Dan, this excess noise and CSD noise plot you have started to speak about is bla-bla. Show a level diagram of your complete system (from source to pwr amp output) and also noise measurements on a complete system and then we can talk. If the audio chain was designed and set properly, then it would have zero audible noise contribution, all the noise would be in the recording itself.
That is where you change the hot side of the primary winding from outer to inner, makes a difference to the coupled in mains hum, possibly to the amount of RF sneaking in too.
Those are the key difference. I sense that Dan has been struggling to somehow improve crappy stuff by a plug and pray method. Unlike your way to cure problems at the root level and ditch the hopeless, his method is still in need of refinement in the areas of measurement, repeatability and reproducibility of result for others to understand.
Currently making my way through Arto Kolinummi's book, (that Jan D. publishes); looks promising for a devoted amateur like me. But I'd be interested in your comments on Bob Cordell's book. I found it amazingly helpful to folks on my level. (He also gives you credit when showing your input design, which Kolinummi doesn't (as yet) seem to. But I could easily be wrong.
Bonus: how fast can you find the error in figure 7.14?
Much thanks as always, and all good fortune,
Chris
It's still the same deal that you heard a year or so ago, but improved since then.
I have run some loopback recordings and I have extracted difference signal 70dB or so down from excitation signal just for the USB soundcard.
In the next few days I will run loopback tests on a couple of consumer integrated amplifiers.
So, I have a residual that sounds like music, except harsh/nasty sounding music, next step is to analyse it and extract time information, ie freq vs decay time information.
Dan.
That´s fine especially in this case where it seems that you completely missed the point.
I´ll try to strip it down to the core; i responded to PMA´s post:
"Pity that no one is able to give a proof in a controlled abx that he could hear about 1% H2/H3 IMD on music. Just talking, no results."
Iow he was stating that nobody could provide evidence that IMD products (i think he meant IM2/IM3) at - 40 dB (or below) .
In my post i mentinoned that in another publication results from listening experiments IMD products were found to be detectable (means were audible) at levels of around -60 dB to -65dB .
No struggles, it is straight up 'plug and play', no questions, no praying.Those are the key difference. I sense that Dan has been struggling to somehow improve crappy stuff by a plug and pray method. Unlike your way to cure problems at the root level and ditch the hopeless, his method is still in need of refinement in the areas of measurement, repeatability and reproducibility of result for others to understand.
Indra you are correct, the proper way is to build quiet systems as JC has strenuously done, however most of us cannot afford such gear, and besides hi-end is a relatively small demographic.
I have stated that if a system is 'noiseless' to begin with the improvements are less....no matter what system (up to and including hi-end $250k systems) the changes are still beneficial however.
I am not aiming at hi-end of gear, I am aiming at typical consumer gear...and stage/musician, PA gear and recording gear as OEM or retrofittable solution.
Please understand this research is not about me, it is about delivering better sound for everybody, from pocket MP3 player through to stadium concert systems (already trialed btw).
@ Bonsai & Syn08,
i was a bit puzzled by your negative judgement about MC´s role in this game (MC means Martin Colloms, or am i mistaken?) .
I remember that the first article i´ve ever read from Colloms was from 1985 where he was writing about differences between passive components like resistors and capacitors. In this article Colloms wrote:
"For the purposes of this article, the majority of the results are anecdotal, reporting what happended - or, as some may prefer to see it, what we thought happened - when a passive omponent was inserted or substituted. These assessments were carried out in the same manner as for other hi-fi products, with much the same attitudes present in the listeners´minds, the main differences being due to the large number of "products"
available.While readers must decide themselves whether or not to take these subjective findings on trust, a small number of controlled double-blind tests were also undertaken and the results from these tests are presented seperately."
(Martin Colloms; A passive Role?; Hifi News and Record Reviews, October 1985, 35)
Overall that looks like a honest assessment, clearly describing what is anecdotical and what is a more objective approach and correctly describing that the results (numbers were included) of the "blind" tests did not provide hard evidence for the said differences.
Iirc John Atkinson wrote several times about "blind" listening tests of amplifers (for example) organized by Martin Colloms, in which JA particiapated before being engaged in Stereophile.
i was a bit puzzled by your negative judgement about MC´s role in this game (MC means Martin Colloms, or am i mistaken?) .
I remember that the first article i´ve ever read from Colloms was from 1985 where he was writing about differences between passive components like resistors and capacitors. In this article Colloms wrote:
"For the purposes of this article, the majority of the results are anecdotal, reporting what happended - or, as some may prefer to see it, what we thought happened - when a passive omponent was inserted or substituted. These assessments were carried out in the same manner as for other hi-fi products, with much the same attitudes present in the listeners´minds, the main differences being due to the large number of "products"
available.While readers must decide themselves whether or not to take these subjective findings on trust, a small number of controlled double-blind tests were also undertaken and the results from these tests are presented seperately."
(Martin Colloms; A passive Role?; Hifi News and Record Reviews, October 1985, 35)
Overall that looks like a honest assessment, clearly describing what is anecdotical and what is a more objective approach and correctly describing that the results (numbers were included) of the "blind" tests did not provide hard evidence for the said differences.
Iirc John Atkinson wrote several times about "blind" listening tests of amplifers (for example) organized by Martin Colloms, in which JA particiapated before being engaged in Stereophile.
How are you getting on with your miniDSP Bill? The main reason I opted for the DCX was that it's a "stand alone" unit, I don't have a PC or laptop that runs windows, I know it limits me these days and here where almost everyone uses it for audio but I'm still holding out from going down that rabbit hole. Windows seems synonymous with PITA from what I read and my few dealings with it, I can do without that.....
large lithium packs need careful care and feeding. DIY chargers for care of ex car batteries don't exist and not even sure commercial ones either unless you literally put the front end of an electric car in. This is so far off the design brief that your insurance would possibly not payout if things do go wrong.
But mainly fire, and the fact that lithium is the gift that keeps giving. A crashed tesla last year re-ignited 5 days later.
Suddenly a sealed lead acid and the silent switcher seem a really really good idea.
Good suggestions, but....
Radios and Radio-Cassettes of the day all ran Double Insulated compartmented transformers like in the pic, so presumably Pr>Sec capacitive coupling is minimal and being a floating AC load there should be no real difference in coupling of line noise to circuitry according to supply polarity ?.
Anybody have any ideas on why the sound differences ?.
Dan.
Last edited:
My thoughts are for line level gear......modern phone auxiliary battery banks provide plenty of energy storage and are safe in my knowledge.
For power amplifiers it's a different story of course but still doable.
Dan.
Still in the love/hate phase. I wish it had more presets and often wish I had the DCX instead but here I am. For advanced EQ duty I have the option of a laptop running ubuntu studio so all bases covered and max heresy
I did think of getting a DEQX for live EQ, but then realised I would need two so went for the laptop approach.
Dan: For preamp a phone booster pack and silent switcher do me nicely. 5V in +/-15v out job done.
Not complicated. You are connecting together two or more AC powered devices.
Depending on the way each one is connected in the AC, you will have a relative difference in their AC "leakages" to the ground. A voltage that you can measure between their grounds before to connect them together. This leakage voltage (high impedance) will generate a parasitic current across the grounds of each device, once connected.
The first thing I do when I connect each device in a system is to measure this voltage , starting with my pre-amplifier alone, choosing the best phase for the amplifier, and doing the same for each device i add after this.
Of course, it is less critical if you use true symmetrical floating signal connections everywhere.
I used during years a little headphone amplifier that I carried from studios to studios. It was both AC powered and batteries (4X9V). The difference in clarity was obvious. Both between batteries, best AC phase, worse AC phase.
In this order from best to worse.
Thanks Tourney, I well understand transformer leakage currents and transformer primary phasing/polarity to minimise chassis/earth capacitive leakage currents.
I am talking about standalone plastic case fully insulated radios and radio-cassette players like the pic and mono versions of the day.
Dan.
I am talking about standalone plastic case fully insulated radios and radio-cassette players like the pic and mono versions of the day.
Dan.
Last edited:
Wrinkle has it right. Where the hot lead is wound closest to the core more noise couples to the core, ground and secondary. There is more leakage capacitance from the core to secondary in addition to primary secondary measurement by itself.
I showed measurements of this a while back.
- Status
- Not open for further replies.