|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
64.13.14.96
In Reply to: RE: Measuring properties or changing operation? posted by bjh on June 24, 2009 at 11:40:54
Um... If I may butt in and dispense with the double negatives I would paraphrase what Tom said thusly: 'If high-end audio equipment was really well designed, cables and environment wouldn't matter'. Since they do, it isn't. You guys are barking right up my tree so I'd like to hop up on my soap box and throw in my two cents worth.
While I largely agree with him, cables themselves are actually the source of some problems, especially triboelectric noise. Anybody that's stepped on a cheap microphone cable has heard that one...
It isn't just the competence, or lack thereof of the designers. Home audio is one of those pathological cases which superficially seems trivial, and yet isn't. While almost anything thrown together by almost anyone will work, designing a product that really works well, i.e. to the point that it is effectively transparent to the ear, is another matter. Here are a few reasons that come to mind..
1. The lack of systems engineering.
-The home audio system topology wasn't designed, it just happened.
2. The lack of meaningful standards.
-To put it bluntly, it's not important enough to governments or industry.
3. Unusually wide bandwidth.
- Three to four decades of signal bandwidth is huge.
4. In-band power.
- Having the 60Hz power supply and it's main harmonics where we hear is bad.
5. Extreme user sensitivity.
- Reducing distortion to where you really CAN'T be aware of it is tough.
6. Conflicting design approaches.
- Nowadays this is less of a problem due to switchers and mixed-signal designs, but traditionally audio guys were clueless about RF and the major rules of thumb are in conflict. It's hard to both star ground and maintain a Faraday shield.
Enough of that. It isn't trivial, there is no agreement on how good is good enough, what factors are important and how they are to be measured and weighted, or what the worst-case environment is that must be accommodated. The latter includes vibration, equipment interface, power supply, EMC and operators both deliberate and inadvertent. I'm trying to think if I know of anybody whose toddler HASN'T put a PB&J sandwich into the VCR.
But... I'm with Tom. The obvious goal is to design gear where all of these variables become, as much as possible, 'don't-cares'.
Rick
"It isn't just the competence, or lack thereof of the designers. Home audio is one of those pathological cases which superficially seems trivial, and yet isn't. "I don't think it's just home audio..... It's **all** audio....... If you think the stuff designed for the studio is of more competence, I'd kindly but strongly disagree. (I personally think Pro Tools is the single biggest monstrosity of a design ever conceived.) Just look at the horrible recordings that have been put out in recent time.
"While almost anything thrown together by almost anyone will work, designing a product that really works well, i.e. to the point that it is effectively transparent to the ear, is another matter. Here are a few reasons that come to mind..
1. The lack of systems engineering.
-The home audio system topology wasn't designed, it just happened."Systems engineering would force a home audio system to be of a "all in one" type design..... Unless every designer agrees upon what the system ought to be, this "lack" will go on indefinitely......
I personally don't think this facet of engineering is any more of a problem than other facets of engineering.
"2. The lack of meaningful standards.
-To put it bluntly, it's not important enough to governments or industry."It shouldn't be important to government, unless public safety is an issue...... (This is why we have UL standards, by the way.)
I do agree in regard to industry...... But it would require an across-the-board agreement amongst many backgrounds, so even if such objective is sought, the chances for successful implementation would be very slim.
"3. Unusually wide bandwidth.
- Three to four decades of signal bandwidth is huge."Not since digital audio has hit the scene. Bandwidth is limited to 22.05 kHz. And because of that, bandwidth in line and power amplification is no longer so important. So I strongly disagree here.
I also disagree in regard to bandwidth being a problem, unless it's so wide, it becomes susceptible to RF interference.
"4. In-band power.
- Having the 60Hz power supply and it's main harmonics where we hear is bad."The only solution to this is battery power..... Jeff Rowland, a so-called "home audio" manufacturer, has undertaken such approach.
"5. Extreme user sensitivity.
- Reducing distortion to where you really CAN'T be aware of it is tough."The user sensitivity has always been "extreme." The problem is too many in the design community are trying to convince the masses that associated discomforts should be tolerated to a degree or that the sensitivity is overrated. This is a big reason why audio design has been disappointing, in my humble opinion.
"6. Conflicting design approaches.
- Nowadays this is less of a problem due to switchers and mixed-signal designs, but traditionally audio guys were clueless about RF and the major rules of thumb are in conflict. It's hard to both star ground and maintain a Faraday shield."It's not difficult to make a star ground and Faraday shield...... It's that the latter has been deemed unimportant. Whether it is or not is subject to debate, but I will say the internal RFI generated from digital devices has rendered the Faraday shield ineffective. Although it should still be used on purely analog components like phono stages.
In regard to "clueless about RF", this is a problem for most designers across the board, not just home audio. The RFI generated by digital devices has changed everything.
Conflicting designs in itself is not an issue. The might both be competent, they might both be incompetent, and anything in between.
"Enough of that. It isn't trivial, there is no agreement on how good is good enough,"
Now here you hit on an issue. When in doubt, raise the bar, not go the easy route and think a lesser design might still fly. This IMO is a big problem in modern audio design. The bar simply hasn't been set high.
"what factors are important and how they are to be measured and weighted, or what the worst-case environment is that must be accommodated. The latter includes vibration, equipment interface, power supply, EMC and operators both deliberate and inadvertent. I'm trying to think if I know of anybody whose toddler HASN'T put a PB&J sandwich into the VCR."
Or oatmeal.
"But... I'm with Tom. The obvious goal is to design gear where all of these variables become, as much as possible, 'don't-cares'."
If the bar is set low, the variables do become "don't cares"..... The problem is there will be someone who will think the design is inadequate.
If the bar is set high, the variables becoming "don't cares" is next to impossible, in my humble opinion. This is why I think listener feedback is paramount to a good audio design. Until we can attain a more-complete picture from a measurements and specs standpoint, the best yardstick for performance will indefinitely be the collective satisfaction of the end users.
Edits: 06/25/09
"It isn't just the competence, or lack thereof of the designers. Home audio is one of those pathological cases which superficially seems trivial, and yet isn't. "
I don't think it's just home audio..... It's **all** audio....... If you think the stuff designed for the studio is of more competence, I'd kindly but strongly disagree. (I personally think Pro Tools is the single biggest monstrosity of a design ever conceived.) Just look at the horrible recordings that have been put out in recent time.
> Well, I don't have much exposure to that field but I think it's better, at least they are using balanced cables! The main reason I usually say 'home audio' is that that's what I'm familiar with.
"While almost anything thrown together by almost anyone will work, designing a product that really works well, i.e. to the point that it is effectively transparent to the ear, is another matter. Here are a few reasons that come to mind..
1. The lack of systems engineering.
-The home audio system topology wasn't designed, it just happened."
Systems engineering would force a home audio system to be of a "all in one" type design..... Unless every designer agrees upon what the system ought to be, this "lack" will go on indefinitely......
> Nah, I don't think so. The main thing that's wrong is the interfaces are poorly specified especially with respect to impedance and susceptibility.
I personally don't think this facet of engineering is any more of a problem than other facets of engineering.
> Boy I do, look at the stuff people hang onto their systems trying to get them to play well together.
"2. The lack of meaningful standards.
-To put it bluntly, it's not important enough to governments or industry."
It shouldn't be important to government, unless public safety is an issue...... (This is why we have UL standards, by the way.)
> It depends upon your view of government. Ours does have some such as FCC part 15, Europe has more. I've come to view some of the CE requirements as a good thing. It makes my job harder but really does result in a sturdier design.
I do agree in regard to industry...... But it would require an across-the-board agreement amongst many backgrounds, so even if such objective is sought, the chances for successful implementation would be very slim.
"3. Unusually wide bandwidth.
- Three to four decades of signal bandwidth is huge."
Not since digital audio has hit the scene. Bandwidth is limited to 22.05 kHz. And because of that, bandwidth in line and power amplification is no longer so important. So I strongly disagree here.
> On the outside that limitation reduces the bandwidth by less than an octave.
I also disagree in regard to bandwidth being a problem, unless it's so wide, it becomes susceptible to RF interference.
> The wide bandwidth makes the signal more susceptible to problems, especially due to passives such as skin effect and core saturation.
"4. In-band power.
- Having the 60Hz power supply and it's main harmonics where we hear is bad."
The only solution to this is battery power..... Jeff Rowland, a so-called "home audio" manufacturer, has undertaken such approach.
"5. Extreme user sensitivity.
- Reducing distortion to where you really CAN'T be aware of it is tough."
The user sensitivity has always been "extreme." The problem is too many in the design community are trying to convince the masses that associated discomforts should be tolerated to a degree or that the sensitivity is overrated. This is a big reason why audio design has been disappointing, in my humble opinion.
> Agreed.
"6. Conflicting design approaches.
- Nowadays this is less of a problem due to switchers and mixed-signal designs, but traditionally audio guys were clueless about RF and the major rules of thumb are in conflict. It's hard to both star ground and maintain a Faraday shield."
It's not difficult to make a star ground and Faraday shield...... It's that the latter has been deemed unimportant. Whether it is or not is subject to debate, but I will say the internal RFI generated from digital devices has rendered the Faraday shield ineffective. Although it should still be used on purely analog components like phono stages.
> I think it is. Basically you want the shield to ground to the box for RF but to the board for audio so if you do a good job you end up with a little of each but don't get the best of either
In regard to "clueless about RF", this is a problem for most designers across the board, not just home audio. The RFI generated by digital devices has changed everything.
Conflicting designs in itself is not an issue. The might both be competent, they might both be incompetent, and anything in between.
"Enough of that. It isn't trivial, there is no agreement on how good is good enough,"
Now here you hit on an issue. When in doubt, raise the bar, not go the easy route and think a lesser design might still fly. This IMO is a big problem in modern audio design. The bar simply hasn't been set high.
"what factors are important and how they are to be measured and weighted, or what the worst-case environment is that must be accommodated. The latter includes vibration, equipment interface, power supply, EMC and operators both deliberate and inadvertent. I'm trying to think if I know of anybody whose toddler HASN'T put a PB&J sandwich into the VCR."
Or oatmeal.
"But... I'm with Tom. The obvious goal is to design gear where all of these variables become, as much as possible, 'don't-cares'."
If the bar is set low, the variables do become "don't cares"..... The problem is there will be someone who will think the design is inadequate.
If the bar is set high, the variables becoming "don't cares" is next to impossible, in my humble opinion. This is why I think listener feedback is paramount to a good audio design. Until we can attain a more-complete picture from a measurements and specs standpoint, the best yardstick for performance will indefinitely be the collective satisfaction of the end users.
> Well, it just takes too long to consider every possible aspect of every design. At some point you have to develop a good feel for what you can get by with. Sometimes you are wrong which adds to your knowledge of what's important.
I don't know if this format will work Todd, I'm using a WP that I'm not familiar with so I apologize in advance if this is confusing.
Still using CDex? I'm really pleased that you tried it and that it worked for you, it does for me. Now there's something. You are a SW guy aren't you? Why does it sound so good?
Rick
Recently I witnessed an improvement in sound with usage of an aftermarket power cord on a SS power amplifier (replacing yet another aftermarket power cord) that rather shocked me.The main reason for the shock was largely due to the fact that the effect ran counter to my experience, that power cords seem to make the least impact on SS power amplifiers vs. source/preamp/low power gear; regardless of how counterintuitive that may seem.
The second reason for the shock was due to the particular SS amplifier, a Bryston 4BST, an amplifier which along with its SST sibling has long ranked in my most hated list. The ills associated with the 4BST/SST, what I would call electronic sheen (a subjective description) seemed to me to have vanished!
Now do we have an engineer present who would like to step up to the podium and propose that the power supply design of a Bryston 4BST is inadequate? I rather doubt we'll witness such at thing. The point is that one can lean on one's engineering principles/background, bring in ideas from other fields and what-not, but the rubber hitting the tarmac is a different story altogether.
---
OH, having had a peek inside a Bryston 4BST ... the one I had on a home trial, that incidentally embarrassed itself against the Kenwood L-07M monoblocks (late 70's production) I was then using ... I was a bit surprised by what I thought was relatively modest sizing of the capacitor banks. Certainly many other current designs provide a heck of a lot more for the power rating. Still from an engineering perspective no doubt the sizing is more than adequate; it's true that (per side) storage *was* greater than what was in the much older L-07M.
(The L-07Ms had had their captive zipcord replaced with garden variety outdoor (orange) extenstion cord (that's how they were when I got them) and IIRC I used the stock power cord provided with the Bryston. At that time I likely wouldn't have had anything much better lying around to try on the 4BST).
Everything matters, don't forget to tweak your placebos!
Edits: 06/25/09
Your findings raise a more-important point.......
Bryston amplifiers have often been deemed "properly designed" by the so-called "scientific" audio community, in regard to how little effect audio cabling should have on sound if interfaced to such designs.
But based on your findings, even this respected amp line is apparently affected sonically by cabling. (I've always suspected this would be the case. I've never come across an amplifier that was immune to cabling.)
So the term "properly designed" is nothing more than an ideal that seems to **never** exist in reality. Yet this idealistic term is often mentioned as factual basis for why audio cables don't or shouldn't impact sound.
It does seem to be mostly semantics doesn't it. In fact as best as I can tell at least one of our resident "engineers" (Don Till) seems to have taken it to a declarative/definition level by essentially stating that if an amplifier's performance varies with usage of power cords then by definition its power supply is not "well designed"... and it seems he's more than willing to include Bryston in the poorly designed group (assuming there's any truth to my report of course!).
So isn't that so very clever! Power cord are hence allowed to effect performance along with a nice tidy reason for why this is the case... our engineer sleeps well at night and all in the universe is in order!
LOL
Everything matters, don't forget to tweak your placebos!
"So isn't that so very clever! Power cord are hence allowed to effect performance along with a nice tidy reason for why this is the case... our engineer sleeps well at night and all in the universe is in order!"
And are you suggesting things are somehow different? As if there aren't tradeoffs made in every component design at every price point. Some of us might immediately condemn a slight high frequency roll off or minor transient shaping at the same time ignore, or even worse claim higher resolution from components displaying problems protecting itself from being influenced by external factors.
deny my call on your definition/declaration!
LOL
Everything matters, don't forget to tweak your placebos!
Why should I deny it? If I establish the level of acceptable performance for a well designed component as being not audibly effected by external influences in normal operating conditions it's not like I'm setting a high bar. Or am I?
And just to clarify this. I don't believe the Bryston as you describe it is working to spec (as advertised), just like my Raysonic CDP that greatly benefits from an "better" power cord. It's not like I think the sound of these things get betters than advertised when external power cords or conditioners are applied. I think only then do they deliver on the promise of the original purchase.
It may be that the particular instance of the amplifier was defective. This could be the result of a component or wiring failure. Or perhaps a batch of instances were defective. This could be the result of Manufacturing performing some "value engineering".
A company that sold audio kits many years ago used to do this kind of "value engineering". The purchasing agent would make "minor" substitutions of component values or tolerances to reduce parts cost. I heard this from the purchasing agent's son.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Of course, there's always a possibility of amplifier being defective - it's just that being affected by power cord change is in no way or shape the sign of "defectiveness".
More experience with resolving equipment, and less preconceived notions in the head - that's best possible cure against such "defects" in amplifiers.
Actually I was quite serious.
I have had a defective amplifier, and it was in the power supply, no less. As it turned out, the defect wasn't at all subtle—it resulted in a moderately loud hum in one channel. The amplifier had separate output filter capacitors for each channel and one of the four had gone bad. The manufacturer had gone out of business and there was no direct replacement available at "You-do-it Electronics" so I substituted four new capacitors of slightly larger size, which fit only when held in by duct tape. Interestingly enough, the result was a significant improvement in sound over what it had been prior to the failure. I presume that the original capacitors either lacked sufficient energy storage capacity when new, or perhaps had gradually lost their capacity over time.
Had the failure been in the input power line filter then I might not have noticed it, at least not until I moved into an environment with lots of EMI. At that point, I might have effected a sonic improvement by replacing the defective internal components or by attaching an external power cord that provides rejection.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
In this instance I have to rely on the veracity of bjh's observations - ie. they are presented in such a way that indicates it's not a one time observation and have been repeatable in a variety of different situations. At least that was my take on the wording of his opinion of Bryston equipment.
In my normal work environment when issues come up the first thing I like to do is to witness them. Whether this occurs in the lab or I reproduce them at my desk kind of depends on the importance and severity of the issue.
I agree with your take it could be a defective unit and the defect could be for one of any number of reasons. However and again - the comments I'm relying upon indicate this is the normal performance one achieves from the Bryston amplifiers in question.
Often when I get to the lab or reproduce the issues as described by a customer, technician, marketeer or even other engineer the issue is actually quite different than the description. I'm giving bjh the benefit of the doubt for the purposes of this conversation.
.
Everything matters, don't forget to tweak your placebos!
by members of self-proclaimed "scientific" community, for effects of substituting different power cords. The same as with, for instance, Van Alstine DAC, that according to manufacturer, was supposed to be immune to differences between transports - but wasn't.
Why would the pseudo-scientific community test for a foregone conclusion??? They might discover the world isn't flat!!! How can you remain an audio regressionist if you test things all the time???? :)
"Now do we have an engineer present who would like to step up to the podium and propose that the power supply design of a Bryston 4BST is inadequate? I rather doubt we'll witness such at thing.
Given that the amplifier as supplied merits the following comment when an after market power cord is installed -
" The second reason for the shock was due to the particular SS amplifier, a Bryston 4BST, an amplifier which along with its SST sibling has long ranked in my most hated list. The ills associated with the 4BST/SST, what I would call electronic sheen (a subjective description) seemed to me to have vanished!"
I would say yes the amplifier power supply filtering is inadequete, based on your comments for the environment in which you made your observations. I can't believe the designers of this amplifier intended the "electronic sheen" and I hardly doubt it takes an experienced golden eared trained listener to hear it.
But alas if one can buy a Bryston power line conditioner or any number of after market power cords to solve the problem who's to complain?
Personally this is the kind of crapola that long ago turned me against the marketing vehicles posing as audiophile magazines willing to tolerate such performance deficiencies to further their vested interests in growing the industry.
Sure power cords and conditioners can and do often make a tremendous difference and most anyone should be able to hear it.
Communications and test equipment owner's wouldn't tolerate for an instant such haphazardly designed products. Funny how audiophiles and the industry in general seems to tolerate it at the same time marketing spending for the work arounds as ways of bringing home improved performance.
I would have added an exception for you.
Everything matters, don't forget to tweak your placebos!
"The point is that one can lean on one's engineering principles/background, bring in ideas from other fields and what-not, but the rubber hitting the tarmac is a different story altogether."
No it's not a different thing, not at all. Whether possibilities jump out at you or need to be sought after does somewhat depend upon what 'other fields' you have been plowing.
So, the power cord changes the sound, what does that mean and to whom does it mean it? Does the product no longer meet it's specifications? Probably not. Does it mean that the specifications are inadequate? Maybe.
Most likely if there is an error it's that in the internal product requirements. Either what you are experiencing was regarded as insignificant or whatever is causing it was missed completely. You can usually rely upon products meeting their production specifications, what's a lot iffier is relying upon those specifications to predict the outcome for all users.
"I was a bit surprised by what I thought was relatively modest sizing of the capacitor banks."
Physical size? Capacitance? ESR? Net energy storage? Capacitors have come a long way the last decade, especially in terms of HF ESR. How much capacitance you need depends upon the headroom and PSRR of the amplifier.
One of the things that's fun about this hobby is all of the counter-intuitive things you find that can affect the sound. I've had rum luck predicting whether I'll like the sound of something without listening to it. Size, color, cost, weight, brand et al are poor predictors. How can you tell a "better" power cord without trying it in a particular application?
Rick
> > "The point is that one can lean on one's engineering
> > principles/background, bring in ideas from other fields and what-not,
> > but the rubber hitting the tarmac is a different story altogether."
> No it's not a different thing, not at all. Whether possibilities jump
> out at you or need to be sought after does somewhat depend upon
> what 'other fields' you have been plowing.
If the exposure to 'other fields' merely informs to discount anecdotal evidence (power supplies are adequate, hence reports are fictions), or dissuades investigation (power supplies are obviously ill-designed, not worth bothering with), which seems to me to be just barely concealed (intentional or otherwise) message(s) in tomservo's post, then I would maintain the 'other fields' applicability is null.
I won't bother with the other points as they seems to me more than a little argumentative oriented.
Everything matters, don't forget to tweak your placebos!
I don't understand where you are coming from. This is a discussion, not an argument. I didn't see Tom discounting the issue as much as advocating that it can be fixed. I'm essentially on the same page but in trying to address your question of whether hearing a difference due to changing a cord meant a defective power supply tried to broaden it out to examine reasons both social and technical that they might especially at the system level even if the individual components were within spec.
Sorry if I waxed too philosophic, no harm intended.
Rick
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]