quieted the screen supply, and is recommended for this reason. Schematic updated 7.25.12
EDIT2: Changed R6 to 6.8K and updated schematic to reflect. Shifted operating points of error amplifiers and dropped screen voltage slightly. 7.29.12 (see revised regulator design later in thread for current implementation 5.12.13)
 

Attachments

  • DSCN0096_R.jpg
    DSCN0096_R.jpg
    206.9 KB · Views: 797
  • 6BD5GT_PSU.png
    6BD5GT_PSU.png
    72.3 KB · Views: 724
Administrator
Joined 2004
Paid Member
Here is a picture of the inside of the power supply chassis just before I drilled the holes and mounted the bottom cover plate.

Note that the much simpler (single) supply detailed in an earlier post is a perfectly acceptable alternative to this design as would be one similar to this using the 6V6 instead of the oddball 6BD5 (different pin out)

I will post any further changes I make to the PSU as I "fine tune" the design. :D
 

Attachments

  • DSCN0098_cr.jpg
    DSCN0098_cr.jpg
    209.8 KB · Views: 758
Nice to hear about your progress, Kevin. And thanks for the chance to listen to the Muscovite. I have to say the Muscovite is a beautiful sounding phono pre (even without the new PS), very accurate sounding (i.e. neutral) and simply doesn't get in the way. Which is saying a lot based on the quality and accuracy of the rest of your system! That's not to say it isn't involving, though. It is! Very resolving but no fatigue for me. It's hard to describe as I haven't heard a phono pre like it before, but I like it! :cheers:

So my question was going to be, if I have 6V6 tubes lying around (I do) is there anything to recommend AGAINST using them - I know you mention some improvements using the 6BD5, but if my system isn't as resolving as yours, will those improvements be likely to be noticeable? Tough question I know. But since I have the 6V6s, I'm likely to use them. Anyway it sounds like you've answered that, so maybe no followup is needed. And from what you've written, the 6V6 shouldn't give me the stability trouble you've had with the 6BD5?

Finally, I'm interested in your filament supply should you choose to post it.

Cheers!
 
Administrator
Joined 2004
Paid Member
Hi Carl,
Regarding the filament supply I recommend the design I originally posted - the only change I made of consequence was to use a 22000uF cap instead of the 10000uF and I do not believe this actually significantly improved the performance of the supply.

With regard to using the 6V6 I think this is an excellent choice and in fact was what I was using in the PSU that powered it when you were here. I think all of the benefits of the current design will be realized with the 6V6 in place of the 6BD5, and to be honest I need to spend a little more time with this tube to really trust it. The way I would put it is I purchased 5 NOS 6BD5GT and the matched pair I bought had one defective tube, so the failure rate out of the box is high IMLE.. We'll see how the rest turn out.
 
Administrator
Joined 2004
Paid Member
Spending some serious listening time tonight, I'm relatively satisfied with the performance at this point, will probably tweak the power supply a bit further once the dust has settled. (We're talking weeks) It's stable and the pre-amp is now vanishingly quiet and sounds good - so the supply has accomplished the design goals.
 
Administrator
Joined 2004
Paid Member
Some minor changes to the power supply since the last photograph. Removed and moved a pair of Clarity SA 4.7uF caps to the phono stages, replaced the 0.1uF feedback caps with Clarity SA and beefed up the grounding to the references and feedback reference points..

No problems so far, one minor thump due to a nearby lightening strike.. (Yeah I just generally ignore the weather outside, figure worst case I'll need some new power transformers.. :p ) Most of the analog gear runs on balanced power from a 600Va medical grade isolation transformer. The GM70 power amps don't like my 1kVa medical isolation transformer very much so they run direct.
 

Attachments

  • W_ClarityCapFB.jpg
    W_ClarityCapFB.jpg
    220.6 KB · Views: 683
Administrator
Joined 2004
Paid Member
It's pretty late, I guess after nearly six hours of continuous operation I'm out of the woods. :D

The improvements with the stereo power supply are more significant than I expected. There is certainly more depth, and additional detail - neither were lacking before, and the sound is not the least bit analytical.

I'm rediscovering some familiar recordings, somehow this pre-amp allows me to see a bit further into recordings than any of my previous efforts.
 
somehow this pre-amp allows me to see a bit further into recordings than any of my previous efforts.

And THAT is what I look for in all my equipment...that is, without it getting too dry or analytical. It's a fine line and I don't quite understand what qualities in the gear allow such clarity without the downsides, but maybe your goal of accuracy is the answer?

I hope I'll have another chance to listen. But off to the wilds of Canada for a while now. :wave2: