After much experimentation with power supplies I came across this latest technology LDO regulator from Linear Technology. Ultra low noise of just 0.8uv! Ultra high and broad DC power cleaning. From 118dB @100Hz to 76dB @ 1Mhz. That is removing between 99.99999% to 99.999% of DC noise.
DC to DC. So just feed with your DC power supply, preferably a linear one with an R-Core transformer. Take a low cost Chinese LPS and clean the power output to below 1uv of noise. A very economical solution to clean power for your audio device. This box is fed by your SMPS or LPS then outputs the ultra clean DC to your device. Rated to 500mA (the USB std) without using noisy transistor boosting.
My experience with my USB audio chain is the lower the noise and better AC and DC grunge suppression the better the SQ. Why buy a $400 to $1000 LPS when all you need is a $80 to $150 LPS and one or two of these?
Available in 5VDC, 7.5VDC , 8VDC outputs. The input voltage can be between 0.5V and 3 volts higher then the output. Simple set-up to feed your USB/SPDIF/I2S converter like the SU-1 - LPS (6VDC)+LT3045 set to 5VDC.
Or your Singxer F-1/SU-1,Pro3z, etc...: Split USB cable - power leg to LPS (6VDC) + LT3045 set to 5VDC.
Using these sota DC noise filters and regulators right at your devices - is a move to a new form of power conditioning called 'Distributed Power Conditioning' as described here by noted audio Engineer Caelin Gabriel - President of Shunyata Resaerch http://www.audioshark.org/power-conditi ... 11713.html
Quote: "b[Your dedicated power lines, rhodium plated outlets and large gauge wires in the wall in no way immunize you to power line noise. Power line noise is picked up from radiated electromagnetic waves all around you. This includes AM and FM radio, wifi, cell frequencies and the list goes on.]b Furthermore noise reduction is always "localized". That means that just because you have a power conditioner that reduces conducted noise does not mean that the noise doesn't re-introduce itself farther down the line. So if you have some form of power conditioning at the electrical panel, the radiated noise will be picked-up by the in-wall wiring that acts like an antennae. This means you WILL have noise on the power line at the wall outlet. b[This is why we talk about the "distributed power conditioning" approach to noise reduction]b."
The use of multiple LT3045 boxes right before the device is just such a 'distributed power conditioning' approach. And not all that expensive. The new power cord design attacks EMI/RFI right at the cord itself - which at 1M can act as an antenna - so working to quash that noise before it enters the DAC or LPS.
Other uses as well. Like splitting the power coming from a LPS to different voltages. Say a LPS outputting 7VDC. You can use a 'Y' DC cable and run one leg to a device needing 7VDC and one of these 'Miracle Boxes' from the other leg to output ultra clean 5VDC. So one voltage LPS can be used to power two devices with different voltages and better yet - due to the LT3045 ultra high noise suppression - noise from different devices on a shared line can now be isolated from each other.
Another use for example running two devices that require the same voltage, say one 7VDC to two of these set to 5VDC. The LT3045's act to clean the DC power and isolate each device from each other. I do this exact thing in my system.
These boards can be bought and assembled yourself - but many folks are not soldering inclined. So I built a few extra ones for folks to try.
Specs: Non-conductive ABS case Silver/Teflon wiring/Nonlead silver content solder Silver plated contact connectors (I tried a few and these had the most secure connection)
Fixed voltage LT3045 board for lowest noise (least resistors) 0.8uv noise, -76dB @1Mhz PSRR Features
• Output Current: 500mA • Low Dropout Voltage: 260mV • Ultralow RMS Noise: 0.8µVRMS (10Hz to 100kHz) • Ultralow Spot Noise: 2nV/√Hz at 10kHz • Ultrahigh PSRR: 76dB at 1MHz • Built-In Fixed Current Limit and Thermal Shutdown • Operating Temperature Range:–40°C to 125°C
LT3045 x1 : IC, Low Noise Power Solution, Linear Technology – United States T495D476K025ATE250 x 1: Cap Tant Solid 47uF 25V D CASE 10%,Kemet – Japan GRM31CR71E106KA12L x2 : Cap Ceramic 10uF 25V X7R 10% SMD 1206, Murata – Japan GRM31MR71E475KA88L x1 : Cap Ceramic 4.7uF 25V X7R 10% SMD 1206, Murata – Japan CRCW0603xxxxFKEA x1: Res Thick Film 0603 1% 0.1W(1/10W) ±100ppm, Vishay – U
I test each box for voltage output (actual output test read out given for each unit) and continuity of connectors, listening test in my system, 1 hr burnin.
I also have: Tinned copper braid shielded 2.1mm DC to USB female adapters (great for powering a DDC with spit USB cable). Tinned copper braid shielded micro USB to 2.1mm DC. Quad star Silver/Teflon wiried 2.1mm to 2.1mm DC short connector cable (switchcraft 2.1mm plugs). Tinned copper braid shielded SATA cables - Asaka, Coboc flat silver, Amphenol silver dual. Thanks for looking