[fsimms]

Frank Van Alstine suggested a conversation be started on what makes a good preamp.   I thought I would try to get it started.

I will ask all of you a simple question - - - what is a preamplifier supposed to do?  What is its basic job in the system other than providing audio level adjustment and source switching?

If you don't know the answers, makes it kind of hard to design a good product, doesn't it.  "Sounds really good" is not an objective answer.

Regards,

Frank Van Alstine

P.S.  We think we have some of the basic answers to my preamp question above and I will be happy to explain in detail herein in general terms (if anyone wants to hear my ideas), not just specific brand or model verbage.

The amplification portion of the ideal practical preamp should have low distortion.   It especially needs to have low odd harmonic distortion.   It should be linear and not have any automatic gain control action.  That is, low levels should be low and high levels should remain high.  What are the precise levels of distortion that should be achieved, I don’t know.  That would need to be judged by educated listening.

There should be filters on the ground and the mains to reject RMF signals coming from the power cable to keep that stuff from getting into the preamp.   The power supply should be shielded from the rest of the preamp.

The grounds from input devices should be able to be isolated from the preamp’s grounds.   I am talking about the single ended returns and the differential shields.   The preamp should be able to float the preamp’s signal grounds so that the outputs can be referenced from the amp’s ground in case the amp doesn’t isolate its inputs.   All of this would be to prevent ground loops.

I don’t make any preamps, so I am sure the above list is very incomplete, but I think that these are some of the issues that should be addressed by a top flight preamp.  Many of these would increase the cost so a judgment of diminishing returns would be at issue so we could afford the product.

Bob

[TheChairGuy]

Dual mono - two separate pathways right down to power supplies - this is a building block of a great preamp.

Many good preamps don't have it (including those I own and have owned)

John

[Imperial]

I was thinking that a nice feature could be variable overall gain, so that one could match any poweramp.
A choice maybe of various types of stepped attenuators, could then have fewer steps, making for better quality there.
Input circuit that could handle a large signal without clipping would be nice as well.
Optional output transformers maybe, to drive a balanced amp with really low input Z.

A well thought out earthing/ground scheme/setup is someting that is important, as fsimms points out.

A feature that I have not seen on so many preamp is that one could have input boards. That is, there is room for many inputs.
But you would only have those you needed plugged in the amp at any given time. Less circuits, less noise, less price.

Say you have a CD, a radio, a TV, and that's it.. two vacant slots then on ye 'ol 5 input scheme then.

If a pre is made up a "bus terminal" with smaller boards, that can be swapped out, one could basically upgrade any part of it later.
It's all down to the design stage, and basic design choices.

I'm sure that if one use less solderred connections and rather sort of develop a way of "clamping" the boards in place.
"clamp" the input boards. "clamp" the powersupply. "clamp" the circuits to eachother. Clamp the outputs circuits to the output sockets.

A smart "spaceframe" or "internal chassis" with modular clamps, that all are made up the exact same way?


Imperial

[denjo]

The volume pot!

It controls the sonic purity and volume resolution by affecting the S/N ratio, left/right tracking and crosstalk. More resistors mean signal has to pass through them and is usually degraded by the quality of the resistor. I remember using a top-of-the-line TKD stepped attenuator but turning the volume pot usually resulted in cracks and pops. The TVC (transformer volume pot) made famous by John Chapman's Bent Audio passive preamplifiers largely eradicates the problems mentioned and retains much of the sonic purity of signals, giving the music a high level of transparency and resolution. That is why there is a lrage following of audiophiles who have gone the passive route.

You can have the best preamp design but a mediocre volume pot will prevent the preamp from sounding good.

Best Regards
Dennis

[Brown]

  Simple, it should be simple. High quality matched parts with a killer power supply using the best transformers available.  A wood shielded enclosure. No coupling caps, use output transformers. Oh and it should be reasonably priced.

[BradJudy]

Frank asked what an ideal pre-amp should do, not how it should be built.  Many of the responses are design decisions and not customer needs.  It seems to be a hobby of audio forum goers to armchair design, even when that isn't what is asked. 

As for my answer to Frank...

As mentioned, low distortion, low noise, linear to a wide bandwidth (wide enough that there's plenty to spare, but not so wide as to lead to potential interaction problems) and good channel separation.  A quality volume control that has good channel matching even at low levels.  Reliable on the order of 10 years+.  Some gain, although a lot probably isn't needed.  Source switching that's reliable and minimally adds to the above issues (noise, etc).  Relative immunity to power line noise/distortion. 

As for features, I don't care about things like phono inputs and tape loops, but I would like things like a home theater bypass and remote.  A second set of output jacks for a subwoofer might be nice, although not needed.  Balance controls are sometimes handy, but I rarely use tone controls. 

[Brown]

Frank asked what an ideal pre-amp should do, not how it should be built.  Many of the responses are design decisions and not customer needs.  It seems to be a hobby of audio forum goers to armchair design, even when that isn't what is asked. 

As for my answer to Frank...

As mentioned, low distortion, low noise, linear to a wide bandwidth (wide enough that there's plenty to spare, but not so wide as to lead to potential interaction problems) and good channel separation.  A quality volume control that has good channel matching even at low levels.  Reliable on the order of 10 years+.  Some gain, although a lot probably isn't needed.  Source switching that's reliable and minimally adds to the above issues (noise, etc).  Relative immunity to power line noise/distortion. 

As for features, I don't care about things like phono inputs and tape loops, but I would like things like a home theater bypass and remote.  A second set of output jacks for a subwoofer might be nice, although not needed.  Balance controls are sometimes handy, but I rarely use tone controls. 

OK then. Two channel stero with a mono switch. Tone controls and a invert polarity switch. A neutral device with 18db of gain. Three inputs two outputs. Tubed rectified separate power supplies. A selector and volume control that has the feel of of solidity. Simple and clean enclosure.

[richidoo]

A perfect preamp must buffer current, add voltage gain in some cases, switch sources, and attenuate while adding no distortion of any kind. It's a tougher job than most other components besides speakers, because the role of the preamp is very complex and offers the most opportunities to add distortion. 

The less distortion added, the the more power music has to move the emotions. Dynamic distortion, harmonic distortion, and noise must be minimized to the extreme that the preamp's design price allows.

The preamp's price is result of an assumption about how low distortion should be. A $1000 preamp will not likely be fed by $20k CD player or play into $100k speakers. So certain distortion levels are accepted because increasingly low distortion levels are exponentially more expensive.

But the resolution of the ear/brain is limited only by software, and the ear always wants less distortion. Driven by the emotional enjoyment of music listening the brain always demands less distortion from an ever more expensive system. When is the distortion low enough? When is your appetite for emotional satisfaction enough? Remember that physiological and psychological homeostasis allows the brain to adapt to any stimulus to prevent becoming a music junky. The tweek that made you cry one day, is "just normal" within a week. Your tolerance for musical ecstacy increases, and with it your desire for even less distortion to intensify music's effects. But thankfully this high is healthy (except for the budget) because good music heals body and mind.

I really like Steve Sammet's philosophy that the preamp ideally should be the only source of voltage gain in the system. Then the largest parts of the distortion potential in the electronics (gain, switching and attenuation) can be centralized and addressed most efficiently. His preamp circuit looks very different from other preamps in design and construction. I admire that Steve puts low distortion above all other concerns.

Euphonic distortion is often accepted or even added deliberately to cover up other distortions in the system of a less pleasant character. When system distortion is reduced instead of covered up with additional, more pleasing distortion then the music is given more power to move the emotions, and is therefore more pleasurable than any kind of euphonic distortion. 
Rich

[Frihed91]

One that is invisible.

[Tyson]

A perfect preamp must buffer current, add voltage gain in some cases, switch sources, and attenuate while adding no distortion of any kind. It's a tougher job than most other components besides speakers, because the role of the preamp is very complex and offers the most opportunities to add distortion. 

The less distortion added, the the more power music has to move the emotions. Dynamic distortion, harmonic distortion, and noise must be minimized to the extreme that the preamp's design price allows.

The preamp's price is result of an assumption about how low distortion should be. A $1000 preamp will not likely be fed by $20k CD player or play into $100k speakers. So certain distortion levels are accepted because increasingly low distortion levels are exponentially more expensive.

But the resolution of the ear/brain is limited only by software, and the ear always wants less distortion. Driven by the emotional enjoyment of music listening the brain always demands less distortion from an ever more expensive system. When is the distortion low enough? When is your appetite for emotional satisfaction enough? Remember that physiological and psychological homeostasis allows the brain to adapt to any stimulus to prevent becoming a music junky. The tweek that made you cry one day, is "just normal" within a week. Your tolerance for musical ecstacy increases, and with it your desire for even less distortion to intensify music's effects. But thankfully this high is healthy (except for the budget) because good music heals body and mind.

I really like Steve Sammet's philosophy that the preamp ideally should be the only source of voltage gain in the system. Then the largest parts of the distortion potential in the electronics (gain, switching and attenuation) can be centralized and addressed most efficiently. His preamp circuit looks very different from other preamps in design and construction. I admire that Steve puts low distortion above all other concerns.

Euphonic distortion is often accepted or even added deliberately to cover up other distortions in the system of a less pleasant character. When system distortion is reduced instead of covered up with additional, more pleasing distortion then the music is given more power to move the emotions, and is therefore more pleasurable than any kind of euphonic distortion. 
Rich

This reminds me of what D'Agostino said about eliminating the current to voltage stages in the CAST based Krell gear.  To whit (Quoted from Stereophile):  "I noticed that every time we did an I-to-V conversion, the converter added noise and grain and messed with detail, so I just said, 'Wouldn't it be wonderful if we didn't convert this at all and ran it as a pure current signal from input to output?' Once we thought of it—well, it would be wrong to say it was simple, but getting out of the voltage-gain mindset was the 'ah-ha!' moment, and the rest was just engineering. And engineering is what we do."

[geowak]

Here's a very non-technical, unprofessional opinion of the blue collar type who does not
have knowledge about building preamps- just listening to them.

1 No noise introduced, no matter what the volume is
2 No hum either, no matter what the volume level is
3 The connectivity you need
4 Simple design, simple to use
5 In my opinion- it does not sound warm, or have a sound of it's own, it simply
passes on the signal that you have selected to the amp

[andyr]

No coupling caps, use output transformers.

So coupling caps implies sonic degradation but transformers = zero degradation?     

Regards,

Andy