Real World All Tube LCR Phono for MC Without SUT

the road not taken.
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Real World All Tube LCR Phono for MC Without SUT

Post by Naz »

I promised to post an update on my LCR Phono design after Jeff Davison got around to building a version. He finished his a long time ago, and what a work of art it is! Sorry for the delay.

It is relatively simple, using building blocks that I've found over years of comparisons to sound the best (to my ears and those of my audio friends at least). Of course, there are many possible variations, some possibly without too much sacrifice. An obvious one is to substitute one or more of the top triodes with a simple plate resistor but the additional PSRR and linearity is worth going the extra mile in my book. Another is to replace them with a simpler SS CCSs but I wanted to keep this all tube.

I chose to go with discrete 7K LCRs and Dave wound some fantastic coils and grid chokes for the job with multi-taps (what a great idea). As it happens, Spice predicted the outcome perfectly in this case and the taps weren�t really needed but still highly desirable I think as you could use them for helping to balance badly Eq'd recordings.

The big lesson for me was that the coils pick up way too much hum in my original design and I had to move the LCRs to follow the second stages, providing almost 30db s/n improvement.

So it measures pretty damn good but how does it sound? Well, with decent components, this is the best Phono I've heard so far. So natural, detailed, dynamic and unfatiguing! And, it's quiet enough to use without SUTs on Carts with only 0.25mV output!

A couple of notes for anyone actually wanting to build one. One, a good regulated PSU will go a long way. Two, you WILL hear a huge difference in different tubes. My personal favourites by a significant margin are Seimens Cca but they are VERY expensive. A mix of tubes might be the best option and some relatively cheap 6H30P DRs might be the ticket in the final stage (just add an extra Sic in the cathode).

I�ve posted both the schematic and ASC and comments or queries are welcome.

Naz
Attachments
naz.jpg
naz.jpg (134.84 KiB) Viewed 65882 times
Phono_LCR_3 Stage_Ref_8-1-10.asc
(11.88 KiB) Downloaded 608 times
Andrew
Posts: 201
Joined: Wed Apr 26, 2006 8:06 am

Post by Andrew »

Nice, Naz, I really want to try this but I'm focussing on getting my existing 50db MM stage quiet enough first I use an E810F into 1k5 into E180F (in triode). I have a box in which to pot my chokes, just need to find time.

-- Andrew
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Hey Andrew, what sort of noise are you getting? If it's tube rush there's not a lot you can do other than use a different tube.

Naz
Andrew
Posts: 201
Joined: Wed Apr 26, 2006 8:06 am

Post by Andrew »

The chokes are picking up hum, they need shielding, I'm sure of it.

-- Andrew
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Thought so, that's why I asked. You'll never fix it, ask me how I know! My original design had the LCR after the first stage. It hummed like a bastard even though the components were in mu metal cans. The only way I could overcome it was to place the LCR after the second stage, picking up near 30db of s/n ... worked a treat!

Naz
bravi
Posts: 7
Joined: Fri May 07, 2010 11:09 pm

Using SS CCS in place of tubes

Post by bravi »

Any experience in replacing the tube CCS with solid state?I have some boards for the Gary Pimm CCS which I have just put together and wish to build this circuit for my next phono stage.

I have used CCS based on cascoded DN2540 in the power amp stages with good results. Hence my interest in using SS devices for the phono stage.
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Funny you should ask, I've not actually tried it in this design but always expected it to work fine with little difference (and it certainly Sims well). However, I recently tried it in a generic Phono design that half the world seems to use and got mixed results that I still haven't resolved to my satisfaction.

In this instance I was comparing SS CCS to a simple Plate R and I also used the DN2540. Overall it was certainly detailed but the sound was also thin and lacked warmth. Bass was tight though. For the heck of it I tried adding a series resistor and although it's not needed or indeed a conventional thing to do, it helped.

This has me curious and I'm waiting for some more CCS PCBs to arrive so that I can directly compare tube Vs sand CCS. I had previously argued that anything in the signal path can be heard and that I expected tube to sound very marginally better but it was all hypothetical. Just when I started thinking there would be no perceivable difference I now have some doubts. I will post my findings in the next couple of weeks.

Naz
bravi
Posts: 7
Joined: Fri May 07, 2010 11:09 pm

SS ccs

Post by bravi »

Thanks Naz. I used 2k Carbon comp resistors right at the gate to avoid possible oscillation. Some posts have indicated values of 330 ohms should be OK. However, my experience is limited to the power amp stages. To reduce dissipation I also used a resistor in series. This was in the tail of the phase splitter.
Will await your experiences in this regard
bravi
Posts: 7
Joined: Fri May 07, 2010 11:09 pm

Building the Phono

Post by bravi »

Just got a message from USPS advising that Dave has posted a set of inductors as per Naz's design. Time for some serious planning and parts collection. I wish to use parts in my bin as much as possible before purchasing values I do not have.

1. Caps in the RIAA circuit. Can I use two of .22 mfd totalling to .44mfd instead of .39mfd? The R in series will be adjusted so that the RC constant remains the same. Any possible issues with this approach? ( The .22mfd are the FT3 Russian Teflons that I purchased some time ago)


2. 300H choke. Can I replace this with a proper grid choke that I have in the box? They are BCP16 from Magnequest. Measurement with my LCR meter indicate inductance at 1800+ Henry. If it is possible to use this will the cap values or the terminating grid resistor get altered?

Comments from the experts on this forum would be very helpful.

Cheers!
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Hi Bravi, First I'll bring you up to date with the SS CCS. Still haven't A/B'd SS vs tube but I have managed to get good results on another Phono using a generic feedback design. Had to increase idle current through the tubes to get some more body, used standard CCS on 1st stage and an R in series on 2nd stage (half normal plate value) which lifted mid bass a tad.

2K carbon comp is OK for grid stoppers.

0.39uF in RIAA is critical. Going to 0.44uF causes drop of 0.5db centred around 150Hz or so. Could get away with 0.18uF in parallel with 0.22uF.

The grid choke value is also important but far less critical. You could make 200H - 500H work but 1800H is just too much.

Something I haven't mentioned (I think) is that I prefer a little bass boost at VLF when using full range speakers, obviously no required when using a sub. The grid choke can be tuned to work wonders here!

Cheers,
Naz
bravi
Posts: 7
Joined: Fri May 07, 2010 11:09 pm

Post by bravi »

Hi Naz. Thanks for the suggestions . Will keep them in mind. I can arrive at the recommended value of caps by using a parallel combination of .22mfd, 0.1mfd and 0.07mfd. The last being made up of two 35nf silver mica caps which have a 0.5% tolerance!.I suspect the ultimate value will be limited by the accuracy of my LCR meter. Looking forward to start this project over the Christmas break.
Cheers!
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

No probs Bravi,

I finally got around to comparing sand CCS with tube. I never cease to be amazed how relatively minor differences from a technical perspective can change the sound so much. Subjectively a pair of Cascoded DN2540 Mosfets sounded a little more detailed than a 6922 on top of a 6922 but were also thinner sounding, as if the the response was skewed towards the highs. Interestingly, the tube CCS measured the same throughout the audio band and had a slightly better HF extension beyond 100KHz. The same was true and even slightly more accentuated using an audio 2SK371 JFet cascoded with the Mosfet on top. FWIW!

Cheers,
Naz
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

I've been asked several times now about simplifying the design by either substituting the top triodes with sand CCS or a simple Plate R. Here is a simple CCS version that measues well and I know works.

Overall it sounds a tad thinner than triodes but subjectively quite detailed. Inreasing the idle current through sections 1 or 3 only, (changing 2 will upset the impedance match) can replace most of the lost richness.

Different Fets will change the overall sound and I've only tried a few. The bottom Mosfet can be substituted with a low voltage JFet such as 2SK369 but the top Fet must be capable of handling high voltage. Total parts cost is only a couple of dollars.

Naz
Attachments
Phono_LCR_3 Stage CCS_28-1-11.asc
(11.19 KiB) Downloaded 559 times
Naz_CCS LCR Phono.JPG
Naz_CCS LCR Phono.JPG (190.83 KiB) Viewed 65009 times
bravi
Posts: 7
Joined: Fri May 07, 2010 11:09 pm

Post by bravi »

Hi Naz,

Great update on your earlier schematic. What are the currents expected in tubes in 1,2 and 3?
Regarding input resistor for MC input can the 47k resistor be replaced with say a 220ohms or lower if I want to use a low impedance Mc cartridge such as OrtofonMC20.

Cheers!

B.Ravi
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Hi Bravi, current through stage 1 is only about 7.5mA and a tad lower through 2. Stage 3 is running at about 9mA for lower OP impedance but could be taken as high as 15mA. It might appear strange running such low currents but linearity is not a problem with CCS, nor is noise an issue, even through the 6C45P where you might have expected an increase due to lower gm at this current.

The 47K input R is just a generic starting point and you can parallel whatever load R you wish.

Also, tube rolling is encouraged and you can expect to hear minute differences even with different date codes. In all cases fine tuning can be effected by experimenting with CCS current through stages 1 and 3.

I also strongly recommend playing with the dampening R across the grid choke. Although I've adjusted for flattest response I personally prefer a higher value which will provide bass boost centred around 20Hz. Depending on your system you might prefer values from 47K - 100K (0.5db -> 4db boost respectively).

Cheers,
Naz
currituckco
Posts: 6
Joined: Fri Apr 29, 2011 5:17 am

Post by currituckco »

Hi Naz,

I'm interested in trying to build a version of your design (the all-tube one). I'm not a very experienced designer myself so I was wondering if you might share the power supply that you designed for it. What kinds of regulation did you use? Chokes, tube rectifier? Solid state rectification and VR for DC to the filaments?

Thanks very much. I'm also wondering what the prices are for the custom iron. (That would be a question for Dave Slagle). I'm wanting to switch from an active tube pre to AVC, but will need to build a new phono stage as well. I'm using denon 103R through cinemag SUTs. The amplifier is a homemade Bugle SET 45 into Altec 604-Ds.

Thanks very much.

kevin
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Hey Kevin,

My personal Ref Phono utilises a very complex hybrid regulator with tube rectification, SS voltage ref and error correction with tube OP so I developed this simpler two tube version for other Phonos I've built and modified which works extremely well, not as accurate but very quiet and very very musical. The best part is what you can achieve with the right tubes. Pre 90's 6H30P DR and a Ruskie 6N6P with the pentagon logo is a killer combo. These tubes, along with the Mullard, make for a relaxed, natural, 3D presentation that is detailed, open and airy!

I much prefer tube rectification and short of going all out with DHT I highly recommend Mullard 6CA4/EZ81 50's Box getter for the job. Following that, a small film cap, max 4-6uF followed by a 20 - 40H choke and a good 100uF cap is enough if you use a regulator.

I always use DC for Phono heaters and a simple 3 pin reg is the easiest. I don't hear a difference between different methods of DC on IDHT heaters but some claim to.

Not sure about current pricing for iron so better check with Dave. I can vouch for his iron though:-)

You should get great sound with your rig and if you are up to the time and effort this Phono and PSU will not disappoint!

Cheers,
Naz
Attachments
Simple Two Tube Regulator.jpg
Simple Two Tube Regulator.jpg (176.02 KiB) Viewed 64102 times
currituckco
Posts: 6
Joined: Fri Apr 29, 2011 5:17 am

Post by currituckco »

This is great, Naz, thanks! A couple other quick questions (thanks for bearing with me.)

What are the specs on the power transformer you used? Do you use a separate filament transformer?

I notice here you're using LM329 to bias the 6H30 cathodes. Would you still be using the Schottky SICs for cathode bias of the 6C45p and 6DJ8's? And I noticed somewhere that you mentioned using 2 SICs in series on the 6DJ8's. Is that your recommendation? Any other changes to the above design?

I think I'll give this a try and see what results I can get, assuming I can source the iron! Thanks so much for your help!

k
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

No probs Kevin. To be honest I'm not sure of the exact specs for the power tranny I just used what I had, a toroid in my ref Phono and a standard EI for others. As long as it it can support the total current for the two channels with a healthy margin (100mA or so will do nicely) and provide the required input voltage plus or minus about 10%. Something in the region of 250-260V AC should do. The values may need tweaking a little for best performance if the input voltage ends up a long way from the spec but that's OK, just let me know and I'll give you the changes.

I've never worried about using separate filament transformers but always ensure that I have plenty of current headroom.

The LM329 is a great little ref for this application, way better than a standard zener but yes I still love SiC Schottkies for bias up to a couple of volts (each provide about 0.8V). To my ears, they better anything I've tried ... and that's a lot!

Good luck with it and feel free to ask any questions.

Naz
currituckco
Posts: 6
Joined: Fri Apr 29, 2011 5:17 am

Post by currituckco »

Sounds good.

I was just looking at the Tango design and was wondering how you arrived at your values for the inductors. The Tango is 1.8H and .45H, right? How did you chose 21.3H and .57H? I'm just curious your method, because I figure the more I learn about the design the better!

Are there any things I should look at besides DCR in choosing the inductors? I still haven't gotten any response to my email to Dave Slagle, though it's only been a few days.
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Dave convinced me to design my own LCR and suggested scaling it up from 600 ohm to 7K3 to make it easier to drive. In the end I chose a slightly different impedance so that I could use standard value caps. I'm glad I heeded his advice because whilst I approached the task with trepidation the result was a near perfect match with my Sim. Dave's suggestion of putting in some additional taps on the inductors for fine tuning actually got me over the line but the RIAA curve was so accurate I didn't even need them.

The DCR is unimportant and if you go with Dave's iron you will have no issues. I'm surprised Dave hasn't got back to you yet, he's normally very responsive.

BTW, although the grid choke is critical to the design its value is not as important as the other inductors. Again, the additional taps really help to fine tune the extreme LF to suit your system. I've found that designing in a moderate amount of boost is beneficial in most systems but it needs to be kept right away from the mid bass area. Increasing the value of the damping resistor will provide boost and changing taps will shift the centre frequency, currently set to around 20Hz.

Hope this helps.
currituckco
Posts: 6
Joined: Fri Apr 29, 2011 5:17 am

Post by currituckco »

Hi Naz,

If I ultimately didn't feel I needed boost in the extreme LF, what would be my course of action? Would I want to omit the grid chokes, and if so would I replace them with a resistor to ground? What would I lose from the design if I lost the grid chokes? Just curious. I ask because I use Altec 604-D's which don't play much in the extreme LF range (20-30hz) anyway. I wonder if a bump there would make them feel more extended or just be lost on them. I like the focus and detail in their bass right now, drums and acoustic basses sound really lifelike and accurate to me.
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Hi Kevin,

Good questions. The grid choke is a key feature of the design. To remove them would actually result in a ~2.5db cut at 20Hz, the response being flat down to 80Hz. It's impossible to cover every aspect in detail here, suffice it to say that if you were running sub(s) for extreme LF extension this drop in response would most likely help, but with full range speakers the lack of bass will be noticable ... detailed but thin sounding.

The values shown provide a flat response to 20Hz and you will need to start with this. I only commented on the LF boost because in every system I have had experience with, unless you run subs, a small amount of boost at the deep bottom end helps overall system balance and it's precisely the speakers that lack bottom end that benefit most. In fact it's amazing just how much better balanced the sound becomes from top to bottom by merely balancing (at least to some extent) the natural rolloff exhibited by the speakers, even though things may already sound good.
currituckco
Posts: 6
Joined: Fri Apr 29, 2011 5:17 am

Post by currituckco »

Awesome, thanks for the explanation!
currituckco
Posts: 6
Joined: Fri Apr 29, 2011 5:17 am

Post by currituckco »

A power supply question -

If I use a 720vct power transformer, will a 6CA4 rectifier be robust enough, or should I use a GZ34? This seems like it would be a suitable xformer, especially given the high 6.3v heater current demand of the circuit:

http://www.edcorusa.com/products/589-xpwr002_120.aspx

it has a 5v tap so I could use a GZ34. i guess the trick would be getting the proper B+ out of it.

I'm playing around in PSUD for the first time, trying out different things.
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Wow, nice trannie! The 6CA4 is more than up to the task and this is not affected by the size of the trannie anyway so only use something bigger if you particularly like it.

One advantage of 720V ct is that you can go with choke input. Should be no need for a small input cap because it will increase input voltage by anything up to 1.4 times depending on its size (which is also a great method of "tuning" the input voltage). I'd go LC with a high H choke (10H or greater) or LCLC with smaller.

The regulator will give you approx 250V B+ irrespective of input voltage and this is not critical anyway. You will want about the right input voltage for the reg to work best.

Naz
fvale
Posts: 9
Joined: Fri May 13, 2011 9:34 am

Post by fvale »

Today I ordered Dave inductors and grid choke in order to build this interesting LCR phono stage. A couple of questions.

Power supply I've in mind is dual mono with old UX-280 globe DHT rectifiers. In order to avoid expensive pre nineties 6H30 tubes in the regulator, would it be possible to adapt a 6N6 here too? Maybe, as current drained is half.....

Also, both to mix tubes and save some money, I'd like to use a 6N6 on last stage hopefully getting a lower output impedance (my preamp is a passive autoformer). Can this be done?

Thank you for your support!
Fabio
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

Hi Fabio,

Glad to have you on board, I'm sure you won't be disappointed. To answer your questions, almost any rectifiers will do. I haven't had any specific experience with UX280s but start with those if you have them, their specs are fine and they do look pretty cool! Ultimately, the sound quality comes down to the individual tube brand and vintage but don't ever underestimate the difference rectifiers can make.

As for the 6H30P, there are many possible substitutes including the 6N6P. You could use the cheaper versions of the 6H30P or change to almost any tube you have with one or two minor changes. To use a 6N6P the easiest way is to reduce the bias voltage on the cathode by about half in order to compensate for the difference in tube gm.

No problem using a 6N6P in the 3rd stage or indeed a 6H30P for even lower OP impedance. I've done both and found that it's really down to the quality of individual tubes. No change is necessary to the circuit when substituting with a 6N6P. You will lose a few db of gain but that shouldn't pose a problem.

Cheers,
Naz
fvale
Posts: 9
Joined: Fri May 13, 2011 9:34 am

Post by fvale »

Hello Naz. I'm also sure I'll not be disappointed!

About UX-280, I used two of them in the tube gain stage for my digital sources and got excellent performances, with a quite incredible sense of ambience that I relate, also, to the good PS (all MKP caps, Broskie's Janus shunt regulator).

Good to know a 6N6 can be used in the last stage. Also, I briefly checked, 6N1P, 6N23 should fit.

I'd be curious to know the reason you chose the 6H30 as many other tubes could be adopted. For a matter of sound maybe?

Today I will order power transformers, separated for filaments and B+. I've some 20H 180R inductors that could fit. A 400-0-400V secondary will give about 335VDC with a drain of 25mA and 325VDC at 35mA. Should be fine... but I'd better to know if tube regulator burns more current! Please let me know. Thanks again Naz.
Naz
Posts: 71
Joined: Wed Aug 06, 2008 4:27 am
Location: Sydney Australia

Post by Naz »

About UX-280, I used two of them in the tube gain stage for my digital sources and got excellent performances, with a quite incredible sense of ambience
That is exactly what I look for in a tube so they will be a great choice
I'm be curious to know the reason you chose the 6H30 as many other tubes could be adopted. For a matter of sound maybe?
Precisely. Of the tubes I had on hand this one sound the best in terms of air and sheer musicality (but only pre 90's). There are plenty of others though so by all means use the ones you like most. The point I can't stress enough is the difference that good tubes will make to the overall sound in any of the PSU positions.
but I'd better to know if tube regulator burns more current! Please let me know
The reg only burns additional current in the cascode portion (6H30) and that's less than 3mA!

Cheers,
Naz
Post Reply