Equipment power capacitive earth leakage currents are indeed important and good reason to run power filtration to reduce at least the BW of mains borne noise.
In my testing and auditioning I am running fully floating system including battery powered Laptop/USB SC to remove such dependencies. It will be interesting to measure and compare system excess noise with and without power filtering.
Dan.
In my testing and auditioning I am running fully floating system including battery powered Laptop/USB SC to remove such dependencies. It will be interesting to measure and compare system excess noise with and without power filtering.
Dan.
Max, I like batteries too, but they are just too expensive to maintain. Still, almost all wall-warts are next to awful in performance and RFI bleedthrough.
I admit to know several who fit your description perfectly, someone even demand DBT result just to flip cables and listen to reverse speaker polarity. But I also understand that some others consider "X is negligible" because they are truly not able to discern any difference whatsoever. I'd consider them to be partially disabled. It is perhaps more appropriate to offer explanation through measurement and/or other means rather than condescending remark that includes the less fortunate.
TX wall warts are all sectioned windings in my experience.
SMPS wall warts have too much longitudinal AC noise/switching noise coupling.
I have run battery powered preamps and dac front end and have found sonic differences according to battery type/chemistry.
For my testing I am running battery power (laptop between recharges) to eliminate confounders such as earth loops.....I am looking just above the system noise floor and also into the noise floor to extract difference signal.
Dan.
Measuring with what?
Also, you do realize that every rail in your laptop is created from a buck or boost switching converter?
Also, you do realize that every rail in your laptop is created from a buck or boost switching converter?
how do USB cables make a difference if the data at the receiving end is it perfect?
Let’s assume the csbles have no EMI issues and there are no earth loopsetc.
Let’s assume the csbles have no EMI issues and there are no earth loopsetc.
Making a good AC - DC supply is not easy. (in my opinion) I try to avoid toroid transformers, because they leak too much. I prefer R core or low leakage E-I (that is what the Blowtorch uses) for best isolation from the power line, and then there is the additional filtering of common mode and differential mode RFI as well as ripple reduction and lack of droop. I usually have about as many passive inductor/caps as I do active filters in the power supply. Still, is it enough? In my heart, I still prefer batteries.
In my experience, it is much easier to filter or regulate out any noise than it is to shield from the magnetic field of the transformer itself. Unless you have a ton of distance to work with or a second box, which brings its own problems.
Exactly. If it's a competent receiver like the XMOS or CM663x implementations, the endpoint is asynchronous and buffered with rate controlled via feedback to the host. The data is clocked out of the buffer with the local, presumably clean clock.
Any other issues like you mentioned are not unique to USB and can affect any other type of non-fiber interconnect.
With Dante, I'd have to see the implementation, but if using Ethernet you now have galvanic isolation in addition...
I sense a market for audiophile air sprays to improve the quality of Wi-Fi and Bluetooth 😉.
Last edited:
Is it more leakage or is it the closer coupling btn primary/secondary , thus no LP filtering?
(I can’t understand why R cores have higher leakage current than R or E-I)
George
Toroidal transformers have much higher primary/secondary capacitance than EI transformers, for obvious reason. Just take your capacitance meter to check this. You will also measure higher AC leak current from primary to secondary. Another obvious fact, I would tend to write primitive fact. Supposing there is no metal shielding layer between primary and secondary.
last night I finally got around to converting my phono stage to battery +Jans silent switcher. Sadly no veils lifted or angels weeping in the living room but does work.
Pavel what I know as “leakage current” in transformers is AC current leakage from primary to core, to earth (affecting earth loops) and is due to capacitive coupling(and this is what I asked JC).
By the same mechanism (capacitive coupling) there is AC current btn primary and secondary as you say (leakage).
George
By the same mechanism (capacitive coupling) there is AC current btn primary and secondary as you say (leakage).
George
Ummm, laptop pc and usb soundcard as a floating two wire connected measurement system.
I am running loopback recording with filters fitted or not fitted around the audio connecting cable.
Sinewave excitation shows no FR, distortion or excess noise changes, however non stationary waveforms (including music) do show bursts of decaying (excess) noise immediately after transient excitation.
The spectral nature and rate of decay of this excess noise is dependent on the formulation of the filters.
I will also run tests of Ferrite filters....I expect confirmation of my subjective findings which also matches that of other listeners.
Dan.
Not really - Destroyer said
That's pretty assertive, and seems worth pursuing, doesn't it?
Which in this broad generalization is truly incorrect. It might be the question if it helped people really to buy the best stereo system for their needs, but there are a lot of people who bought a certain kind of reproduction system after getting "DBT" results.
At the bottom it depends on the design of experiments where "blind tests" are part of the experiment. I should have said "seriously flawed experiments" instead of "....flawed blind listening tests" .
A clearly expressed hypothesis/question that will be examined should be the first step and then starts the design of an experiment which might include a controlled (blind) listening test following a protocol that is best suited for this specific task.
Unfortunately a lot of this process is usually missing which leads to a kind of exploratory data sampling without really knowing what is done but nevertheless followed by wide ranging categorical conclusions from any results.
Overall something that violates the rules of propper scientific work.
Last edited:
You also heard differences between 2 files with equal checksums. How shall I use your listening reports after that event do You think? Seriously - help me to understand how I shall use your reports with that in mind...
//
//
- Status
- Not open for further replies.