Jump to content
The Klipsch Audio Community
Jeffrey D. Medwin

Thinking of doing a Dynaco / DYNA PAS Preamp

Recommended Posts

Sometimes, as a DIYer in audio building, one gets an idea ( like me on 3-1-20 - to extensively re-do a  DYNA PAS ) and you have to LIVE with the idea for awhile, just to see..... a) how it feels to oneself and ....b) to determine if is it the best feasible course of action.

 

After 19 days of doing the above, I have concluded that I will NOT re-do a PAS. 

 

Instead, I will soon commence to do a from-scratch-build of a " phono-line " preamplifier, something I designed in Q4, 2019, after two months of planning and simulations.  

 

This will be point-to-point wired.  It will employ two 12AX7 sections in the phono stage.  But instead of a negative feedback RIAA ( Dyna PAS, ARC SP3A1, etc .) it will be a " passive " or feed-forward RIAA EQ.  A passive RIAA circuit with two 12AX7s, would be roughly - somewhat like Jim Hagerman's basic Coronet 1 or 2 or 3 phono stage.   

 

There will be one major and important ( to me ) change to the basic audio circuit.  Instead of capacitor-coupling the first 12AX7 to the second 12AX7  stage, I will eliminate that capacitor-coupling totally and directly-couple the two 12AX7 gain stages.  Just connect both stages using Cardas polished silver 19 AWG wire. A direct-couple is the simplest and best sounding way, IMHO, to connect two tube stages.

 

For the line stage, I am leaning to using a single 12B4-A triode tube per channel, operated so it can drive a 10K load.  ( Its low mu, 6.5 times ) .  Of course, no cathode follower will be employed.  Jim Dowdy ( Dowdy Lama ) has provided a 12B4-A operating point suggestion that I like, and will adapt.

 

The plans for a power supply, is what will set this preamplifier totally apart from all others, ever done before in tube audio, to my knowledge.

 

I intend to use a highly over-specified, beefy power transformer, more suitable for powering an amplifier than a preamp. I will LIKELY use two tube rectifiers, perhaps 5AR4s.  The B+ filter will start out as L1/C1/L2/C2/L3/C3, with Ls  under 10 Ohms DCR, and all B+ Cs being WIMA DC LINK film caps.  An all-film-cap B+ supply. 

 

Importantly, most importantly, I intend to DOUBLE-SERIES -SHUNT-REGULATE each B+ node, feeding the plate resistors of all the triodes inside the unit. 

 

Double series shunt regulation is absolutely wonderful to hear, it has been a listening revelation.  My 2019 KT88 DC amp has it, for its front end stage, and we love listening to what that does.  It acts to provide the maximum in signal resolution that a tube is capable of, IF fed right.   Lovely -to- hear detail, and dynamics, when listening.  Each triode stage, will get that double-shunt treatment - in the new preamp build. 

 

Additionally, each plate resistor will also have a proprietary Dennis Fraker ( Serious Stereo ) Final Filter.  This is applied ( an L/C, a small but low DCR L, and a high quality C ), on every individual tube's plate resistor, with "zero lead length".  This puts a separated B+ supply - right at / on each tube's point-of-use.  Ohh-wee !!  These L/Cs buffer each triode section, from interacting with the other stages, through a shared B+ supply.  

 

If I recall, the audio circuit might draw 70 milliamperes, but the multiple SHUNTING, to stabilize the B+es, will draw 110 milliamperes, so.....it does about 180 mA. of total draw !!  A power supply such as I have described, is unique in audio - to any preamplifier.

 

This will be my first tube phono / line stage build ever.  And I hope to use it henceforth, and not feel any need to ever have to replace it.  I am so curious to execute the new design, mainly to hear this  -  and to enjoy a large record collection.

 

Jeffrey Medwin

 

 

    49284328_12B43-20.JPG.cedf8128aa63812bbd5da24116c483ab.JPG

 

 

See:   https://community.klipsch.com/index.php?/topic/187956-autoformer-volume-control/&do=findComment&comment=2471221

 

 

    

Share this post


Link to post
Share on other sites
23 hours ago, DirtyErnie said:

This seems like an appropriate thing to put here:

 

Spice and the art of preamp design

 

Thanks for posting that. Actually, its pretty far off course in relation to what I plan to build. 

 

Yes, Norman did an impressive job documenting things, and simulating things.

 

The majority of my approach, and what "I" find to be important in a build, is quite different from Norman's, which was a rather conventional approach.

 

This planned from-scratch preamp build will be unique to me Ernie.  It will, on-purpose, in several areas, depart from everyone else's conventional practices.  ' Mostly ....   in how fastidiously the power supply is executed. 

 

I've already outlined this, ( how and where ) in prior posts.  Thanks for contributing sir.

 

Jeff 

  • Thanks 1

Share this post


Link to post
Share on other sites

Jeffrey, what is the DC current draw difference between zero signal and maximum voltage output going to be?

 

 

Maynard

Share this post


Link to post
Share on other sites
On 3/24/2020 at 11:11 AM, tube fanatic said:

Jeffrey, what is the DC current draw difference between zero signal and maximum voltage output going to be?

 

 

Maynard

 

 

I will let you figure that out.  You tell me.  Two 12AX7s at 0.75mA. in phono stages and a 12B4-A as line stage idling at about 33 mA. , per channel.

 

Maynard, are you are wondering, why so much - 110 mA. shunted current ??   

 

That comes from me PSUD simulating the supply, and using / adding shunt currents to get better waveforms, section-to-section in the power supply's multiple sections, PLUS, having a minimum of two filter sections shunted, preceding / feeding each triode

 

A total brute-force power supply.  "Out of the box" thinking, this is my own unconventional DIY audio design.  VERY few parts, all is passive.  No solid state parts.

 

I will construct this phono preamp on a usual standard - sized chassis, about 11 by 18 by 3 3/4 deep, 14 gauge steel with welded corners.  A favorite choice, cost VS high performance, for each shunt resistor is an  ARCOL Chassis Mount, HS series. 

 

I do not want to dissipate a lot of R-SHUNT heat into the chassis, so I may mount everything ARCOL using several large gold annodized heatsinks, on top - in the center area of this chassis,  ( between the power transformer in the rear, and tubes and RCA jacks up-front. )  ' Will place these heatsinks  side-by-side on one 1/8 th inch aluminum plate, but not attached intimately to the chassis.  That way, those ARCOL Shunt Rs will not transfer excess heat, on a long term basis, into the unit.  This power supply SHUNTING may generate around 44 Watts of heat dissipation continuous, but through about eight or more 50 Watt rated resistors. 

 

Why do this?  Double shunted, all-film cap B+ supplies sound really good, fantastic to me !!   That's why.  So, why not ?? 

 

Jeff 

                                        1612092664_GoldHeatsink3.jpg.7ddc4e6e0fd59782a7b3bd6cdef6eef9.jpg

                                     

 

                                                                      

 

 

Share this post


Link to post
Share on other sites
On 3/25/2020 at 12:44 PM, DirtyErnie said:

Sounds good, can't wait to see the finished product.

 

 

Hi Ernie, 

 

Here is what I have done, so far, as to chassis layout - notice, Power Transformer, which has large fields, is tucked away in the rear, and actually catty-corner to phono input 12AX7s.  Two 5AR4s as tube rectification. :

 

    1879900913_P1010048EDITED.thumb.jpg.30b00aa60accbccd7bc42b6de4f012f3.jpg

 

 

Audio circuit draws 70 mA., but my passive B+ DOUBLE SHUNTING, draws another 110 mA., so entire stereo phono/line stage will draw a whopping 180 mA.   Choke input, Flywheel, followed by two L.S.E.S. chokes, no chokes over eight Ohms in  DCR.

 

     

Share this post


Link to post
Share on other sites
On 4/15/2020 at 10:23 AM, DirtyErnie said:

Beastly!  Nice work!

 

Decided to add a LINE STAGE input, to access the 12B4A line stage tube.  This photo shows new re-positioning of tubes, and newly added pair of LINE RCA jack INPUTS, and a strategically placed new SELECTOR SWITCH ( Pnono /  Line) . No, what is pictured is not the ugly Selector Switch knob I will use.

 

Notice how short I have made all the wire paths, into and out of this preamplifier, VS: what is typical.  Preamp's Output RCA Jacks are on right-hand vertical side of chassis.  Unseen at this angle.

 

   966422035_P1010004edited2.thumb.jpg.3667f93b42da62c90d465cffd6b7b4ab.jpg

 

If you appreciate Classical music, please learn about, and LISTEN to this person's conducting :

 

                                                         https://en.wikipedia.org/wiki/Sergiu_Celibidache

 

Enjoy.

 

Jeff

 

BTW, tube line-up is     PHONO  :  12AX7 , 12AX7,      LINE  :  12B4A

Share this post


Link to post
Share on other sites

Whew, the first thing you do to develop a new design is load-test the power transformer, under a known line voltage, in my case 123.3 VAC from the wall. 

 

I did this carefully - safely,  last night, so I can order parts and get this build under way.  This testing under load took 141 Watts worth of resistor power dissipation to do, with this preamp's  ( very high) current draw, of around 160 mA.   Out from storage came four 100 Watt and over, power-rated resistors, to get the correct ohmic load.  ( Simple Ohm's law stuff here. )

 

Once one knows the loaded voltage, on half the power transformer's secondary, you Ohmiate the power transformer's windings and apply the DCR and VAC figures into PSUD 2, a free on-line simulation.

 

Well, I worked 2 hours doing PSUD simulations and computations last night, and another 5 hours this morning.  Seven hours and counting !!  Used four legal pages of space, so far,  to do computations.  The trick, in a directly coupled circuit, is to get ALL the voltages you want, to be had with STANDARD VALUE resistors, and, get the two directly coupled stages to have proper bias, and a balance of decent operation points, for both tubes, while they act as ONE . 

 

To get a needed resistive value, one can not series-connect Rs, anywheres, 'cause it produces SKEWS and it simply sounds terrible.  One can parallel two to three resistors, to get a value, but the resistors MUST all be of the same precise value, or you will also get skews.  Equal value paralleled resistors, will generally sound better than a single resistor, if the multiple resistors are matched WELL to each other. 

 

( That is why I have been harping on you " tube rollers" to use two tightly matched 2 Watt QUALITY resistors , in place of what you have, for your amp's input tube's plate resistor. It becomes an easy, cost-effective upgrade in performance, more-so than tube rolling in my experience ).

 

After seven hours, I still do NOT have the direct couple computed and simulated where I would like it to be, using standard resistors.  There are NINE sections to my B+ filtering, to the 12AX7 input tube, of my phono stage.  I can juggle 2 or 3 variables, in each stage of filtering.  I may do another two hours of work, and IF one does not get it, you simply think about it ALL, and in a day, or two, or three, you come back and try again.

 

My intended result, will be worth my effort, and that alone keeps one in the ball game !!    Just lettin' off steam, writing this, after seven hours at it !!!  I need to next, inspect the "menu" for Ohmite Brown Devil wirewounds, as a third set of power supply resistor value choices.  Where is that Ohmite catalog ??

 

Jeff 

 

1542637426_SNIPofSupplyMasterEDITED.jpg.f7055a90656e23354373de9eb03018bb.jpg

 

                                                                                 

                                                                                        Double Series Shunted Power Supply

 

  775798777_TwoPairsofLINEINPUTJacksedited.thumb.jpg.5053f564ee42442336ab4990df1ce73a.jpg

                        

 

EDIT:

 

The design task was woefully underestimated !!   It took 3 1/2  full - time days of work ( Computation and Simulations ) to get the circuit to balance as originally envisioned.  'Am physically and mentally sapped tonight, ....but also elated.  There were 26 legal-sized pages used, to take notes and do documentations.  On the 42nd PSUD simulation /  iteration, it all gelled.   This was a corona-virus precaution / mandated stay-at-home non-stop effort, except for sleeping and eating. 

 

' So happy to have achieved the goal - which was for the multiple power supply sections and direct coupled phono stage's 12AX7  tubes to balance, voltage and current-wise, using standard value resistors.  On simulation number 42, this got to within 0.1 volts of the goal.  Considering that this supply starts out at over 510 VDC, one tenth of a volt is truly good enough !!

 

I own about 10 to 12,000 L.Ps, so the incentive to design and build  the best possible performing preamp is high.

 

Have you ever seen such an overbuilt preamp ?  This starts with a Power Amp sized power transformer, and uses two 5AR4 rectifiers, and seven Ohm and less DCR filter chokes.  How about use of TWO 10 AWG m22759/11 ground bus returns from the power transformer to ground, one dedicated only to phono grounds, the other to the 12B4A line stage??  There is actually, much much more to this design than that.   I can't wait to hear this on various L.P.s !!   

 

Come to think of it, John L. Hasquin, an E.E, built "something" somewhat like this  about 14 years ago., for his own personal use.  He loved it.  His, as I recall, used solid state rectification and pentodes, whereas  mine employs proven-over-time to me triode tube choices.

Share this post


Link to post
Share on other sites

The newest stock Hammond 167 Series 6 AMP power transformers are small, which is nice, but they run far too hot for 30 to 100  year reliability.

 

I use these transformer in my latest gear, amps use dual EH5U4GBs, and this preamp  - dual 5AR4s.  These rectifier tubes used in pairs, require a 6 Ampere 5 VCT power transformer.  Hammond's 167Q5 is a good choice.  

 

Dennis Fraker shared a great DIY idea, which I implemented yesterday.  I drilled holes for air cooling in their end bells.   Not yet repainted. 

 

See photo.

 

  235283029_HOLY167R5alaDF.thumb.jpg.97008a09717d65b78ad398d235dcd2c4.jpg

Share this post


Link to post
Share on other sites

        1159382529_P10100164lengthwisespreadapartEDITED.thumb.jpg.7de5545f3d478070e8e4862638b1ed61.jpg

 

                                                              It is lovely, not being limited to a DYNA PAS chassis.

Share this post


Link to post
Share on other sites

This is a basic stereo ATTENUATOR mounting assembly ( after two attempts ),  as of 4-22-20. 

 

It will become a two channel Volume Control for the new preamp design, to control the entire audio system.

 

It is comprised of two pairs of WIREWOUND precision three-turn pots, which will all turn together, via MXL toothed belts and non slip pulleys.  ( This is the equivalent of four precision wirewound pots, all turning on one common-to-all shaft. )

 

758215133_4-23-20SECONDAttenuatorAssembly.thumb.jpg.0165f4355db0e1b9fc4d7568811c2966.jpg

 

 

A pair of high quality linear taper wirewound pots is used for each channel. Each pair is to be wired backwards to each other, to become a constant impedance " L Pad " Volume Control.  As one pot increases, the other correspondingly decreases in resistance, and the output remains a constant impedance.

 

I have opted for one control, one Volume Knob, to turn both channels simultaneously.   I prefer a single Volume Control, rather than adjusting two controls, each time a slight volume change is desired.

 

Normally, a passive ATTENUATOR control would be used by me as a separate system component.  By incorporating this assembly inside this preamp design, driven by a high-current 12B4A linestage,  it eliminates using an extra interconnect and two pairs of signal-degrading RCA jack connections.

 

KISS .

 

Jeff Medwin

  • Like 1

Share this post


Link to post
Share on other sites

I would settle for just the ruler.

Share this post


Link to post
Share on other sites

Newly designed  ( 4-24 and 4-25--2020 )  mounting assembly.   

 

One assembly used for each of the preamp's ( six ) triode sections. 

 

Forces short lead lengths, and has attachment points to build in a three dimensional manner. ( Most folks trend to lay out, build in two dimensions. )

 

   391065138_P1010036EDITforPUBLICuse.thumb.jpg.3afb0ae5bbeaf2693b9d31037b8dfd0b.jpg

 

 

   1500283116_P1010041REV1.thumb.jpg.6a1b03f2f2280fd54da4fff616826eb8.jpg

 

 

This new mounting assembly uniquely solves three design goals at once : 

 

1) it securely attaches each 15 uF  WIMA cap inside the preamp, 

 

2) it positions each B+ power supply cap, within one inch of the tube socket

 

3) and it allows for easy one bolt installation or removal, if needed,  for servicing or future modifications.

 

Note  Imagine this, each 12AX7 RIAA input tube has 19 parts associated with this assembly. Each 12AX7 middle stage has 18 , and each 12B4A line stage has 25 parts .

 

This new mounting assembly is truly necessary, to keep this preamp build compact, uniform in layout, and orderly.

 

This Saturday, 4-25-20 is a great day to start on the chassis, carefully position and center punch most holes - perhaps do some drilling.   14 gauge steel, hand tools only.

 

Jeff

 

 

  1566863504_P1010048EDITED4.thumb.jpg.1999cb5e5ac49444125e8e5f1ceedb08.jpg

 

  432430016_P1010029(2).thumb.JPG.8adc34fd66143ae44910c4b3d2e61804.JPG

 

  

 

 

  

Share this post


Link to post
Share on other sites

NOT two, but THREE full days of tedious hand labor :

 

Started with a " blank " chassis.  Three days !!  Center-punching, drilling and either tapping ( 6-32) or  Greenlee chassis-punching holes through 14 gauge steel.  Lotsa' hand filing .  Hand tools only.  This chassis and bottom cover weighs in at 15 pounds.

 

Next, unbolt the heatsinks and send this off for Powder Coating, Midnight Blue WRINKLE.

 

  P1010048.thumb.jpg.425d00a2cabdaa5866da6c93c2d28d1b.jpg

  

 

  1422721141_P1010053EDIT2.thumb.jpg.128dab6a076ceb6843805b4fcdaa886f.jpg

 

  

 

 

  

 

 

                                               Rectifier tubes Filament Transformer, with painted  covers, powers dual 5AR4s,  Mullard reissue :

 

  P1010043.thumb.JPG.e88e782e19dffdd67c14a0a5532c1f9c.JPG

 

 

Share this post


Link to post
Share on other sites

5-1-2020 ................Progress Photo :

 

   1867831533_BOTTOMCOVEREDITED.thumb.jpg.6bf284f7c2f25573e4e0988b625b1963.jpg

 

                                                                                         Ordered Powder Coat 5-1-20.

Share this post


Link to post
Share on other sites

5-3-2020   

 

Today I cut out plexiglass, and shaped it .  It will be used to mount ALL RCA jacks, rather than attach the jacks, with their own insulation washers, directly to a metal chassis. In effect, this DOUBLE insulates all RCA jacks. 

 

I chassis punch 3/4 inch holes, where RCA jacks are to go on the chassis, but the jacks themselves get attached to the plexiglass, which is bolted onto the metalwork, with the RCA JACKS properly positioned in the chassis.

 

   534476701_P1010047(2).thumb.JPG.3d34527abd34b449486207373142ad88.JPG

 

 

Why not ??   If I am going to go to the trouble to DOUBLE series shunt regulate ALL the B+es, I can DOUBLE insulate ALL the RCA jacks also.  This way we minimize potential Eddy currents, and Sammy currents also.   :-)

 

Jeff  Medwin

Share this post


Link to post
Share on other sites

5-7-2020                                             CHASSIS at Powder Coater's Shop.                                                  Be ready 5-11-2020, Monday.  

 

 

    P1010017.thumb.jpg.56ab9d3e003525f1673cd6b1c43f5b82.jpg

Share this post


Link to post
Share on other sites

5-11-2020 

 

Back from the Powder Coater's shop.

 

Ordered lots of parts, last four days from six different providers.

 

I am so glad to have this chassis finished and designed as it is !!!

 

Have been brainstorming last few days on different " Building Module  " configurations, to be able to attach ALL the many parts, for each triode, on it's own building module, one that positions parts for SHORTEST possible lead lengths, and in THREE dimensions spatially.

 

 Bare 14 gauge steel chassis alone, with bottom cover, weighs 15 pounds.

 

Jeff

 

  P1010020.thumb.JPG.e224654667ba1bd2ed58d9e833adaa59.JPG

  1697615276_P1010025NEWedited.thumb.jpg.90d350be89221de309e44278891c3b69.jpg

 

  

P1010018.JPG

 

792350516_P1010039edited.thumb.jpg.d8822249deb07c5193f226a443e61fa5.jpg

  • Like 2

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...