K&K Audio / Lundahl Transformers

Welcome! Need support, you got it. Or share youe ideas and experiences.

Return to K&K Audio / Lundahl Transformers

K&K Audio / Lundahl Transformers

Re: Brady, Toccata & queries

24.136.237.79

[ Follow Ups ] [ Post Followup ] [ Thread: Display All Email ] [ K&K Audio / Lundahl Transformers ]

Dave Davenport

In Reply to: Re: Brady, Toccata & queries posted by gdahl on November 12, 2003 at 10:21:42:

Hi Gary,
You wrote:
The K&K phono stage doesn't have a line stage. The Amity was designed to be driven by a line stage, such as the Raven (currently being revised). The K&K balanced line stage is essentially the same circuit with different tubes, and would seem to be a perfect choice for this application.
Which tubes has Lynn chosen for the latest revision of the Raven?
Dave Davenport

Follow Ups:

Re: Brady, Toccata & queries - gdahl 21:55:19 11/12/03 (21)
Hi Dave,
We are still sticking with the ECC99. The changes are (for now): using an input transformer (1676, or maybe the 1684) and a balanced DACT attenuator instead of the TX-102 (which in its current form is not suited for phase splitting), ditching the parallel-feed topology (which makes it pretty much the same thing as the Amity's driver stage), and using the 1674 as an output transformer instead of the Sowter 9111. We also will try a few ideas that we have for minimizing (or, better yet, eliminating) DC offset.
I don't know if Lynn endorses all of these changes, but I am building mine this way. We'll decide what we prefer after we have had a chance to test it, listen, and compare some alternatives.
Gary Dahl
"So many tubes, so little time..."
- Re: Brady, Toccata & queries - Dave Davenport 06:59:57 11/13/03 (19)
  Hi Gary,
  Well the ECC99 is certainly a lively tube and I can not fault you for using it here, to me it is a matter of preference.
  Lynn's page is somewhat out of date and still shows the TX-102 for an attenuator, so I am not sure how you have the DACT configured. Can you elaborate?
  Ditching the parafeed??? Why? I can only suppose that you have had some good results with the Amity configuration. Or maybe a desire to eliminate the parafeed cap? But what about the Ultrapath cap? And if you can balance the circuit to get rid of the DC offset for this, isn't that also applicable for a parafeed configuration?
  Dave Davenport
  - Re: Brady, Toccata & queries - gdahl 12:54:06 11/13/03 (18)
    Hi Dave,
    The original Raven used 5687, 7044 or 7119. These were chosen for their linearity, as well as their suitability in terms of gain and plate resistance. When the ECC99 became available, they replaced the previous tubes because of their sonics. We haven't tried any others yet.
    As for the DACT attenuator, two decks per channel are used. The grounds are common. Each end of the input transformer secondary feeds the input of its deck, and each wiper goes to the grid of its respective ECC99 section. The DACT's tight tolerances keep the phases properly matched. The attenuator's input impedance is essentially constant at all volume settings, maintaining consistent termination of the input transfer and an unchanging load for the source component.
    For simultaneous level control of two channels, a 4-deck DACT (CT2-10K-4) works perfectly. For dual mono use, two 2-deck attenuators (CT2-10K-2) will serve.
    As for ditching the parafeed, you're abolutely right--it is because of what we learned when working on the Amity/Aurora amplifiers. The Raven is actually very much the same circuit as we used in the driver stage of my Auroras, excepting the attenuators and the actual transformers. The first version of the Aurora that we built used a current source feeding each ECC99 plate, and a blocking cap between primary halves. Lynn was disappointed in the sonic results, because it sounded slower and less transparent than did his original Amitys. A few cap swaps later, we realized that the blocking cap was highly audible, so we reverted to the Amity's driver topology, which feeds the CT of the LL1635 from a low impedance (VR shunt reg). The 1635 obviously liked this a lot better; the sound became quicker and more transparent, and the cap coloration disappeared. We realized at that point that we would want to try the same thing in the Raven.
    Incidentally, I have been using the push-pull version of the LL1635, and no longer use a balancing pot between the cathodes. Measured low-frequency performance doesn't indicate problems with residual offset current. Furthermore, I am less convinced that abolute "dead-center" makes such a huge sonic difference, after having the opportunity to listen to an amplifier equipped with a knob and a meter for the purpose of nulling the offset.
    As for the ultrapath cap, it seems to be much less audible than a coupling cap or parafeed blocking cap. This makes sense, considering that in a Class A push-pull circuit, it only carries difference signals, which are considerably lower in level than the main (balanced) signal. (See Lynn's "Loop Distortion" article for further info).
    For all of this, we'll have to build it, listen and measure before we can conclude whether the same thoughts that improved the amplifier will do so in the Raven.
    Gary Dahl
    "So many tubes, so little time..."
    - Loop Distortion by Lynn Olson (Open in New Window)
    - Re: Brady, Toccata & queries - Francois G 10:54:43 11/14/03 (6)
      Gary,
      Regarding volume control -- have you tried "bridging" the primaries of the input transformer with a single potentiometer instead of using the dual control in each side of the primary? If I understood correctly Dave and Kevin listened to various options in the K&K Raleigh line amp design and preferred the bridging method, in spite of the fact that it changes the loading on the input transformer at changed volume settings.
      I shall appreciate your comments, and Dave's too.
      Francois
      - Re: Brady, Toccata & queries - Dave Davenport 21:23:16 11/14/03 (4)
        
        In Reply to: Re: Brady, Toccata & queries posted by Francois G on November 14, 2003 at 10:54:43:
        
        Hi Francois,
        Gary is correct in what he says about the balance and constant reflected input impedance of the configuration he is using. I know that Lynn has used this configuration for a while but I had not asked them about it before. That is why I was interested. Kevin and I have compared three different differential attenuators. The first is that which Gary is using, the second is the simple bridge and the third is a hybrid, which is a bridge with a center tap ground. Like Gary, I like the technical characteristics of the one that he uses; however I prefer the sonic characteristics of the simple bridge. I must say that it is only a slight difference, but nonetheless a difference. To me, I will almost always go with what sounds best even if I think another approach is technically superior. I used film pots for my comparison, but I think it is an apples-to-apples comparison. To digress for a bit, sometimes something will sound better at one stage of incrementally improving a design (a puffy way of saying "tweaking") and later after several changes, that way that used to sound better no longer does. But getting back to the attenuator, I use the bridge because to me, it sounds better.
        Now, I am going to open a can of worms with some speculation. I'm not sure how to say this... but I think chasing balance is the wrong approach. Balance is important only because you have a balanced transformer. For example, in the output stage, I agree that the balanced push-pull parafeed is not the best sounding choice between several alternatives. If you look at the output circuit on the schematic for my line stage on the K&K Audio site you will see that it is a differential circuit but the transformer primary (although there are two sections in series) is really a single, non-balanced circuit. The two halves of the primary are not pushing and pulling, but rather acting as one like in a single-ended circuit. Now, the important thing with this is that there is a single signal loop rather than two signal loops like in push-pull. So with a single signal loop, there is nothing to balance. If one side were a little weaker than the other, it would still work okay. Now taking this same thought to the input, I think the single unbalanced loop has the same advantage. If this is true, and I am speculating here, than perhaps this could be why the bridge sounds better.
        Dave
        
        Re: Brady, Toccata & queries - gdahl 23:05:34 11/19/03 (3)
        
        In Reply to: Re: Brady, Toccata & queries posted by Dave Davenport on November 14, 2003 at 21:23:16:
        
        I wonder if the results of your comparison of attenuator configurations might have been compromised by the film pot you used to test the traditional balanced configuration. There is a good chance that the film pot (Alps?) had some degree of tracking error between phases. Even a few percent of difference would add significant distortion, but this effect wouldn't occur with the bridging approach. As you said, AC balance is important.
        As far as chasing DC balance, I suppose it only matters to the extent that the transformer is affected by it. These little amorphous beauties are devastatingly sensitive to *any* offset--Gary Pimm ran some distortion tests recently, and found that even a few microvolts messed them up. What we get in return for this sensitivity is their wonderful liveliness and low-level dynamics. But if we use them as line output transformers, we need to either accept capacitor colorations (parallel feed) or have a *really* good servo. On the other hand, iron-core output transformers such as the LL1620 aren't nearly so touchy about offset. Allen Wright's amps had those wonderful meters and balancing knobs, but I didn't notice sonic degradation when minor DC imbalances appeared on the meters.
        Lynn had always wanted to avoid parallel feed in the Raven, but was unable to find an output transformer that would allow this. I think I have now found the transformer we were looking for, the LL1680/5ma. This unit can deal with residual offset, so we will be able to run it the same way we run the Amity's driver stage. The LL1680's voltage headroom will also be a plus. But of course, I'll have to build it before we can find out if the sonics come out as we hope!
        I would be very curious to compare the sound of a parallel-feed line stage (such as yours) with the revised Raven. Perhaps the differences would turn out to be just a matter of small flavor variations. But we sure did like the changes in the Aurora's sound that resulted from replacing its parallel-feed driver stage with the current version.
        As for the attenuator, my real hope is for a TVC that can properly split phase. John Chapman tells me that the nice people at S&B are presently working on such a project.
        Gary Dahl
        "So many tubes, so little time..."
        
        Re: Brady, Toccata & queries - triodenuts 02:10:33 01/01/04 (0)
        
        In Reply to: Re: Brady, Toccata & queries posted by gdahl on November 19, 2003 at 23:05:34:
        
        Hi Gary Dahl!
        Do you know how can I get in touch with John Chapman from Bent audio? Their website is not exist anymore. How can I get in touch with him?
        Thank you!
        
        Re: Brady, Toccata & queries - Dave Davenport 20:08:23 11/21/03 (0)
        
        In Reply to: Re: Brady, Toccata & queries posted by gdahl on November 19, 2003 at 23:05:34:
        
        It could very well be the case that my comparison of attenuators was compromised by slight imbalances between the potentiometer sections.
        Thanks for the insight and I'll have to think about this some more.
        Dave
        
        Re: Brady, Toccata & queries - Dave Davenport 20:07:28 11/21/03 (0)
        
        In Reply to: Re: Brady, Toccata & queries posted by gdahl on November 19, 2003 at 23:05:34:
        
        It could very well be the case that my comparison of attenuators was compromised by slight imballances between the potentiometer sections.
        Thanks for the insight and I'll have to think about this some more.
        Dave
      - Re: Brady, Toccata & queries - gdahl 15:47:05 11/14/03 (0)
        
        In Reply to: Re: Brady, Toccata & queries posted by Francois G on November 14, 2003 at 10:54:43:
        
        Hi Francois,
        I haven't tried the "bridging" method. As for Lynn's method, it works especially well with the DACT CT2 because of its tight resistance tolerance. The resistance sections of typical film pots don't track accurately enough to ensure equal attenuation of both phases, but the DACTs do.
        With the bridging method, if I understand it correctly, the input impedance decreases as the volume is turned down. The result is a lower reflected seen by the source component through the input transformer. Using the DACT makes it possible to present a consistent load to the source component at varying volume settings.
        I wouldn't be surprised if Kevin and Dave preferred the sonic results of the bridging method over the results obtained using two decks of film pot per channel, but I don't know if they compared it with a DACT in true balanced configuration. If they did, it might just be a question of a choice of sonic flavor. My preference is based on my feelings about the technical aspect of the issue; as I said, I didn't try it both ways.
        Kevin and Dave obviously have good taste and technical expertise, so I don't doubt that the bridging method works well too.
        Gary Dahl
        "So many tubes, so little time..."
    - Re: Brady, Toccata & queries - Tuba 21:02:35 11/13/03 (0)
      All this sounds very exciting! I'm really looking forward to your results of your "ideas for minimizing (or, better yet, eliminating) DC offset" so that amorphous core output transformers can be used. THis would really complete the picture :-)
      Cam
    - Re: Brady, Toccata & queries - Dave Davenport 13:35:52 11/13/03 (9)
      Excellent post Gary, thanks.
      You wrote:
      When the ECC99 became available, they replaced the previous tubes because of their sonics. We haven't tried any others yet.
      I have had good results with both the 6H30 and 6N1P. I would be interested in your assessment of them in your circuit.
      You wrote:
      Incidentally, I have been using the push-pull version of the LL1635, and no longer use a balancing pot between the cathodes. Measured low-frequency performance doesn't indicate problems with residual offset current. Furthermore, I am less convinced that absolute "dead-center" makes such a huge sonic difference, after having the opportunity to listen to an amplifier equipped with a knob and a meter for the purpose of nulling the offset.
      I have suspected that this might be the case but haven't had the time to check it out. My thoughts are that AC balance is more important than DC balance, but there again, it is only a speculation.
      Thanks,
      Dave
      - Re: Brady, Toccata & queries - JamesD 01:03:05 11/14/03 (8)
        
        In Reply to: Re: Brady, Toccata & queries posted by Dave Davenport on November 13, 2003 at 13:35:52:
        
        I have found that ac balance is the critical factor in PP stages. I use a pot to tweak the balance in each stage for best sonics or if I have one available, a spectrum analyser for best harmonic distortion profile (smooth roll off of harmonics). I was taught this by a club of old time tube engineers nearly 30 years ago. Of course this needs periodic resetting...
        Now I'm playing with CCS and true differential stages as per the latest discussions. To early for conclusions yet as sometimes they are better than the traditional and sometimes not...
        I abandoned parafeed fairly early on for much the reasons given above. I prefer neutrality and transparancy and good iron on its own seems the best way of acheiving this...
        ciao
        James
        
        Re: Brady, Toccata & queries - Dave Davenport 06:21:17 11/14/03 (7)
        
        In Reply to: Re: Brady, Toccata & queries posted by James D on November 14, 2003 at 01:03:05:
        
        Thanks for the comments James. Where is the pot in the circuit and what is being adjusted?
        Dave
        
        Re: Brady, Toccata & queries - JamesD 07:51:04 11/15/03 (6)
        
        In Reply to: Re: Brady, Toccata & queries posted by Dave Davenport on November 14, 2003 at 06:21:17:
        
        Hi Dave,
        The adjustment is to valve's operating point varying anode current, cathode voltage and/or grid bias voltage, I tried varying Va but found it inefficent. I aplly it through one of many methods:
        1) Pot between two cathodes with wiper to shared cathode resistor.
        2) pot between two cathodes with wiper to earth and fixed bias to C.T. of pp input transformer.
        3) cathodes to earth or shared cathode resistor or individual cathode resistor with cap connecting cathodes and individual fixed bias adjustment to 'cold' end of individual input transformer windings with windings in anti-phase i.e. pp. This typicallly also is cap coupled from 'cold' end of windings to cathode to complete the input current signal loop.
        4) pot between cold end of individual windings with wiper to fixed bias source again cap from bottom end of windings to cathodes.
        5) pot as individual cathode resistors with cathodes cap coupled.
        or any combination of the above methodes. I quite often mix fixed and cathode bias. I tend to always use the cathode as the input and output signal reference point rather than 'ground' per se. I often use bipolar psu so this helps keep things conceptually simple.
        I don't beleive in ground planes for valve audio circuits - it's just a source of noise. I use transformers to isolate each stage from each other electrically and like pp over SE except in musical instrument amps.
        Hope this is interesting.
        ciao
        James
        
        Re: Brady, Toccata & queries - Dave Davenport 08:23:51 11/18/03 (5)
        
        In Reply to: Re: Brady, Toccata & queries posted by James D on November 15, 2003 at 07:51:04:
        
        Hi James,
        Yes this is very interesting, thanks for the informative post. I have studied it and have been thinking about it and have a couple of questions for you.
        First, and probably most important, how does it sound? How does adjusting the operating point effect the sonics, and what differences do you hear between the different methods?
        I think that a capacitor between the cathode and cold end of the input is technically correct. However I have tried it and not heard any difference. Do you hear a difference?
        I think, if I understand your post correctly, methods 1,2,3,and 5 change the operating point of the tube and are what I call "DC balance. Method 4 is what I call AC balance; the pot doesn't change the DC bias but the relative levels of the two signals. Is this correct? What are the sonic differences?
        You use a mix of fixed and cathode bias. I have tried both fixed and cathode bias but not a mix. What is your experience with a mix?
        Dave
        
        ...Long and winding reply... - JamesD 11:56:35 11/21/03 (4)
        
        In Reply to: Re: Brady, Toccata & queries posted by Dave Davenport on November 18, 2003 at 08:23:51:
        
        Hi Dave,
        Sorry for the delay in replying, I'm 'on the road' and have only occasional 'net coverage'. Anyway to your questions...
        1) How do the various changes affect the sound?
        Hmmm... Very good question! As always in a high-resolution system every change makes a difference but lets have a general stab at it. I have found the character of the sound changes with changing distortion profile (no surprise there..) The distortion profile comprises several components such as harmonic and intermodulation etc etc. Fortunately tube amps with zero interstage feedback have the least number of distortion mechanisms and so I'm can generalise certain conclusions.
        For harmonic distortion the profile should show a decrease with between order 6dB and 18dB per octave with 12dB being optimum. It should also have orders higher than 6th in the noise floor. This provides the least THD generated sonic profile - moving away from this profile makes an amp sound tuby or ss in character - depending on which way one moves. So I try to voice my amps to achieve this. I also try to achieve the lowest overall THD figure as lower is better and increases the level of fine detail or overall transparency in the amp.
        For Intermod. then I try to minimise the ability of any stage to affect any other stage - this is also affected by the op point chosen - and I find that very well dynamically balanced op points in pp can be quite sensitive to intermod mechanisms as they are very, very transparent.
        I have found that I can use any of the mechanisms for adjusting the amp and get the same level of sound quality. Not always the same sound as each sounds slightly different and it is personal taste as to which is best. This was what caused me to try combining methods to try and get the best combination of 'voice' to the amplifier - or rather the most neutral to my ears... I have also found that at this level of finesse there is no absolute right way to configure any given stage � different valves and overall topologies change which method works best. Different components � particularly different transformers have a big influence on which method is best too...
        What I am tweaking for each time is the least sonic signature from the amplifier � to my ears and the most resolution. I find this also equates to the most relaxing sound from an amplifier i.e. a low distortion and high-resolution system is also the most relaxing and emotionally involving� it lets you listen deeply into the music or just ride it�s surface depending on your mood.
        This is not terribly descriptive of each method and how it changes things but I have not found � yet � that there are hard and fast rules except that the dc balance point is not the point of best sound� play with things and see what works best. Of course this introduces all sorts of other variables into the testing � such as what sort of day one has had so far, etc. etc. So when I have found a point I like I leave the amp for a few days and then change things back to how they were before and after a few days see if I miss the previous change � if I do then the change goes back in and stays.
        2. I think that a capacitor between the cathode and cold end of the input is technically correct. However I have tried it and not heard any difference. Do you hear a difference?
        Yes � most of the time but certainly not always, probably about 60:40. Generally I only hear it once the amp is close to being optimised unless the IT has some issues (in-circuit resonances) that this dampens. I try all sorts of things out including neutralising caps � they work in some circumstances and not in others. The neutralising caps are very system dependent as I always over size the current delivery of my driver stages in my designs but they can help and can be heard � sometimes�
        3) I think, if I understand your post correctly, methods 1,2,3,and 5 change the operating point of the tube and are what I call "DC balance. Method 4 is what I call AC balance; the pot doesn't change the DC bias but the relative levels of the two signals. Is this correct? What are the sonic differences?
        Conventionally this is correct for Class A1, I had a Class A2 pp amp that was in A2 for no signal and then Method 4 did move the dc op point.
        Having said that, I don�t really distinguish between the dc and ac op point when thinking about how to design and balance the amp. I think of static i.e. no signal conditions and dynamic conditions i.e. with signal. We listen to the dynamic condition so the static condition is only the starting point for setting the amp up. As L.O. has remarked, triodes are three terminal devices with differential inputs and outputs. The valve doesn�t care about topology or components etc. it just responds to the instantaneous voltage and current differences present on its terminals � there may be some �lag� or �memory effect� but the instantaneous �output� is directly related to the instantaneous �inputs� so I�m after the best sound under dynamic conditions and that can only be obtained by trial and error and listening� test signals are all quasi-static or highly repetitious or single event to enable us to capture and measure various parameters. This can be a good guide to getting into the right ballpark but the final set up is always by ear. Design and build by using your head but listen with your heart!
        So when I am adjusting the 'balance' of the amp it is only the 'ac' balance that really concerns me and I set this by ear using trial and error. Of course I measure everything I can think of because I hope that one day I will find a corresponding set of parameters that I set by measurement and get it right first time... some hope :-) As I stated earlier, I try to get the most neutral sound from the amp that is also the most relaxing and has the most resolution. My standards of neutrality are BBC based and so are centred on accurate reproduction of the human voice, piano, violin and chello followed by trumpet. I'm lucky in that my family members all play various of these instruments so I am always hearing them.
        By resolution I am refering to musical resolution i.e. the mix of timbre and timing that communicates the true emotional value of a piece of music. I find I am in agreement with a lot of what James Boyk writes about music and sound reproduction. I listen for the subtle beauty of music.
        Dave, I haven�t directly answered your questions � mainly because I don�t know how to yet. I�m still learning about zero feedback valve amps and I find that good ones are so transparent that anything and everything changes how they sound and there is very little overall consistency between amps but each individual amp is consistent with itself so some set of interactions that I don�t yet understand will one day explain all this�
        Hope this is interesting�
        Ciao
        James
        
        Re: ...Long and winding reply... - Dave Davenport 20:45:02 11/24/03 (3)
        
        In Reply to: ...Long and winding reply... posted by James D on November 21, 2003 at 11:56:35:
        
        Hi James,
        You wrote: Hope this is interesting�
        
        That was a wonderful post, one of the most thought provoking that I have read in a long time. I will certainly try some of those things that you mentioned in my next design. It has taken a while to respond because I wanted to digest it all.
        
        You wrote: For Intermod. then I try to minimise the ability of any stage to affect any other stage - this is also affected by the op point chosen - and I find that very well dynamically balanced op points in pp can be quite sensitive to intermod mechanisms as they are very, very transparent..
        You lost me a little bit here. What do you do to minimize the ability of any stage to affect any other stage?
        You wrote: What I am tweaking for each time is the least sonic signature from the amplifier � to my ears and the most resolution. I find this also equates to the most relaxing sound from an amplifier i.e. a low distortion and high-resolution system is also the most relaxing and emotionally involving� it lets you listen deeply into the music or just ride it�s surface depending on your mood.
        Yes, yes, exactly.
        You wrote: My standards of neutrality are BBC based and so are centred on accurate reproduction of the human voice, piano, violin and chello followed by trumpet.
        Again, right in line with my philosophy. I have several good female voice recordings that I use for a reference, but no good male voice. Do you have a recommendation there? The best piano recording that I have found for hearing differences when I make changes or tweaks is Keith Jarrett�s �The K�ln Concert.� I have used it as a reference for over 20 years. I haven�t focused on strings but do like brass instruments and also the saxophone for listening evaluations. Maybe the sax, because I used to play sax so I am very sensitive to the sounds of that instrument.
        
        You wrote: Design and build by using your head but listen with your heart!
        I�ll pin that one up on my wall :-))
        Now a question: Are all of your balanced differential circuits traditional push-pull topology or have you tried any of the differential parafeed designs?
        And if you would, please drop me a private note off-list so that I will have your your e-mail address. I expect that I will have a question or two for you when I try some of the things you covered.
        Ciao,
        Dave
        
        Re: ...Long and winding reply... - JamesD 05:57:18 11/29/03 (0)
        
        In Reply to: Re: ...Long and winding reply... posted by Dave Davenport on November 24, 2003 at 20:45:02:
        
        Hi Dave,
        Thank you for your reply. I'll attempt some answers below and then I'll send you an email.
        Regarding Intermod. Dave asked:
        > What do you do to minimize the ability of any stage to affect any other stage?
        Ohh...where to start. A bit of background. For twenty years, on and off, I designed, with various friends and colleagues, lots of solid-state audio circuits for hobby and professional use. During this time I came to realise that spurious signal injection into each stage (my broad definition of intermod) caused most of the residual problems once topology and components had been optimised. Of course the obvious feedback derived intermods are well documented but there are many other mechanisms and these apply to valves every bit as much as to ss. Here is a list of most of the mechanisms that I try to control...
        1) PSU related intermods. Voltage and current related. Regulation issues and pulse issues.
        2) Earthing related issues.
        3) RFI issues.
        4) Signal current return loop issues.
        5) Capacitive coupling - both ground plane and wire-to-wire.
        6) Inductive coupling - both ground plane and wire to wire
        7) Feedback issues - Local or interstage or global.
        Of course a lot of these are different views on the same fundamental problem - and so a solution to one might help solve another or make it worse...
        Fortunately by working only with valve Class A(1&2) circuits a lot of problems are minimised compared to ss - but to get the best results all aspects must be covered ( and probably more - I'm discovering other sensitivities all the time...).
        I have used a lot of interstage transformers in my ss design over the past ten years as I found they helped control a lot of the above problems. Needless to say most of my valve designs use interstages too. In particular the ability to break the grounding scheme for each stage from each other stage is a godsend!
        I tend to use star psu distribution too with local smoothing and or regulation as I think fit. The power feed is also taken through a forward biased diode to help isolation from the star point.
        I don�t like metal chassis! I know shielding can be useful but any metal chassis introduces too many capacitive and inductive coupling issues for my liking. I use wood, acrylic and open chassis design a lot.
        I could write lots more on intermod� maybe one day I will and put it on a website. And don�t get me started on guitar amps and intermod!!!
        Dave asked:
        > �but no good male voice. Do you have a recommendation there?
        Just a few:
        1) Anything by Roch Voisine (Canadian singer with a quite pure male voice � sort of a male equivalent to Eva Cassidy in style).
        2) Anything by Chris Isaak except for his live work where he is often a little flat�
        3) Peter Green�s Fleetwood Mac �Live at the BBC� especially �Man of the Word�
        4) Led Zeppelin �Live at the BBC�
        5) The Very Best of Louis Armstrong
        6) Richard Burton �Under Milkwood� and �Burton at the BBC� � both spoken.
        7) Richard Harris �MacArther Park�
        Then for Choirs:
        1) Treorchy Male Choir �A Garland of Welsh Songs�
        2) Handel �Messiah� � several versions.
        3) Mahler Symphony No. 8 �The Resurrection� � several versions.
        There are others but these are my consistent references and I have heard all except Chris Issak live more than once.
        Playing an instrument really opens up ones ability to judge how well it is being reproduced. I play guitar and bass badly but it still helps. The rest of my family play several instruments quite well so that helps keep me honest! Playing Sax would be a real help. All those woody tones right in the middle of the frequency range!
        I haven�t heard the Keith Jarrett � now I will search it out � thanks!
        Dave asked
        > Are all of your balanced differential circuits traditional push-pull topology or have you tried any of the differential parafeed designs?
        I have tried parafeed differential and I found I could hear the cap signature quite clearly when compared to non-parafeed interstage. It had it�s own strengths � particularly the ability to tune the extreme l.f. performance to work optimally with the rest of the system. In general I prefer to remove the cap signature from the midrange and treble whilst allowing the lf to go it�s own way� I have been toying with a valve sub-woofer amplifier and making that parafeed to take advantage of the lf tweaking it allows�
        As Gary Dahl says ��so many valves�so little time��
        Ciao
        James
        
        Keith Jarrett�s �The K�ln Concert.� - Bas Horneman 14:54:15 11/25/03 (1)
        
        In Reply to: Re: ...Long and winding reply... posted by Dave Davenport on November 24, 2003 at 20:45:02:
        
        Hi Dave,
        Very nice album...am listening to it now as I type. Read your post this morning...and went looking for it in my collection ..I also have and love "Works" from Keith. The best recording of a piano I own is Nojima plays List by Reference Recordings..especially the 2nd Track.."La Campanella" it is really beatiful. I first heard it at an audio show...and knew I had to have it. Just like I knew I had to have your PP line stage...;-)
        Cheers,
        Bas
        
        Re: Keith Jarrett�s �The Kõln Concert.� - Dave Davenport 19:04:33 11/25/03 (0)
        
        In Reply to: Keith Jarrett�s �The Kõln Concert.� posted by Bas Horneman on November 25, 2003 at 14:54:15:
        
        Hi Bas,
        I have quite a few of Keith Jarrett's recordings, but haven't heard "Works." I'll have to get it. I have also head good things about Nojima Plays List, I guess I'll have to get that too.
        I had used the K�ln Concert on vinyl for years as one of my reference recordings and ran out to get it when it was released on CD. Well I was disappointed - it sounded terrible. Hard, ringing, aggressive sound. Then as I improved my system the CD sounded better and better with each improvement. Now it is excellent. That CD is merciless for showing up the slightest defect. There are some changes that I have made that that is the only CD I can hear the difference on. I suggest that you keep it handy as you go through the changes to your power supply and see if it is as good tool for you as for me.
        Dave
- "Ditching the parallel-feed topology"...interesting - JLH 04:10:36 11/13/03 (0)
  Gary,
  Can you enlighten us on why you are choosing to drop the parafeed topology? I'm interested in the thought process that is going on here. What about the current balance in the line output transformer? It just seems weird to hear one of you guys move somewhat away from the "perfect diff balance with zero distortion" state of mind. I can only guess you will be using something like Allen's current balancer on the cathodes?
  Rgs, JLH

This post is made possible by the generous support of people like you and our sponsors:

You can not post to an archived thread.