There was a study done by a Japanese firm and it is somewhere in the AES papers. However, my personal experience shows me that a steel chassis can be audible. In fact, I had the Parasound JC-3 changed from steel to aluminum, where possible, and found an audible improvement. I do not know why this was so, but I would not release the JC-3, until we changed to aluminum. The JC-3 has been awarded an A rating, this is one of the reasons.
I realize that this seems strange, but sometimes we just don't know WHY something can make so much difference, because it can be virtually unmeasurable. I just follow my experience. Works for me.
I realize that this seems strange, but sometimes we just don't know WHY something can make so much difference, because it can be virtually unmeasurable. I just follow my experience. Works for me.
I have very bad experience with steel enclosures in instrumentation for power (high-current) testing labs. It was absolutely necessary to avoid steel or mumetal and to use thick copper or aluminum. Of course, magnetic flux density and intensity of magnetic field was in orders higher than at home audio, but, on the other hand, signal level higher than the lowest audio signals.
For audio, I suspect even the phenomena like transformer magnetic flux coupled into steel metal enclosure. It is not only "low order" and in some cases the change of enclosure from steel to thick Al is measurable at the bottom line of the FFT analysis of the preamp output signal.
For audio, I suspect even the phenomena like transformer magnetic flux coupled into steel metal enclosure. It is not only "low order" and in some cases the change of enclosure from steel to thick Al is measurable at the bottom line of the FFT analysis of the preamp output signal.
I also have tried conductive plastics for audio, John. The result was not good. The unit was quite sensitive to placement with regards to another instruments. That means, placed on top or aside something that had a big transformer inside resulted in worse S/N and hum components audible and measurable. IMO conductive plastics might be used with consumer audio or mid-fi, but not with high end.
This is the nightmare world of fine tuning a complete system to get optimum sound. The first time I went down this road I burnt myself out with frustration, because I couldn't get a handle on what was going on. Later on, I learnt to pace my mod's: try something, let it settle down until I had a clear picture of what the long term benefits were, then move on to the next thing. It's taken years to get a decent understanding of what matters, and it's mostly none of the usual, obvious stuff.
The best measuring tools have always been my ears, I don't use any of the standard analysers and I'm certain I haven't lost out on anything by not having those on hand ...
Frank
Of course, PMA, you are correct. However, to save a good deal of money invested in the case, I was wondering what is next best? Perhaps most of the added 'EMI' in our environments, today, is electrostatic or line induced. We could fix those sources without a heavy metal case, couldn't we?
The scourge now mostly for the "low-end" is mandated switchmode a.c. adapters, to eliminate or reduce standby current drains from lossy double-bobbin transformers --- Balakrishnan's colorfully-named "energy vampires". The old adapters were bulky and inefficient, fairly strong magnetic field emitters, but had the virtue of low interwinding capacitance. And for the mag fields, distance helps a lot as the field falls off as the cube of the distance. Now, not only are the adapters themselves strong interference sources, generating mostly conducted high-frequency noise modulated by the mains and loading, but the coupling of junk from the line is much greater. And more and more junk is being dumped onto the mains.
In consumer electronics the convention of a relatively low capacitance coupling for the internal supplies or plain iron-copper external supply meant the whole shooting match tended to sit at about half of the mains voltage and move together. Hardly perfect, but if the electrical shielding were reasonable and the magnetic emitters positioned optimally and at some distance from pickup loops, you could get by. With the mandated switchers it becomes a lot harder. Or, before that, the referencing of some equipment (satellite stuff, computer sound cards) to the safety ground made things no longer move as one, and provoked line related noise.
All of the above can be dealt with in various ways, including balanced modes and very good common-mode chokes, but these are not that cheap and not without other potential penalties.
The last time this “unmeasurable” effect has resurfaced in this thread, was only a month ago. See links below. (previous bursts of interest were with shielding of input x-formers and shielding of cartridge bodies, more than a year back)
Do all questions still remain open? Come on people. There is enough of input (Jneutron has a multiple of times contributed on the issue) for realistic solutions.
FLEETWOOD MAC - Oh Well (1969 UK TV Performance) ~ HIGH QUALITY HQ ~ - YouTube
Are you looking for a housing of your preamplifier, that is a perfect electromagnetic shield to internal and external fields 0.1Hz –100THz, one which will also re-emit nothing back, 4 pi space, at any miniscule distance?
Your Jinni heard your (last) wish and noded “Yes Master. At no cost. Right now”.
Floop!
It has the shape of a ball. It is a hollow large ball.
Jinni laid in front of you the detailed engineering drawings.
The internal hollow space is as large as you wanted it to be.
The wall material has a varying electrical conductivity through it’s thickness, starting and ending at 0.000… IACS on both sides of the wall and in between increases to a value >>100, following a very wide, symmetrical, bell shape distribution.
It’s magnetic relative permeability follows the same distribution, starting from and ending to 1.000…, reaching in the middle of the wall thickness a value of some millions.
The thickness of the wall is >>1m.
There are no openings and no discontinuities of the wall material.
Now you only have to use it.
(silly) George
George, what is the problem? I KNOW HOW to make good, relatively effective cases, BUT they are SO EXPENSIVE! They DOMINATE the building cost. Is this a good balance? I was hoping for some SHORTCUTS that might be almost as good, yet not cost anyway near as much.
However, the usual engineering solution would be to use a relatively thin steel case. BUT I will NOT use a steel case, because I can hear them when they are in physically close contact with my designs. I last tested this the the JC-3 phonostage design.
Now, let us go forward with some new approaches! Graphene? Silver foil?
However, the usual engineering solution would be to use a relatively thin steel case. BUT I will NOT use a steel case, because I can hear them when they are in physically close contact with my designs. I last tested this the the JC-3 phonostage design.
Now, let us go forward with some new approaches! Graphene? Silver foil?
A sphere with no openings. what rating on the practical scale is this?
Congrats Morinix.
The highest rating!
It’s the price one has to pay if he is looking for theoretically perfect solutions.
Now, start thinking how this construction can become usable.
With every step you make, one elementary condition of theoretical perfection will be violated and it’s effects will be canceled.
Consider the importance each canceled effect has on your goals.
George
However, that said..Amperes law states the magnetic field (B) around an electric current is proportional to that current. And Faraday's law states that the voltage generated in a loop of wire is proportional to the rate of change of (B) flowing through the loop.
Neither law invokes H, so if H is a non linear function, it does not directly impact either Ampere's law nor Faradays. So that's how I'd explain it, if I knew what I was talking about...
The non linearity of a BH curve will certainly impact the inductance of the system, and when you start to saturate, that inductance will drop. All these factors will come into play with the primary current, so the IR drop and varying impedance of the primary to its driver can cause all kinds of nasties..
BTW, I looked up hyperphysics to confirm what I've said w/r to Ampere and Faraday..really nice site, University of Georgia I believe..
jn
Jneutron
I really hadn’t noticed this part of your post.
I am surprised that I see H and B changing roles.
Maxwell refers to the entities of:
H as the magnetic force of the field and to
B as the density of the lines of force of the field. It names B also as magnetic induction.
Both are vectors.
Vector of H functions over lines. Line integral denotes work.
Vector of B functions over surfaces. Surface integral denotes flux.
He explicitly stated that magnetic force produces magnetic induction. (A Treatise on Electricity and Magnetism Vol I, Preliminary, Forces and Fluxes, page 11, see attachment)
On BH diagrams, H is the abscissa on the horizontal axis.
B the ordinate, is on the vertical.
In modern terms, B is the degree of magnetization a material obtains in response to an applied magnetic field of strength H.
The factor of proportionality μ (B= μH) is not a constant. It is an attribute of the medium through which the lines of magnetic force travel.
Any variation or non linearity of μ distorts B.
George
Attachments
What distance from steel reduces the effect to audibly zero? Find that, make a cube of thin steel with double those dimensions and use spacers to position the components to be in the middle of the cube.
May look awkward, but certainly cheaper ... !
Frank
Aluminium and copper for chassis are the two best non-ferrous materials. Cant get lower cost either. It cost what it cost.... wall street has figured it out -- commodities are the way to make the most money. Watch out for metals and food and the like. Thus smaller physical builds are the way to go for lowest cost of chassis/materials. Thus, iPOD/PAD et al. For the Highest-End its practical to physically seperate/isolate or move noisy elements/circuits away from sensitive circuits.
You need to worry about both field-coupled as well as direct-coupled modes of unwanted signal/noise pickup. -RNM
You need to worry about both field-coupled as well as direct-coupled modes of unwanted signal/noise pickup. -RNM
Last edited:
Erratum
Sorry Morinix. I misread your post. (non practical scale)
So on the practical scale, the rating would be very low (for this application).
The congrats and all the rest remain the same.
George
Congrats Morinix.
The highest rating!
George
Sorry Morinix. I misread your post. (non practical scale)
So on the practical scale, the rating would be very low (for this application).
The congrats and all the rest remain the same.
George
If the chassis is a cost driver I would expect there to be plenty of data showing the detrimental audio effects, and supporting evidence that a thick chassis is the best solution.
Its not apples to apples but all the low signal level instrumentation (HP analyzers etc) seem to get by with thin multiple electrostatic shields, and I would expect these are often used in a harsher EMC environment.
There are also always practical trade-offs, an electroless nickel plated aluminum chassis may be discarded by some for its magnetic properties, ignoring the improved contact impedance and resistance to oxidation.
I really do like the look of a thick aluminum chassis, very different from the cheap Bud box experiments which may have housed the equivalent prototype.
Thanks
-Antonio
Its not apples to apples but all the low signal level instrumentation (HP analyzers etc) seem to get by with thin multiple electrostatic shields, and I would expect these are often used in a harsher EMC environment.
There are also always practical trade-offs, an electroless nickel plated aluminum chassis may be discarded by some for its magnetic properties, ignoring the improved contact impedance and resistance to oxidation.
I really do like the look of a thick aluminum chassis, very different from the cheap Bud box experiments which may have housed the equivalent prototype.
Thanks
-Antonio
Well written.
I try to keep the explanations as simple as possible, but no simpler. You'll drive others nuts if you push E/M theory too far here.
But then, I'd have company, no??
jn
To quote the maravingian.. there is only cause...and effect..
- Status
- Not open for further replies.