A paper I wrote 11 years ago.
66.61.56.29 |
||
| Posted on July 18, 2016 at 20:03:35 | ||
|
Posts: 6936
Location: America's Heartland Joined: May 27, 2001 |
I thought this might be of some interest to you guys. I came across it in my files and decided to post it. It has been edited for tense, spelling, etc.; but nothing technical was changed except to increase meaning or clarity. I'd be interested in your thoughts. If I were writing this today, I might put imaging first. I have also developed some personal preferences since then; but nothing that violates my 2005 thoughts. Things that really make a difference in a Stereo System, and Those That Don't. 14 August, 2005 From years of experience and experiments, theory, and more recent confirming experiments and computations the following list is a compilation of the things to really worry about. Everthing else is tied for last. 1. Frequency response of the overall system in your room is the single most important parameter. Of the things that affect the overall, the room and the speakers, and their interaction, is the most important by far. In frequency response we are not just talking about 20 Hz to 20 kHz +/- 3 dB. Rather we are talking about the overall trends (rising, falling, flat) peaks and dips, their magnitude and location, and the character of the bass relative to the rest. All of this referenced to your head location in your favorite listening spot in the room. 2. Imaging is really important, maybe as important as frequency response. This includes lateral spread, depth, precision of positioning, stability, width of the sweet spot, and even height. If the image can locate the instruments or voices accurately, then masking effects will be different than with monaural or poorly imaging systems. This is a good part of the elusive "definition". Separate the instruments and voices in space, and you hear them individually. But crowd everything to the center, and they mask each other and definition and clarity suffers. 3. Room placement of the speakers relative to each other and to the room boundaries and to the listening position. Includes room boundary effects, mutual coupling, and room modes, all as seen from the listening position. Sometimes you might have to move to find a better spot. Also includes issues such as being in the near or far field, ratio of direct to reverberent sound, the speaker's radiation pattern, speaker height relative to the floor, and to the listener's ear level. 4. Room treatments including overall room absorbtion, bass traps, diffusion, and absorbers placed at reflection points to kill early arrival reflections. Furniture, book shelves, carpeting, wall hangings can all serve to help the room acoustics. Even a well placed ceiling fan, if it's running, can help with diffusion. 5. Speaker radiation patterns. Dispersion in horizontal and vertical planes, launch angles and lobing at crossover, dispersion angle changes at crossovers, sidewall and floor and ceiling reflections, ratio of direct to reflected sound. 6. Room symmetry with regard to reflections, the listening location, and equal cable lengths does seem to be important to imaging. On a mono source, the two speakers should sound alike from your listening position. 7. Dual mono amplifiers seem to help with clarity of sound on loud passages and also with imaging. Power supply isolation and regulation for preamps, tuners, and power amps is important if they are in the same chassis. A primary argument for separates, but not necessary IF the power supplies are well done in integrated amps and receivers. Not many are! 8. System power ratings. A comprehensive look at speaker power handling and sensitivity versus personal loudness tastes, room size and absorbtion, and the amplifier's continuous and dynamic power. In the room of choice, at the levels you prefer, on the music you like, is there sufficient speaker sensitivity and power handling to cope safely with the amplifier needed to achieve your goals for level? If the levels are high, then one needs to consider compression effects as well as thermal and mechanical considerations. Amplifier continuous power relates to speaker power handling, but dynamic headroom relates to peak levels obtainable on transients. 9. The primary interaction between components in the system is the amplifier/speaker/speaker cable interaction. The sum of wire and amplifier impedance interacts with speaker impedance to form a voltage dependent voltage divider and shifts speaker frequency response. (John Atkinson in Stereophile refers to this as the Ohm's Law effect) Parameters to be concerned with are speaker impedance curve, amplifier output impedance Vs frequency, and cable DCR and impedance, and whether the speaker was voiced for solid state or tube amps. Also to be considered is the amplifier's ability to handle low impedance loads and supply their current demands versus the speaker's minimum and average impedance. The amplifier's ability to tolerate capacitive loading is also a factor in speaker and cable choices. The 5% rule of thumb holds for solid state amplifiers and typical SS voiced speakers, but not for tubed amps or transformer coupled SS amps. 10. Component input and output impedance matchups are not often a problem, but do crop up and need to be watched for. Basically, the receiving component's (like a preamp input) input impedance should be at least 10-20 times the output impedance of the source component (like a CD player output). Long, or very high capacitance interconnects may contribute as they are seen as part of the input impedance of the receiving component. If the output impedance is not known, a test may be necessary to determine load sensitivity. Input impedances should be high, around 50,000 ohms, and primarily resistive with low capacitance. Many do not meet this criteria, and it needs to be watched for and considered. 11. Patch cords (AKA, interconnects) should be well made, with good strain relief at the connectors, with gold plated plugs, well shielded, and low in capacitance. Some components are unusually sensitive to capacitance on their output. This is evidence of a poorly designed output stage. Longer and higher capacity cables make them edgy and bright. For these few use very short cables of low capacitance construction, or get a new piece of gear that's well designed. 12. Clean, oxide free, and tight connections are critically important at all points, but especially the speaker cables. The wonderful sound of new cables may just be the clean new contacts. 13. If there is a phono system, then the capacitive and resistive loading of the cartridge by the interconnecting cables and the input circuitry of the phono amp are critically important. One must also worry about placement of the turntable to avoid floor vibrations and acoustic excitation of record and arm resonances. The cartridge/arm resonance due to interaction of the cartridge stylus compliance with the cartridge/arm mass must be in the 5-10 Hz range to avoid structural vibrations from footfalls, traffic, etc. and from acoustic vibration from the speakers. This range makes isolation easier and allows the cartridge to ride record warps with minimum output. A good table suspension is important. The record must be "nested" on a dead and vibration damping mat or vacuum chucked. A closed cabinet or closet is a good idea to reduce acoustic excitation of the record. Because of the very low output voltage of phono cartridges, gold plated connectors are a must have. If the phono preamp or transformer is an outboard device, then you must worry about its interface to the preamp input. 14. If you have a tuner, than an antenna is critical. A cheap tuner with a good antenna will outperform a great tuner with a crappy antenna. Although the better tuner may have a better sound quality. What Doesn't Really Matter. 1. Interconnects of exotic materials or construction. Radio Shack Gold at $10.95 for a 1 meter pair, are well made, have gold plated connectors, and moderately low capcitance. They are all you need. Even John Dunlavy thought so. 2. Isolation is important for turntables, and possibly for tube gear due to microphonics. But is unnecessary for solid state amplifiers, preamps, tuners, cassette decks, and CD players unless they are in an unusually violent location. If the floor is very flexible, consider wall shelves or an isolated rack. Cones and spikes are not isolators. Springs and air cushions are isolators. Stacking components is no problem provided cooling is not impeded. Hockey pucks make great 1" spacers at $1.00 each. 3. Extra heavy component cases are a waste of money. An NAD C350 works just fine with a 20 gauge case and plastic front panel. (I have to admit, I like the look and feel of heavily built gear, but that's psychological, not sound quality) 4. Power cords do not need to be special. Even at 1200 watts, they are only handling 10 amps. If they get warm in use they are too small. If the plug gets warm in use, the contact is poor. Bend the prongs or install a new wall outlet. Unless you live in NYC or next to a factory, power conditioners are not needed. A surge protector, however, is a great idea. So is a system ground back to the surge protector, not to the wall. 5. Speakers "break in" after a few hours of use. Other components don't. Things do warm up, and tubes age rapidly at first and need frequent re-bias and/or rebalance. But even tubes will settle down after the first 30 hours or so. Warmup should take no more than 30 minutes. My SS stuff settles in after 15-20 minutes. Give your new stuff a week, if it doesn't sound good then, it never will. Do not wait until your return privilege expires. The dealer will try to tell you to wait longer to let it "break in". 6. Exotic speaker cables. 12 gauge should work for any speaker of 4 ohms or more if runs are under 25'. If runs are longer you do need to worry about DCR and inductance. But even so, you only need inexpensive cables. The DIY Cat 5 cables and the Kimber braided cables make sense for long runs and their design reduces cable inductance and the impedance at higher frequencies. But their capacitance is much higher and may cause amplifier interactions.
|
|
| "Making sense", posted on July 19, 2016 at 14:56:03 | |
|
Posts: 37673
Joined: May 12, 2000 Contributor Since: April 5, 2002 |
Such requires a complete understanding of all the relevant variables. Many engineers begin with the wrong assumptions and/or believe they are omniscient. Some, however, are open to understanding what their senses tell them. :) One engineer's view on the "science" of power cords
|
| RE: "Making sense", posted on July 20, 2016 at 12:28:50 | |
|
Posts: 11287
Location: Arizona Joined: February 4, 2003 |
What about copper vs silver? In relation to internal wiring, AC lines and speaker wires?
|
| RE: I'm with Jerry on this one, posted on July 20, 2016 at 13:29:42 | |
|
Posts: 11287
Location: Arizona Joined: February 4, 2003 |
Mogambi sp?? and Kimber
|
| I agree -- you got it all right. (nt), posted on July 20, 2016 at 15:48:40 | |
|
Posts: 1674
Location: Connecticut Joined: April 16, 2002 |
nt
|
| RE: "Making sense", posted on July 21, 2016 at 06:21:04 | |
|
Posts: 37673
Joined: May 12, 2000 Contributor Since: April 5, 2002 |
You can make your own that work just as well.I know because we've done it. Yes and no. If you view my profile, you'll find that I use a mix of DIY and commercial power cords. The better DIY cords are based upon Belden 83803. They're better than stock but don't touch the Harmonic Tech ones used with the power amps. Tell us about yours and to what you've compared it. Who's we? :) As for Steve's article, there is precious little there I find that agrees with my experience. Who uses "stranded rubber" cords? Shielding bad? Who cares about RFI anyway, right? He is, however, welcome to his opinion. Mine follows that of companies like ARC (Sain Line) and VTL (Shunyata) who use shielded aftermarket products in their own evaluation systems.
|
| Perspective, posted on July 21, 2016 at 08:28:25 | |
|
Posts: 37673
Joined: May 12, 2000 Contributor Since: April 5, 2002 |
I started making my own interconnects using Belden stock back in the late 70s. The Acoustat X had tube amps in the bases and required long runs which were difficult to find pre-made at the time. Initially, I used DIY Belden, then Audio Technica and eventually Esoteric Audio which used better connectors. I still have a short run of those today. The office based in-house system poses a similar challenge where I use Belden 1505F and terminated the leads myself. You really can get decent cable for a modest amount. One of many areas where I agree with Jerry is the concept of "non-bright". I've found that all matter of untrapped RFI causes a false brightness that masks resolution. In order to really get noticeable improvements in that regard, I've found that you must step up the investment a bit. Kimber certainly makes such along with JPS Labs, DH Labs, Nordost, etc. Be prepared to spend about $400 for even 1-2M runs. Initially, they might sound "dark", but I've found that translates to "dead quiet". It is then when you start hearing more nuance. On the other hand, I am primarily a speaker guy and have always centered a system around a great speaker. With your system, I would consider a change there if anywhere. The Zaphs appear to be well designed mini-monitors with a low crossover for better driver blending. As for me, however, I find tall line sources provide a more realistic presentation. Have you ever heard Maggies in your system? They offer a money-back guarantee on their entry level model. The only downside is they do need a couple of feet breathing room behind them. You can minimize that requirement in a small room using more toe-in. I've found that everything in a system can make a difference. After a while, one noticeable difference here and one there starts to add up. I would put cables towards the end of the priority list, but continue to find that for the highest performance systems, they can make meaningful differences. The final systems of HP I heard at Sea Cliff used Nordost Odin throughout. I can't say that all of the phenomenal transparency - and I don't use that adjective lightly - was due to the cabling, but he was convinced they certainly contributed.
|
| RE: A paper I wrote 11 years ago., posted on July 21, 2016 at 08:28:26 | |
|
Posts: 14371
Location: East Coast Joined: October 4, 2001 |
Interesting! What cartridge? Dave
|
| BTW..., posted on July 21, 2016 at 09:49:56 | |
|
Posts: 37673
Joined: May 12, 2000 Contributor Since: April 5, 2002 |
Did you end up trying out Chris' V twist ICs? I had to smile about the conversation between Duster and you about shielding. Did you notice that the premier VH Labs cable is shielded? :) "...has a double shield of foil and silver plated copper spaced AWAY from the signal conductors." One reason why my Harmonic Tech power cords are pricey is that like the above, they use silver conductors for hot and neutral.
|
