The best way to design an audio system.


What is the best way to design and assemble a high-end audio system?

Should you first adopt a system philosophy?

Or should you just pick out a component you really like and build a system around it?

These seem to be the 2 most popular methods of assembling systems that I have read here on the Audiogon forums. Of course, I have my own thoughts on this subject, but I am interested in how everyone else feels about this, and what are the reasons for their opinions. Considering that we have alot of new people on the Audiogon these days, it may be helpful for them to read about how the "old timers" and "experts" configure their systems and why.
twl
I think this is a very relavent and thoughtful question for A-gon. I have two systems that were built around different principles, but it's not exactly clear (even to me) if it was a philosophy or a component, because it evovled from one to the other. First is the reference system. This started from hearing a pair of Martin Logans and being extremely impressed with their transparency and imaging capabilities. I decided this was going to be the cornerstone of the reference system. So, already, is it the philosphy of the transparency and imaging or the component? I'm not really sure to tell you the truth--but I knew I wanted the sound that M-L could deliver. However, in building the system--as with most--there were weaknesses that were exposed as the system got better through upgrading components. The most problematic weakness was the bass. I had listened to Genesis 200s and wanted to recreate bass at that magnitude and realism. Now that is virtually impossible to go from a 12 inch woofer and try to compete with 16 servo driven woofers--but always set the bar high and see how close you can get. The first step was to actively bi-amp the system. This helped quite a lot, but still was soft in the bass. Even the Krell KMA-160 on bass didn't quite give the dynamics I was trying to acheive. At this point we have departed from a component in trying to acheive a sonic goal. The next step was to replace the bass drivers to Focal Audiom drivers. We reinforced the bass cabinet and changed the damping material. I did this with the help of some engineers to figure out how well it would be mated with the panel. It turned out extremely well and meshed better with the panel than the original woofer. At the same time I was able to get realistic dynamics. Was it up to par with the Genesis 200s--no, but it would match or surpass virtually any other 12 inch driver I had found. I could go on about the evolution of this sytem through the source components--but I don't think that answers your question any better than this one specific example.
The other system was a family room system, that originally started with the concept of a low cost monitor system. I purchased Epos EP-12 and a solid state amplifier--which I would up hating. I replaced the amp with Cary tubes to give the system some warmth and character. Not trying to design a high resolution system at this point, just something that would be enjoyable to listen to. I later replaced the Epos with Red Rose R-3s. These really sound great with tubes and now the system evolved into a high resolution system that also has incredible bass for such small monitors. With these monitors it was now time to increase the resolution by going to NOS tubes. The system as you can well imagine, which started with one goal in mind has changed dramatically (but it's still enjoyable to listen to--so that goal didn't change).
You have to calculate the room dimensions, size and distance of actual listening area to be used and know the frequency and SPL range that you want to reproduce with authority. Pick speakers that are suitable for those specifics and don't over or under do the project at hand. This means looking for something that is reasonably efficient, has the appropriate radiation pattern and does not have a wild impedance curve. The more that you compromise those basic requirements, the more complicated your job finding suitable support components will be.

No matter what you do, if you have speakers that are inappropriate for the room or your specific requirements, everything else that you do will strictly be an attempt at a "band-aid". You'll spend all of your time trying to correct your initial mistake rather than enjoying what you already have.

If you've done your homework with the speakers, this will narrow down the field of amps quite a bit, so that shouldn't be a problem. As such, you should go back to the source components and find something that is both user friendly and offers the sonics / features that you prefer. When it comes to digital, I think that a lot of people never hear how good it can really be as they don't spend the time to really set things up as good as possible or experience several different front ends within their system. They settle for less because they've never been exposed to "good" digital. Remember, the rest of the system can't reproduce what the source has never recovered. Sources are PHENOMENALLY important.

From there, move onto the "backbone" components, i.e. preamp and power amp or integrated. With an excellent source and a set of speakers that WILL work in your room for your specific situation, your goal is to find a preamp that offers the versatility that you need while doing as little to the signal ( in or out of band ) as is possible other than to regulate the amplitude. A "straight wire with gain" if at all possible. A power amp needs to take this signal and amplify it to the levels required to efficiently respond to your speakers needs WITHOUT being "pushed". This means the amp has to have the ability to deal with a wide range of impedances, levels of reactance, output levels, etc...

If you've done all of that, the rest is a matter of fine tuning the "voice" of the system. This can be done with various cables ( power, interconnects, speaker ). Don't forget that a rack or whatever it is that you intend to support the components with CAN and DOES affect what you hear. Do some research in the archives here as to what works well. A basic suggestion would be to use a very rigid support structure with lightweight shelves that are free-floating.

Treating the room acoustics to varying degrees will also tend to highlight minute differences that you might not otherwise have taken notice of. This is due to frequency response and / or timing errors. These differences would have otherwise covered up the subtleties that good systems make obvious but do so in a fashion that seems completely natural.

Once you've gotten to this level, you are well beyond addicted and you'll soon be giving us advice about what to do and the tweaks that we need to try : ) Sean
>
Before I began putting together a high-end system I ran pro sound for a rock band & got a good education in the basics. A lot of them carry over to home audio & the biggest culprit is AC. It's amazing how much noise one dimmer switch can put into a system. Then there's phasing, everything being correctly grounded, shielded cables only going where they belong & so on.

As for my system, what I've tried to do is recreate that "live" sound as best I can & have gone through a few different set-ups to get what I'm after. You'll never know how a piece will work in your system until you hear it there. You can get a pretty good idea from hearing it in another application but it's not the acid test.

Then there's synergy. You might fall in love with a certain component but it just won't work in your system. You can either build your system around that one component or try something else. Or build a second or third system around that one component.

If you're happy with the sound of your system then when you want to upgrade or change something you can evaluate that one change & make a decision regarding that one change. If you're not happy with the sound then you have to evaluate what you don't like about it & go after that particular thing until you get satisfactory results.

For me its been a combination of hearing a piece I liked (i.e. speakers) & going after that sound and sometimes its just been experimenting in a certain area (i.e. SS amps) because I wasn't quite happy with what I had. The rest of the time it was trying many different items (i.e. cables) until I settled on one I was happy with.

I suppose it's a combination of some sort of system philosophy & favored component(s) although I would throw in trail & error as one of the main ingredients.
I don’t have anything that rises to a philosophy. I’m more of a member of the “Church of what’s happening now.” I think this is because everything in audio is such a compromise. The idea of going after “The Absolute Sound” seems ridicules (good advertising pitch though). If you just look at what we use for speakers (cones, ESL, etc) they are very imperfect. You can’t have it all. I like what I have and then I listen to what someone else has put together and it gets something out of the music that I haven’t heard for awhile and it sounds good and I tweak. I don’t think we audiophiles are monogamous. It’s like your favorite food, you eat it every day and it gets a little stale.

Personally, I especially like dynamics and transient response lately. Actually for some time now. I tweak for that. This has led me to efficient speakers. Horns and single drivers, light cones without much excursion etc… I’ve been on a little DIY tube binge too for a short while. I’d like to come up with all sorts of philosophical reasons for liking SETS but in the end it’s just that they sound pretty good and are simple enough that even I can half understand it.

There is something about the first watt of power theory that seems interesting. I mean if you have a 90db and 100db sensitivity speaker what kind of power are you using?

90dbW/m speaker
Driven with 0.5 watts = 87dB
Driven with 0.25 watts = 84 dB
Driven with 0.125 watts = 81 dB
Driven with 0.0625 watts = 78 dB
Driven with 0.03125 watts = 75dB
Driven with 0.015625 watts = 72 dB
Driven with 0.0078125 watts = 69 dB (1/128th of a watt)
Driven with 0.003986 watts = 66 dB
Driven with 0.001953 watts = 63dB
Driven with 0.000976 watts = 60dB (about 1/1000th of a watt)

100dbw/m speakers
Driven with 0.5 watts = 97dB
Driven with 0.25 watts = 94 dB
Driven with 0.125 watts = 91 dB
Driven with 0.0625 watts = 88 dB
Driven with 0.03125 watts = 85dB
Driven with 0.015625 watts = 82 dB
Driven with 0.0078125 watts = 79 dB (1/128th of a watt)
Driven with 0.00390625 watts = 76 dB
Driven with 0.001953125 watts = 73dB
Driven with 0.000976563 watts = 70dB (about 1/1000th of a watt)

A medium room and normal levels (85db?), of course you’re sitting a few feet away, but you're only using .03125 to .0625% of one watt. About 1/4th of a watt w/ 90db speakers

No wonder TWL is running that little single watt w/ Lowthers if it performs good down there. I remember reading a piece by Kurt Strain where he would optimize his amp circuits for different 1/10ths of the first watt depending on the speaker load and sensitivity! How many systems are optimized for that?

Oh, I almost forgot, I keep everything on bubblewrap!

PS: If anybody is interested in a list of DIY tube electronics related materials available on-line drop me an email. A couple of us are trying to put together intro/beginners material that, if we get our act together, will be posted over at VALVE magazine and we need some honest feedback.

I remain,