HiFi Haven Group Project - Building The "World's Best Tube Phono Preamp"

Thanks guys! There is no magic circuit though. I once read something like this---the final sound quality is 50% design topology, 50% parts selection, and 50% layout and build quality. Meaning that these are all critical factors.

100% right! :)

So let me ask this question. If you could do it again, what would you do differently? Also, were there any choices you made due to cost or expediency?
 
The D3a/5687 circuit details evolved over 10 years and I tried many variations before the version shown. Here is just a quick off the top of my head list of some of them:

* D3a—-pentode vs triode
* 5687 vs 56 vs 27
* Biasing for each tube—battery fixed bias vs cathode resistor vs LED
* Plate loading for 5687—resistor vs plate choke vs CCS
* RIAA eq—-High impedance network vs Low impedance
* Different rectifier tubes
* B+ with tube regulator vs old school iron and big caps
* Heater supply—direct ground reference vs DC reference
* Separate heater transformer vs windings on B+ transformer
* Different B+ transformer with higher voltage

And that is before the part swapping. I have tried different types of resistors and capacitors for nearly every part on the schematic, in some cases half a dozen or more different types, before settling on the ones shown.

So to answer your question, No, there really isn’t anything I would change in the design topology or part selection. And part choices were made based solely on what sounded best to my ears without cost considerations. Unfortunately as I have said before some of these parts are no longer available. The BlackGate power supply caps are no more, and I have no idea what to recommend as a similar sounding alternate. The 2uf 600v V-Cap on the output is no longer made, but here I would feel comfortable using a single .47 or 1.0 Copper V-Cap instead.
 
Looking at the schematic again reminded me I also tried choke input filtering (with a different power transformer), different chokes in the outboard supply, different RCA jacks and different IEC inlets (4 of them).

Two things I didn’t try are a mercury rectifier and LCR eq. I was really tempted to try a mercury rectifier but I took it off the list after I had a chance to use a Jeffrey Jackson phono preamp for a couple weeks. It had a somewhat similar topology and tube choice but used an 83 mercury rectifier and an output transformer instead of a cap coupled output. Without checking my notes, I don’t recall whether the eq was LCR or RC. In any event I thought my D3a/5687 phono sounded better overall so I basically took the mercury rectifier off my list of things to try. Given the reputation of mercury rectifiers for superior sound quality, that may have been a mistake. I did try a number of non-mercury rectifiers including most recently a Sophia 274B Aqua, but the RCA 5R4GY is still my favorite.

Anyway I think the basic circuit is a proven one and is fairly easy to build, and it would be a good platform for other DIYers to use for making their own part selections. My only recommendations that differ from the schematic are to use a 300 ohm Riken gridstopper resistor on the D3a and a 47K input load resistor instead of the 100K in mine. The 100K is a better load for my particular setup.
 
One other important omission. I forgot to credit Stephen Robinson in the UK with assistance in the initial design of the D3a/5687 preamp. Stephen has a website called Izzy Wizzy Audio where he describes several of his DIY projects. One of these tracks the evolution of his phono preamp through roughly 20 versions. My first version back in 2007 used an audio circuit similar to his Mk 15 but without an output transformer, and it had a much simpler power supply. From there we went in somewhat different directions but Stephen provided very helpful assistance on several occasions in modeling changes to the RIAA network for example. Whatever your design priorities may be, his website makes for interesting reading. Here is a link to it:

http://www.izzy-wizzy.com/audio/preampnew.html
 
This thread is brilliant. I have always fantasized about LCR eq in a phono setting, but this thread has me thinking more about layout, parts and design execution. A big thank you to Salectric for sharing.

A quick question tho for Salectric .... with your proposed D3a/5687 preamp circuit, the resulting gain would still require the use of a step up for low output MC's? I must have missed in the previous postings where gain of this circuit was mentioned.
Cheers
Frank M
 
Frank, yes, a SUT is definitely required for low output MC cartridges. I use a 1:10 transformer (20dB) which with the roughly 40db gain of the phono preamp gives me 60db of total gain. Even without an active linestage, that gives me plenty of gain for my 0.35mv Benz.

Coincidentally I was just listening to a record at very loud volume (at least for me) and putting my ear right next to my speakers I could just barely hear a low level hum and a very faint hiss. But if I tried running a LOMC directly into the D3a and turned up the volume to compensate for the missing SUT the noise level would be a lot higher.
 
I certainly have my biases, and I haven't built Sal's preamp, but I would be willing to wager that pentode wiring the D3A and directly coupling it to the 5687 would be a step toward making what I have thought for several years now would be the best possible tube phono preamp. These are pretty difficult challenges though!
 
I certainly have my biases, and I haven't built Sal's preamp, but I would be willing to wager that pentode wiring the D3A and directly coupling it to the 5687 would be a step toward making what I have thought for several years now would be the best possible tube phono preamp. These are pretty difficult challenges though!

What are those challenges?
 
Pentode wiring the D3A means giving the screen grid its own stable DC supply. This nearly eliminates the input capacitance of the first stage (always a good thing) and increases the gain significantly.

Direct coupling means removing the coupling cap between the two tubes. The signal levels are small and the 5687 doesn't need a ton of plate voltage to operate, so this isn't an unreasonable approach.
 
Speaking of pentodes...

So -- I'll readily admit that I know squat about the design and construction of a good (vacuum tube) phono preamp -- but I do know that Kevin Kennedy, in whose ears and skills I have considerable trust, has been working for years around a couple of designs that (at least locally ) have attracted a fair amount of interest.

http://www.diyaudio.com/forums/analogue-source/213769-muscovite-6s3p-tube-phonostage.html
http://www.diyaudio.com/forums/analogue-source/246756-muscovite-mini-6-9-6z9p-phono-stage.html

Other than to mention it/them, I am not sure I have a question.
If I did have a question, I guess it'd be whaddya think?

I think I may have (???) heard a fairly early iteration of the original design at a hifi shindig in eastern MA a few years back -- cannot swear to it, though.
 
The second preamp isn't too far from what I have in mind, but the 6J9P has about half the transconductance of the D3A (but they are otherwise quite similar).
 
Back
Top