Volume, balance, oh my…

December 7, 2016
 by Paul McGowan

As part of our new affordable line of electronics, Stellar, we’re building what’s known as a “pre-DAC” though I am not big on that term.

A pre-DAC is a preamp and DAC combo. Analog preamp, digital DAC, one box. The term I prefer is Control DAC and that’s what the new Stellar piece is called.

The biggest challenge in building such a device is not the DAC. Sure, that takes a great deal of skill and technology of which we will certainly lavish tons upon it. But the real challenge in building a sonically wonderful analog/digital instrument turns out to be the preamp. It took Bascom King the better part of 50 years to come up with a preamp as good as the BHK Signature.

Preamps and power amplifiers are perhaps among the most challenging products to design when expectations are high (and when aren’t they?).

Looking deeper into specifics of preamp design challenges we find the volume control is number one. And worthy of discussion.

I thought perhaps we’d spend a few days talking about preamps and volume controls as the subject moves me to write about it.

Let’s start at the beginning (novel idea, that). A preamplifier consists of three essential elements:

  • Input selector
  • Volume/balance control
  • Gain stage

All three are vital elements in successful designs. Failure to pay attention to any of the three gives less than grand performance when sonics matter—and in our world, sonics always matter. A lot.

Tomorrow I will detail some interesting notions and challenges with the input selector section, then we’ll move on to the meat of the discussion, the volume control.

Subscribe to Paul's Posts

30 comments on “Volume, balance, oh my…”

  1. If designers are so good (taking into consideration parts being better now) why don’t they build their best component first?

    I don’t understand the need for a balance control. They add more complexity and distortion to the component. And if needed something is wrong elsewhere.

    1. I am not sure I fully comprehend this question, but if it is why designers don’t put their best foot forward first, then make scaled down lower cost versions second, that’s exactly what we’re doing.

      I agree with you on balance controls – unfortunately they are required by people and equipment without it won’t sell. That’s the voice of experience.

      The trick then. How do you make a balance control without adding anything? That’s what we do. Now, balance controls are free both sonically as well as cost wise.

      1. My question was more of a comment instead. People work for 30 years to make the design they should have made at the beginning. If designers are so smart they would build their last product first instead of practicing for so long. This goes for everything not just audio.

        1. Yes, but it’s the 30 years of work that teaches them how. It’s like asking why someone who’s studied piano for 30 years didn’t just start out playing that good. Doesn’t work that way, unfortunately. Skills and knowledge are earned over time.

      2. I’d say that a balance control is a really odd duck. It actually UNbalances the music. As one adjusts such a control further and further toward the right, for example, there’s less and less of the left signal emanated into the room. The violin section is being progressively removed until at full right it may all but disappear. Even before that extreme, it doesn’t take long for the musical contributions of the various performers to become out of whack with the studio and with the intentions of the artists, producers, and engineers.

        A real balance control would be more of a panning control. At full right a combined mono signal would emanate from the right speaker with no loss of the folks one the left. We could then actually adjust left-to-right positioning without mucking up the music…if that ever seems important enough to do at all.

        I’m amazed that people hesitate to buy audio gear without a fader type balance control.

  2. What are your plans or hopes for the DAC section in comparison with the Direct Stream? Same question with the preamp section and the BHK Signature preamp? And finally, do you have a sense of the price point on this new equipment?

    1. It is not in the same league as DirectStream or Junior, but it’ll be damned good. Based on the ESS line of Sabre DAC chips, Stellar’s DAC will be among the top contenders within its price range – which will be $1,500 for the entire piece. Analog preamp and DAC in one box.

  3. Volume control:
    A comparison of theory and sound experience (considering cost/effort compared to other design topics influencing sound) regarding very good potentiometers, single resistor designs and other solutions would be very interesting.

  4. I like preamps and I consider them the 2nd choice to make behind the selection of the loudspeakers.
    I remember when passive preamps seemed the rage and I was so excited to borrow one and plug it into my system.
    I was disappointed that the music just seemed to fall flat without any oomph; no size or strength or power.
    It seems that preamp gain is vital to music projection.

    1. This is a systems problem. Gain capability should not change sound so long as the output is set at the same level.

      One of the Connecticut Audio Society preamp shootouts was done using CD, modified Dynaco Stereo 70 and Quad ESL63. The winner was a transformer based passive preamp, and a Creek passive preamp was close. The high input impedance of the tube amp and the low output impedance of the SS source change the priority. The preference for the impedance shift of the transformer is a significant clue.

      Depending on source and power amp, a passive pre can exhibit spectral tilting. It could also be that your ears are “broken in” to your active preamp’s distortion characteristics or frequency response shaping. Human break-in is a primary reason for differing speaker preferences.

      1. I’ll go along with your analysis especially regarding that I may have found the preamp distortion characteristics appealing to my ear as it integrated with the system. BTW I loved that system setup which consisted of 2 Hartley 24 in. subs driven by a Hafler 500 amp, 4Quad ESL 57 Mid range speakers and 4 Decca Super Ribbon tweeters all driven by Dynaco 70’s raised up by a dozen cinder blocks. The pre amp was an Apt Holman and the output was actively crossed at 92 and 7000 Hz.
        I just loved that stereo. I still have that stereo in my basement as my secondary stereo but it just does not seem the same. Maybe I’ve changed.
        I also agree about the human break in factor and if I audition someone elses stereo I feel like I need to reprogram my brain for the new sound.

        1. WOW! That system could be the basis of a “Destination High End audio venue”. I have to ask – what is the physical arrangement of the speakers? The Hartley subs have the potential to be great or awful depending on room and placement, likewise the finicky ESLs and ribbons.

          Looked up tests on Apt preamp. Unless you have tone controls engaged, I am going to scratch the euphonic distortion theory. Back to the “impedance mis-match leading to spectral balance problem, possibly involving interconnects”.

          I would run the woofers up a bit higher assuming they were close enough. I generally try to match the midband to the human hearing bandpass, 400Hz-4KHz. This minimizes Doppler distortion in the critical midrange. It serves as my selection criterion for drivers, rather than picking drivers and then using crossovers to flatten the system.

          1. Thank you for liking that system. I actually stumbled into it and that was one long story. That system was put together a long time ago when I was still single and it did not cost me all that much. I also did not appreciate how wonderful the room it was in was.
            The Quads now are stacked by a custom made wooden stand with the ribbons positioned in the middle similar but I think nicer than the Levinson stack unit. They are in a basement room built for them many years ago. They had been side by side raised up on 3 cinderblocks each. No way that system would have been allowed in the living room of the home my wife and I moved into.
            Upstairs now, years later, I have put together a different system which interestingly enough uses another dozen cinder blocks to pick up the front 6 loudspeakers. It is larger and louder sounding than the HQD system but I think less sweet.

  5. I had an equivalent design problem and solved it by analog thinking.

    We were targeting the pharmaceutical industry with a new Carbon analyzer for ultra-pure water. Our marketing department worked for a few years getting the Pharmaceutical Manufacturing Association to write a new regulation for Carbon impurity limits in drug making because we had proven it was essential to maintaining sterility of process water in the semi-conductor business.

    The strict PMA regulations (written by their army of product liability lawyers) required all process measurement devices be calibrated in place. All prior art for Carbon analysis was either redundancy, or a room sized calibration station that could maintain a flow of continuously filtered ultra-pure water. Switching a valve upstream caused a Carbon upset, stopping the flow for a few seconds would cause it to rise, opening a port or injecting a calibration standard was out of the question.

    The compromise was switching a calibration resistor into the front end of the conductivity sensor. This was an AC circuit (any DC on a chemical cell causes plating, space charge and electrolysis problems) measuring 100nS (nanoSiemens) full scale, or 10 MegOhms. SO, my challenge was to make an input selector that was reliably accurate to .1% of 100nA AC. Normal gold over nickel contacts of mechanical connections have a minimum current rating of 10uA. This holds for connectors, switches and relays.

    This is when I found out that all good electrical connections require melting metal. You can weld, braze or solder but mere touch is not enough. Mechanical connections work because either friction or arcing melts a few asperities to create the connection, and you need sufficient energy to make and maintain this melding of the electron clouds. I searched catalogs for days, but nobody had beaten this apparent limit of surface physics.

    Finally, my non-linear brain made a connection. My new HP 34401A used a relay to switch between the front and back terminals, and this six digit DVM had 100mV, 100MOhm and 10mA scales, so the LSD was 10nA. I removed the cover and got the relay manufacturer and part number, but then could not find it at any US distributors. Fortunately our receptionist was born and raised on Honshu, so she called Corporate headquarters in Tokyo and followed the information chain to find that these relays were only sent to the US for Hewlett-Packard production, and were “protected inventory” so we couldn’t buy them in the US. She arranged a supply chain through our Japanese distributor, and the product was a success.

    The electronic surface that gamed the laws of physics is proprietary so it is still magic AFAIK – but it works!

  6. I never understood why there should be so much fuss over pre-amps, a system element only good if it has no impact on the sound. By using integrated amplifiers or the volume control in a PWD DAC, it took me 35 years as an audio owner to finally get a stand-alone analogue pre-amplifier, only because I needed to add a few more sources.
    The device comprises a pair of 64-step 1db attenuators and an input selector. It has 3xXLR, 3xRCA inputs and similarly 5 outputs. It is totally passive. It works a treat and cost $800. Just under half of that is the wholesale price of the attenuators.
    Seems to do the trick.

    1. In studio/stage recording, mic preamps are a big deal, and I think this applies to phono and stereo preamps as well. Gain staging is one of the reasons – it’s nice to be able to crank the gain at one point or another in the chain, giving you more of the character you want, and less of the subsequent sound and noise you don’t necessarily want to amplify.

  7. I would add that, having used PS Audio digital attenuation in the PWD DAC, and replaced it with a very similar but much cheaper DAC that also has excellent digital attenuation, and there are loads of other similar DAC/pre’s on the market, why not just do that at a more attractive price?

  8. I look forward to such a combination, because it eliminates some interconnects and that’s a huge advantage. Also, the internal interfaces are much better optimized than general purpose output buffers of a DAC. I wish you would include treble and bass controls. This is a “must have” feature for me and fortunately some high end manufacturers such as Meridian and Dan D’Agostino have included tone controls in their ultra expensive gear. Because so many recordings are a bit on the bright side, and lowering the treble a couple of DBs makes them much more listenable. Likewise, the bass in recordings is not always at the desired level. This can be due to the recording, the room, the amps, the speakers etc. I know tone controls are not considered “audiophile” but for me it’s absolutely essential. Another item on my wish list for such a combo is two buffered analog outputs in addition to the main outputs that are intended to the amps. One for a subwoofer, and one for possibly a headphone amp etc. Finally a “nice to have” feature is a digital output that represents the settings of the volume & tone controls. The purpose of this is to facilitate some signal processing intended for side and rear channels, or a headphone amp with a built-in DAC etc. I think I saw someone mention that a balance is not necessary; I can’t disagree more. So often I get recordings that are out of balance. So often amps are very slightly off. Even cables can skew the balance.

  9. You are saying that you are intending to build a low budget equipment in the $1500 range using off the shelf ICs etc. I wish you’d also consider building a high-end version of this using your best DAC circuitry and buffering etc. I always found that eliminating a preamp, even a very expensive one, has improved the sound. Provided of course that the DAC has level control or by using a passive device to attenuate.

  10. Good day Paul:

    One of the things I like to do is experiment. I used to have an Audio Research SP-3 pre-amp (passed away after too many experimental surgeries). I removed the stereo/reverse stereo/mono selector switch and soldered in a couple of short copper wires where the selector attached to the PC board and the sound improved. I did the same with the balance selector and the sound improved again.

    It would seem that selector switches and potentiometers degrade the analogue signal. I would prefer a pre-amp that has minimal amount of selectors and switches for better sound. I understand what you are saying about what the customer wants. It would be nice for manufacturers to off a special order for minimalists like me.

    With that in mind, does digital displays with touch screens degrade the sound signal?

    1. My Prima Luna pre has no balance control, Alps analog attenuator for volume, ( volume controls only reduce the preamp’s gain), reed switches on the input selector, point to point wiring in the signal path, vacuum tube rectification, and no ic’s in the signal path. -But no balanced in’s or out’s! I like a simple design, but miss a balanced design even though the sound is heavenly. -We always want more, don’t we?
      John W.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram