Noise reduction -why the rage now ?


Is it better science ? Or more needed today with the ’surge’ of switching power supplies, smart phones and Wi-Fi routers ? Then, higher-frequencies in digital interfaces -now megahertz and gigahertz.

The problem with the "needed more today" theory is that one, listeners are hearing things they never heard before. One report is from Paul of PS Audio (on his blog) stating that a 4K power cord transformed his system. Yet he sells a Power Plant.

Two, many reports of lower noise (in the last 10 years) were in systems with linear power supplies, not switch-mode.

Three, for digital, many people were using AES/EBU or i-squared interfaces. These don’t have the noise concerns of USB.

Fourth, many people listen at night. They know better, due to the problems of a noisy day-grid.

Yet, there it is -large gains from the latest power conditioners, ground blocks and after-market power cords. To me, today’s science is much better at doing this. Along with the "need today".
jonnie22

Showing 2 responses by kijanki

Two, many reports of lower noise (in the last 10 years) were in systems with linear power supplies, not switch-mode.
Benchmark Media would not agree with it.  They lowered noise in their DAC by 10dB (S/N= 116dB to 128dB) by replacing linear power supply with SMPS.
Their power amp AHB2 has SMPS and is one of the quietest power amps available (S/N=132dB).  Technical director John Siau claims that such noise performance is not even obtainable with linear supply.  All their products contain SMPS.  Same goes for Rowland.  Rowland uses them in all products including preamps, where efficiency is not important.  They explain benefits of SMPS in application notes on their website.  

Most of linear supplies are in fact primitive switchers that operate at 120Hz.  They switch at maximum voltage and produce current spikes of huge amplitude.  They also produce very narrow voltage spikes, when rectifier diodes at the peak of the full wave sinewave get polarized in reverse direction.  These extremely fast spikes can couple to any circuit.  In contrast, well designed SMPS, switches in zero current, zero voltage and produces very little noise, that is easy to filter out (because of high frequency).   SMPS is also line and load regulated with fast response to line and load changes, while linear supply in power amps is completely unregulated and needs a lot of capacitance to keep amp quiet and voltage steady.  Since these large electrolytic capacitors are inductive they slow down response to fast transitions.

SMPS got bad rap from crude applications on computer power supplies and such cheap things as dimmers.

So why still so many linear power supplies?  For two reasons IMHO.  Because designing high quality SPMS is not easy and because most of people (including you) still believe that SMPS are very noisy (designers simply respond to demand)
Some noise might be originating at home (computers, dimmers, electronics, appliances, WiFi etc.) and some is coming from the outside on power lines, from cell towers and radio stations.  Many years ago I had loud noise coming from my TV (that was in standby) resembling police communication.  Every time it happened big grey unmarked car with a lot of antennas was passing by.  His radio was likely out of wack polluting airways.  

You mentioned better results when listening at night.  In part it might be due to FCC regulation that requires radio stations to lower power at about 6PM, since signal propagation improves at night.