Grant - Let me add that these narrow spikes have width (conduction angle) and therefore peak amplitude dependent on impedance of power transformer and ESR of capacitors (Schaffer's diagrams). Often with huge banks of capacitors in parallel (very low ESR) and big transformers peaks can be really high. It causes big voltage drops in power cords but also heats-up transformers since rms value of power (related to heat) is much higher than average value hence heating windings, while high frequency content heats-up the core. This high frequency content of the very high current spikes is also radiated by the cord. Linear power supply is in reality a switching power supply operating at 120Hz where "ON" time is proportional to load.
As for telling others what they should or shouldn't hear I'm guilty of it as well. I expressed opinion that Benchmark DAC1 is insensitive to transports and digital cables knowing that it has huge jitter suppression (2Hz jitter bandwidth) in order of >100dB at the frequencies of interest (kHz) on the top of very low amplitude of the sidebands (-60dB) created by the jitter. Many people confirmed it but also many found difference in their systems. I don't know why and cannot explain it, but it was wrong for me to say they imagined it. Perhaps something else changed - like electrical noise (FCC requires transmitters to cut power around 6PM), warming of the gear (often ferrofluid in the tweeters), ground loops or just simply I'm wrong and Benchmark is not that perfect. Now I only say that I cannot hear any difference IN MY SYSTEM.
As for telling others what they should or shouldn't hear I'm guilty of it as well. I expressed opinion that Benchmark DAC1 is insensitive to transports and digital cables knowing that it has huge jitter suppression (2Hz jitter bandwidth) in order of >100dB at the frequencies of interest (kHz) on the top of very low amplitude of the sidebands (-60dB) created by the jitter. Many people confirmed it but also many found difference in their systems. I don't know why and cannot explain it, but it was wrong for me to say they imagined it. Perhaps something else changed - like electrical noise (FCC requires transmitters to cut power around 6PM), warming of the gear (often ferrofluid in the tweeters), ground loops or just simply I'm wrong and Benchmark is not that perfect. Now I only say that I cannot hear any difference IN MY SYSTEM.