RF noise: mechanisms and audibility

RF noise, grounding and power

Answers on this topic

• Do digital cables (USB, coax, glass or plastic optical) make a sonic difference, and if so through what mechanism?
• How does RF noise get into a DAC, and why does it make the sound brighter?
• If the data reaching the DAC is bit perfect, can the source (laptop, phone, streamer, CD transport or network gear) still change the sound?
• Should I put ferrites on my BNC or USB cables, and which kind?
• Should I use optical or coax into my Qutest?
• Why is the M Scaler a separate box instead of being built into the DAC?
• Why might a brighter or more detailed change actually be a step backward?

Everything he said on this topic (249)

One-line summaries in our words, oldest first. Each link opens the original.

• 2014-04-16Watts says he prefers optical as smoother and more natural; it has more jitter than coax, which his DACs remove, but reduces ground-induced RF noise and noise floor modulation. Head-Fi, Apr 2014
• 2014-04-16Watts compares PCM and DSD strengths and weaknesses, says a long filter fixes PCM timing while DSD loses low-level detail, and explains why Hugo filters DSD to PCM. Head-Fi, Apr 2014
• 2014-04-17Watts says Qute and Hugo use isochronous USB with timing from the DAC, so reported sound differences are not jitter but RF noise, sounding like reduced noise floor modulation. Head-Fi, Apr 2014
• 2014-04-26Watts explains Hugo noise with Bluetooth sources as amplitude-modulated 2.4 GHz picked up by headphone cables, and suggests looping the headphone cable through solid or split ferrite cores. Head-Fi, Apr 2014
• 2014-05-13Watts says Hugo's USB chip takes timing from the FPGA, but the DPLL makes that minor; input differences come from RF noise causing noise floor modulation, so he prefers optical. Head-Fi, May 2014
• 2014-05-13Watts lists ways RF noise enters, says many audiophile cables are RF-poor, some like noise floor modulation's brightness, and advises digital connections with good RF performance. Head-Fi, May 2014
• 2014-06-09Watts recounts 1980s mains cable experiments that showed RF noise mattered, leading to a 12-stage RF mains filter, and explains RF-induced noise floor modulation, audible even at -200dB. Head-Fi, Jun 2014
• 2014-06-09Watts says USB affects sound via jitter and RF noise; Hugo solves jitter internally, add-on USB units add RF noise though isolation may help. Head-Fi, Jun 2014
• 2014-06-22Watts re-tests optical vs coax on Hugo, finds optical warmer with better separation, and explains how RF-induced noise floor modulation harms separation and timbre. Head-Fi, Jun 2014
• 2014-06-24Watts reports the 33-pound Olimex USB-ISO on Hugo's SD USB input (44.1/48 kHz only) gives better depth, focus and separation, consistent with better RF isolation. Head-Fi, Jun 2014
• 2014-10-01Watts says USB cables change sound through RF noise causing NFM and brightness, while signal-dependent ground currents degrade depth; galvanic isolation addresses both. Head-Fi, Oct 2014
• 2014-10-12Watts says his Olimex tests used a noisy Dell, so an iPhone may need no isolation; smoother sound and better depth indicate lower RF and correlated noise. Head-Fi, Oct 2014
• 2014-10-27Watts says RF in Hugo's ground plane causes bright noise floor modulation and zero-crossing currents degrade linearity and depth; the Olimex removes LF currents and matched optical for him. Head-Fi, Oct 2014
• 2015-04-15Watts confirms isolation follows the USB chip, which draws PC power, so Hugo lacks it; laptops cause earth-loop and RF problems phones do not, and isolation removes audible laptop differences. article, Apr 2015
• 2015-11-21Watts explains Dave's DSD filtering, how out-of-band noise causes noise floor modulation heard as brightness, the 60 kHz HF filter helping even CD, and PCM+ versus DSD+ modes. Head-Fi, Nov 2015
• 2016-01-03Watts says Dave is insensitive to the digital source if the data is bit perfect, though the rest of a system may still be affected by RF noise and jitter. Head-Fi, Jan 2016
• 2016-01-04Watts explains that source differences heard with Hugo came from RF noise (noise floor modulation heard as brightness) and correlated ground currents degrading depth, which Dave's galvanic isolation removes. Head-Fi, Jan 2016
• 2016-01-18Watts explains that out-of-band RF noise intermodulates into noise floor modulation that sounds bright, so his DACs are limited to 50 or 60 kHz; Dave's HF filter targets DXD noise. Head-Fi, Jan 2016
• 2016-01-18Watts says the target is no out-of-band noise; Hugo is 50 kHz -3 dB, with no measurable noise from 23 to 50 kHz on CD; 1 MHz noise matters more. Head-Fi, Jan 2016
• 2016-01-27Watts heard no difference on 2Qute between its stock switching PSU and a car battery, unlike Qute, expects none with Dave, and says internal DAC RF noise dwarfs PSU noise. Head-Fi, Jan 2016
• 2016-05-13Watts gives four reasons jitter is a non-issue on DAVE, including the pulse array's jitter insensitivity, and says reducing RF noise sounds warmer and softer and is more transparent. Head-Fi, May 2016
• 2016-05-13Watts doubts DAVE is sensitive to source RF via the mains and explains how RF intermodulating in analog electronics causes noise floor modulation that sounds harder and brighter. Head-Fi, May 2016
• 2016-05-14Watts reports DAVE via headphones sounded smoother and more natural with his laptop on battery than on mains, a small change he says is consistent with lower RF noise. Head-Fi, May 2016
• 2016-05-31Watts gives two reasons adding noise to zero bits can help a DAC, small-signal non-linearity and signal-correlated DSP supply noise, and says DAVE's 350 dB noise shapers justify re-testing it. Head-Fi, May 2016
• 2016-06-21Watts says DAVE's USB isolation leaves 2 pF coupling, about 79 ohms at 1 GHz, and plans a JitterBug test to find whether laptop RF arrives via mains or USB. Head-Fi, Jun 2016
• 2016-06-27Watts plans to try a microRendu, saying reported smoothness suggests lower RF noise, though a battery source with no earth ought to inject the least RF into DAVE. Head-Fi, Jun 2016
• 2016-06-28Watts says airborne RF is no problem for DAVE because its machined aluminum case screens it, while mains and I/O connections are carefully RF filtered. Head-Fi, Jun 2016
• 2016-07-18Watts explains how source RF couples into DAVE through the USB isolator's 2 pF capacitance and ground plane, raising noise floor modulation, and why a battery-powered laptop avoids it. Head-Fi, Jul 2016
• 2016-07-24Watts says his noise shapers run at 2048 times (104 MHz), digitally filtered to that rate for RF noise and noise floor modulation reasons, making the sound smoother and darker. Hifistatement 6:13
• 2016-08-05Watts describes DAVE's power system: multistage RF filtering, a 3 milliohm output-stage supply and a discrete reference under 0.1 milliohm; he says noise floor modulation is audible to -200 dB. Head-Fi, Aug 2016
• 2016-08-10Watts says player differences depend on DAC isolation, OS and program; Chord's Windows driver resends errored packets, and with DAVE on a battery laptop he expects no differences. Head-Fi, Aug 2016
• 2016-11-19Watts says it can be done on BNC inputs but not DX outputs, without sonic benefit, as pull-ups keep unused BNC inputs inactive and their ground joins the ground plane. needs context Head-Fi, Nov 2016
• 2016-12-18Watts says ASIO and WASAPI are both bit perfect; DAVE shows no difference via isolation and battery laptop, yet ASIO sounds slightly better on Mojo, likely from less processor noise. Head-Fi, Dec 2016
• 2017-01-10Watts says Hugo 2's USB is not galvanically isolated because the USB chip would drain mobile source batteries, but RF filtering and isolation on USB power and ground are improved. Head-Fi, Jan 2017
• 2017-02-05Watts says his USB cable tests used headphones only, cable RF may affect pre-amps and power amps, and early differences vanished when he separated the USB and headphone cables. Head-Fi, Feb 2017
• 2017-02-28Watts heard almost no optical versus USB difference on Hugo 2 with a mains laptop and headphones, crediting RF-filtered USB ground and VBUS, separate input supplies and 144 dB crosstalk. Head-Fi, Feb 2017
• 2017-02-28Watts says USB needs VBUS 5V connected, but it is heavily RF filtered and only a flag to the FPGA, so feeding a clean 5V line is pointless. Head-Fi, Feb 2017
• 2017-03-04Watts explains that headphones keep source noise common mode, while a grounded power amp drives ground-loop currents through Hugo 2's ground plane, now limited by a USB RF ground filter. Head-Fi, Mar 2017
• 2017-05-17Watts says re-timing has no benefit for any source, but RF levels can matter; the Remedy would sound worse or no different in his system, so spend on music instead. Head-Fi, May 2017
• 2017-06-08Watts says galvanic isolation mattered greatly on Hugo 1, but Hugo 2's RF filtering and supply isolation (over 144 dB separation) cut its benefit tenfold; he suggests listening to USB versus optical to test sensitivity. Head-Fi, Jun 2017
• 2017-10-20Watts says not to bother, since Hugo 2's USB inputs have built-in RF filtering on both ground and VBUS. Head-Fi, Oct 2017
• 2017-10-21Watts says Hugo 2's USB has full RF filtering, he heard no change with many USB cables, doubts cable ferrites matter, and heard no optical vs USB difference from a laptop. Head-Fi, Oct 2017
• 2017-11-29Watts says USB cables sound identical or, if audiophile, add RF, brightness and noise floor modulation; USB ferrites did nothing, but 2.5 GHz ferrites sounded marginally better on Blu Dave. Head-Fi, Nov 2017
• 2017-11-29Watts explains that 2 GHz ferrites improve depth by blocking signal-correlated ground bounce from FPGA logic transitions, while random RF causes noise floor modulation and brightness. Head-Fi, Nov 2017
• 2017-12-02Watts says he heard no optical vs USB difference on Hugo 2 from his laptop with headphones, a small one with grounded outputs, and credits improved USB RF filtering. Head-Fi, Dec 2017
• 2017-12-03Watts says his ferrite tests were on Blu 2 to DAVE, USB ferrites made no difference, and the warmer, smoother-sounding ferrite with better depth is the more transparent choice. Head-Fi, Dec 2017
• 2017-12-05Watts says mains cables change sound via RF noise intermodulating in analog electronics, so random noise pumps with the signal, confusing timbre perception and making dark instruments sound brighter. RMAF 11:24
• 2017-12-25Watts says such devices could sound warmer by cutting RF or brighter by adding it; good RF USB cables all sound the same, and a 'tuned' one generated RF. needs context Head-Fi, Dec 2017
• 2018-01-04Watts says USB data is differential, so ground and VBUS can be isolated from common-mode RF with chip ferrites, which is why Hugo 2's USB closed the gap with optical. Head-Fi, Jan 2018
• 2018-01-08Watts says he heard no USB vs optical difference on Hugo 2 with a laptop and headphones, so his DPLL is sonically perfect, and the warmer source is better. Head-Fi, Jan 2018
• 2018-01-11Watts says airborne RF is unlikely to induce ground-plane currents in DAVE's thick aluminum chassis, and without ground current there is no pick-up by the analog electronics. Head-Fi, Jan 2018
• 2018-01-11Watts now attributes USB's slight brightness on DAVE to RF through ground loops and the isolators' 2 pF, not timing, since Hugo 2 USB sounds like optical. Head-Fi, Jan 2018
• 2018-01-12Watts says without the solid aluminum block, external RFI would create currents on DAVE's ground plane; the chassis screens it. needs context Head-Fi, Jan 2018
• 2018-01-12Watts says a battery-powered laptop via USB matches optical; USB's slight brightness is RF pick-up, not better timing, and vanishes when the laptop is unplugged from mains. Head-Fi, Jan 2018
• 2018-01-16Watts says with low-noise sources USB equals optical, while noisy sources favor optical; if they sound the same, USB is preferable as it covers all files. Head-Fi, Jan 2018
• 2018-01-24Watts says RF noise from the Blu-ray player likely adds noise floor modulation, making it brighter, and whichever source sounds warmer is technically better. Head-Fi, Jan 2018
• 2018-01-25Watts advises comparing a USB battery pack with Qutest's included PSU, saying a battery pack outperforms a linear PSU by eliminating mains RF noise. Head-Fi, Jan 2018
• 2018-02-12Watts says ferrites on digital signals have no downside, but on analog cables non-linear inductance makes transient timing amplitude dependent, so RF treatment of headphone cables needs care. Head-Fi, Feb 2018
• 2018-03-08Watts says it is not an issue if the PSU has effective RF filtering and the galvanic isolation works at GHz frequencies, but these effects are extremely sensitive. needs context Head-Fi, Mar 2018
• 2018-03-11Watts says switchers beat linear regulators on Hugo 1 through lower RF noise, which causes noise floor modulation heard as brightness, and advises trying a USB battery bank first. Head-Fi, Mar 2018
• 2018-03-11Watts says transformer-coupled PSUs give negligible RF isolation, so multistage RF filtering is needed; 1MHz to GHz noise matters more than 50 to 100 kHz switching leakage. Head-Fi, Mar 2018
• 2018-03-17Watts describes DAVE's development: individual RF filtering of digital sections, care over ground-plane return currents and layout, two ground planes, and Chord's patience over time and prototypes. Head-Fi, Mar 2018
• 2018-03-20Watts says complex USB gear adds RF brightness, optical is the benchmark, and gives WTA overload margin, M scaler latency (630 ms, video mode 104 mS) and 16.6-bit accuracy. Head-Fi, Mar 2018
• 2018-05-17Watts says a high-end server had better tonal balance than a slightly dark battery laptop, but he hears its RF noise clearly; the laptop's depth improvement was easy to hear. Head-Fi, May 2018
• 2018-05-17Watts says servers sound brighter with an illusion of transparency, hallmarks of RF noise, mostly via RF-sensitive power amps; a laptop feeding only Dave's USB gives the lowest RF. Head-Fi, May 2018
• 2018-05-17Watts says with bit perfect USB data only RF noise can change the sound, by creating noise floor modulation that sounds brighter; jitter has no influence on his DACs. Head-Fi, May 2018
• 2018-05-24Watts says the supplied PSU sounds as good as a battery in his setup, linear supplies often pass mains RF, and a hard bright sound signals RF-driven noise floor modulation. Head-Fi, May 2018
• 2018-07-18Watts says extra shielding will not help; ferrites, like BNC galvanic isolation, raise the RF coupling impedance between Blu 2 and Dave grounds, with the worst problem around 2GHz. Head-Fi, Jul 2018
• 2018-07-26Watts reports no difference, says Blu 2 avoided problems he had with the Hugo M Scaler, and never needed USB isolation since the noise comes from the FPGA. needs context Head-Fi, Jul 2018
• 2018-07-29Watts says optical and USB sound the same on Hugo 2 with headphones but differ once grounded; coax lacks galvanic isolation, so RF raises noise floor modulation and sounds brighter. Head-Fi, Jul 2018
• 2018-07-29Watts says optical is warmer and smoother from complete galvanic isolation, USB comes close, and coax cannot take RF filtering without degrading lock. Head-Fi, Jul 2018
• 2018-07-29Watts says sound quality can only be impaired by RF noise through the ground, which optical avoids, and source jitter is not an issue with his DACs. Head-Fi, Jul 2018
• 2018-07-29Watts says optical is perfect galvanic isolation with near-zero coupling capacitance, while every electrical digital connection carries ground RF noise that isolation coupling capacitance passes at GHz frequencies. Head-Fi, Jul 2018
• 2018-07-31Watts says it is largely immune; RF corrupting other devices is the issue, so he plugs TT2 and the source in on opposite sides of the room. Head-Fi, Jul 2018
• 2018-08-26Watts says power banks beat mains on RF noise (10 MHz to 10 GHz), linear supplies lack RF filtering, RF causes audible noise floor modulation, and batteries avoid ground loops. Head-Fi, Aug 2018
• 2018-09-03Watts recommends optical, warning against the RF noise of other inputs. Head-Fi, Sep 2018
• 2018-09-03Watts says certified USB cables sound the same, but one audiophile cable, servers and reclockers sounded brighter with poorer depth from RF noise; a battery bit perfect source is ideal. Head-Fi, Sep 2018
• 2018-10-14Watts says Hugo 2 has USB RF filtering and measured no effect from 2uS input jitter, heard no USB vs optical difference, and advises avoiding source gadgets that add RF. Head-Fi, Oct 2018
• 2018-10-27Watts says the Hugo M Scaler is not sensitive to RFI, and that Dave is more transparent while M Scaler with TT2 is better for timbre, flow and separation. Head-Fi, Oct 2018
• 2018-12-12Watts says the cable depth difference was easily heard, 350 dB is unmeasurable, any error is audible, and digital cables' RF creates audible noise floor modulation. Pursuit Perfect System 29:21
• 2018-12-12Watts says discrete parts allow a very low-noise, low-impedance reference to avoid noise floor modulation and distortion, and that RF noise, a big NFM source, needs filtering throughout. Pursuit Perfect System 34:07
• 2018-12-12Watts says every circuit point has RF decoupling, with quad-layer ground planes, a digital PLL eliminates source jitter, and TT2's noise floor is minus 178 dB with zero measurable NFM. Pursuit Perfect System 34:53
• 2018-12-12Watts says TT2's inexpensive supply was customized after trying four or five to match battery performance; batteries sound warmer because they lack mains RF noise, which creates NFM. Pursuit Perfect System 38:09
• 2018-12-12Watts says RF-induced NFM sounds brighter and is mistaken for transparency but becomes tiring within an hour; he then states TT2 gives 9.3 V RMS and five amps of current. Pursuit Perfect System 38:55
• 2018-12-13Watts says he judges improvements by checking timbre, space, detail and separation separately, and that RF noise pumping with the signal sounds bright and etched, faking transparency. Pursuit Perfect System 25:00
• 2018-12-13Watts explains 2 GHz ground-plane noise crossing the BNC isolation's 5pF; clip-on ferrites help Blu 2 but degrade the Hugo M Scaler, solid-core help slightly, cables beyond 2m don't. Head-Fi, Dec 2018
• 2018-12-16Watts says he has never seen an M scaler EMC issue, finds it odd since red is low power, and suggests repositioning, earthing via a BNC input, and moving cables. Head-Fi, Dec 2018
• 2019-02-24Watts explains that noise harms sound via ground-plane current causing RF intermodulation and noise floor modulation; optical avoids it unless mains-powered sources form an electrical loop. Head-Fi, Feb 2019
• 2019-03-11Watts says Qutest isolates USB with RF digital isolators of 2 pF coupling plus GHz filtering, not transformers, so added USB devices raise system RF noise rather than improve isolation. Head-Fi, Mar 2019
• 2019-03-16Watts says USB VBUS powers the decoder on the isolator's source side, so only ground loops let sources matter, and added source complexity can sound like improvement while degrading timbre. Head-Fi, Mar 2019
• 2019-03-17Watts says Qutest's supplied PSU is ungrounded, and SMPS have low coupling capacitance and RF filters linear supplies lack; a USB battery pack removes mains coupling, with tiny differences. Head-Fi, Mar 2019
• 2019-07-25Watts says his DPLL uses only frequency once locked, so even 2 uS of source jitter has no effect, and external flip-flops remove FPGA jitter, verified with switching stopped. Head-Fi, Jul 2019
• 2020-01-02Watts says optical is the best input because it carries no RF noise; RF entering non-linear analog electronics intermodulates with the audio, spreading noise across the bandwidth. Passion for Sound 10:34
• 2020-01-02Watts says the result is a lump of noise in the audio band linked both to the audio signal and to the random RF noise. Passion for Sound 11:15
• 2020-01-02Watts says RF causes noise floor modulation; adding RF filters to Hugo 2's USB made it sound like optical, unlike Hugo, with no measurable change, so unmeasurable modulation matters. Passion for Sound 12:11
• 2020-01-10Watts advises against adding a USB-to-optical stage, saying reported gains are RF brightness mistaken for transparency, and a Dell laptop worsened his APx555 THD and noise by about 2dB. Head-Fi, Jan 2020
• 2020-01-18Watts explains how a mains-powered USB-to-optical converter can create RF ground-current loops a DAC may intersect, advises keeping sources simple, and notes Ethernet isolation still has capacitive coupling. Head-Fi, Jan 2020
• 2020-04-01Watts says 55 kHz PSU noise is irrelevant to the M scaler; 2 pF isolation coupling at 2 GHz matters, the ungrounded supplied SMPS helps, and linear PSUs worsen it. Head-Fi, Apr 2020
• 2020-05-04Watts says USB matters through RF noise and signal-correlated ground currents, that his tests find noise floor modulation at -180 to -200 dB audible, and Chord USB filtering suffices. Head-Fi, May 2020
• 2020-05-31Watts plans to angle the ribbon mic stand for equal source distance, says amplitude encoding matters more than phase for stereo, and says RF effects depend on isolation and filtering. Head-Fi, May 2020
• 2020-06-05Watts says his listening is now very consistent, unlike the early 1980s when sound was best at 2 AM before he discovered RF noise causing noise floor modulation. Head-Fi, Jun 2020
• 2020-06-06Watts says it signals more RF injected into the DAC, raising noise floor modulation that falsely seems more detailed and increases listening fatigue. needs context Head-Fi, Jun 2020
• 2020-06-25Watts says modulated white noise brightens a saxophone, valves have local feedback, high overload margin and good RF linearity, and RF noise intermodulating with the signal causes noise floor modulation. Part-Time Audiophile 16:54
• 2020-08-11Watts calls amps that short inputs reckless, and says his MSI laptop beat a server, which sounded bright with false transparency from RF-induced noise floor modulation. Head-Fi, Aug 2020
• 2020-08-13Watts says the best source injects least RF noise into the DAC ground; a battery phone is lowest noise; the USB decoder sits on the isolation's source side. Head-Fi, Aug 2020
• 2020-08-23Watts explains optical injects far less RF noise than USB because the receiver has no processing, ferrites on optical made no difference, and it blocks source ground currents. Head-Fi, Aug 2020
• 2020-09-30Watts says his DACs use switching regulators, which sounded better than linear ones from Hugo onward, and that random RF noise, not fixed switching noise, causes noise floor modulation. Head-Fi, Sep 2020
• 2020-10-19Watts says a poster's USB versus optical difference on Dave fits RF raising noise floor modulation, though he heard only a tiny change, and says he has heard NFM below -180 dB. Head-Fi, Oct 2020
• 2021-01-25Watts says the treble extension and tighter bass heard with an Sbooster supply are noise floor modulation from more RF, and SMPS RF filters give softer, warmer, less fatiguing sound. Head-Fi, Jan 2021
• 2021-03-07Watts keeps music locally, says jitter is not a problem for his DACs but RF is, so keep sources simple, as extra boxes add RF heard as false brightness. Head-Fi, Mar 2021
• 2021-03-22Watts attributes differences between CD players into Dave over optical to mains RF noise and ground loops, and says bit-perfect source differences come from jitter, RF noise or signal-correlated noise. Head-Fi, Mar 2021
• 2021-04-01Watts says disconnecting the M scaler charger avoids the source-M scaler-Dave loop, CD players inject RF into mains, and changing Dave's PSU invalidates the warranty. Head-Fi, Apr 2021
• 2021-04-02Watts says clip-on ferrites made his Hugo M Scaler sound brighter with poorer depth, suspects resonance with its internal 2GHz chip ferrites, and recommends solid-core ferrites. Head-Fi, Apr 2021
• 2021-04-03Watts says supply ferrites are fine, any trace of noise floor modulation seems audible so he expects desktop two-box, and stock cables with a battery M scaler matched Wave Stream. Head-Fi, Apr 2021
• 2021-06-26Watts says a linear PSU only degrades or matches TT2's stock unit, since random mains RF raises noise floor modulation and brightness, and regrets not adding more RF filtering. Head-Fi, Jun 2021
• 2021-10-04Watts describes SPICE RF filter work showing random RF noise, unlike switching noise, causes noise floor modulation audible to -350dB; linear PSUs pass it, SMPS and batteries do not. Head-Fi, Oct 2021
• 2021-10-13Watts advises certified USB cables over DIY, says audiophile cables can add RF brightness, explains ground-plane RF currents and Mojo's ferrites, and suggests an optical test. Head-Fi, Oct 2021
• 2021-10-18Watts covers phone RF loops with Mojo, USB cable dressing, 2G buzz, Netflix audio, USB galvanic isolators in desktop DACs and M Scaler, and why brighter cables mean more noise. Head-Fi, Oct 2021
• 2022-02-05Watts explains Mojo 2's USB-ground ferrites, when external filters might help (optical is best), how crossfeed mimics loudspeaker listening, and reduced 2G buzz. Head-Fi, Feb 2022
• 2022-02-09Watts says switching noise from DC-DC converters is easily filtered while random mains RF passes through linear supplies, so power banks are fine and linear PSU warmth is NFM. Head-Fi, Feb 2022
• 2022-02-11Watts says audiophile linear supplies lack RF filters while SMPS need them for EMC approval, batteries off the mains avoid the issue, and active charging might explain the poster's report. Head-Fi, Feb 2022
• 2022-02-12Watts says SMPS tones do not cause NFM unlike random RF, recalls switching regulators beating linear in Hugo, and prefers power banks for isolation from RF ground currents. Head-Fi, Feb 2022
• 2022-02-14Watts says PSU break-in signals RF noise causing noise floor modulation, better fixed by RF filters, and that design knowledge matters far more than parts cost. Head-Fi, Feb 2022
• 2022-02-21Watts says putting the source and M scaler on one battery keeps source RF from the DAC; the M scaler's USB isolation exists only to cut noise reaching the DAC. Head-Fi, Feb 2022
• 2022-02-22Watts says RF only degrades low-level linearity if signal-correlated, his DACs' RF output is low and uncorrelated, and analog pre-amps add coloration rather than beating Chord's volume control. Head-Fi, Feb 2022
• 2022-02-26Watts says a working powered speaker will not damage TT2, but its ground connection could carry RF currents that brighten headphone listening, so try switching it on and off. Head-Fi, Feb 2022
• 2022-03-14Watts says linear supplies with large transformers and electrolytic capacitors filter RF poorly and add magnetic noise, so switch-mode supplies are always better for RF and stray magnetic fields. Head-Fi, Mar 2022
• 2022-03-19Watts says it depends on the system; USB can fundamentally sound the same as optical, and the issue is circulating RF noise currents. MFG Audio Reviews 57:44
• 2022-03-22Watts says the units have internal regulation and RF filtering, so better supplies make no difference or sound worse (brighter) from RF noise of big linear PSUs. article, Mar 2022
• 2022-04-12Watts says DAVE is earthed because its PSU is inside the chassis, TT2's double-insulated supply needs no earth, and earthing is moot since GHz noise is the main problem. Head-Fi, Apr 2022
• 2022-05-01Watts says Mojo 2's USB decoder has its own regulation and RF filters, he is independent of Chord, and two headphone sockets were kept for reliability and sharing. Head-Fi, May 2022
• 2022-06-19Watts says his designs already use many RF filters and separate regulators to prevent source and PSU problems, but perfect immunity may prove impossible. Head-Fi, Jun 2022
• 2022-07-18Watts calculates DAVE's PSU-induced errors at -198dB from the linear regulator and -234dB from the SMPS, saying RF noise is what changes sound with supplies, and SMPS has less. Head-Fi, Jul 2022
• 2022-08-19Watts says the M Scaler is separate because its processing creates gigahertz ground-plane noise that is very hard to filter inside a DAC, so an external scaler will always help. AVstore 29:50
• 2022-08-19Watts says that generates less ground-plane noise, making it slightly easier for the DAC, a very small difference just detectable in measurements. needs context AVstore 40:05
• 2022-09-14Watts recounts RF filters warming the sound since the 1980s, and says a null result with Mojo 2's flat UHD DSP HF filter shows the effect is noise floor modulation. Head-Fi, Sep 2022
• 2022-10-03Watts says USB sounding different from optical on TT2 is RF noise through the isolator's 2pF coupling, causing brightness, and a £9.29 5m plastic optical cable equals his glass one. Head-Fi, Oct 2022
• 2022-12-21Watts says the effect is barely measurable near the FFT noise floor and unheard by others, suggesting the Feliks amp picks up display multiplexing noise through grounding or RF. needs context Head-Fi, Dec 2022
• 2023-01-11Watts says star earthing is useless for RF noise and cannot be used for SPDIF, which needs a controlled 75 ohm return, causing lost lock or thumps. Head-Fi, Jan 2023
• 2023-02-04Watts says the streamer must be separate from the DAC because processing noise, even nanovolts of RF, degrades performance, and separation helps contain it. Dusty Audio 51:22
• 2023-02-05Watts says in 2013 or 2014 he would have chosen linear supplies, since switching creates RF noise, which causes noise floor modulation and a brighter sound. Soundnews 24:50
• 2023-02-05Watts says portables must use switching regulators, and they surprisingly sounded warmer and smoother than linears, but switching regulators need RF filters. Soundnews 26:49
• 2023-02-05Watts says RF filters go on both the input and output of the switching regulator to remove switching noise, which helps RF noise. Soundnews 27:41
• 2023-02-05Watts says SPICE lets him fully analyze filters and switching stages before building, measuring at picovolts or attovolts, and he has used it to improve RF filters. Soundnews 29:02
• 2023-02-05Watts says a switch-mode supply's regular PWM signal is easy to filter to nanovolts with a few parts, but filtering random noise needs a filter orders of magnitude more complex. Soundnews 29:53
• 2023-02-05Watts says regulator PWM intermodulates into discrete harmonics outside the audio band, but random RF noise creates in-band products and noise floor modulation; switch-mode supplies' filters help. Soundnews 30:41
• 2023-02-05Watts says, turning to power amplifiers, that audiophile linear power supplies have no built-in RF filters. Soundnews 32:29
• 2023-02-05Watts says SPICE showed filtering random mains RF from 100 kHz to 10 GHz to nanovolts is very hard; linear supplies have no filtering, switching supplies filter both ends. Soundnews 32:43
• 2023-02-05Watts says EMC and safety rules force RF filters into switching supplies, making them better than unfiltered linears, which also suffer 50 Hz ripple harmonics. Soundnews 33:28
• 2023-03-01Watts says 2pF across the M Scaler's USB and BNC isolation gives about 80 ohms at 2 GHz, letting current reach DAVE's ground plane; optical may beat USB, source depending. Head-Fi, Mar 2023
• 2023-03-26Watts says switch mode supplies have RF filters on input and output while audio linear supplies lack them; RF makes sound brighter, an artificial illusion of transparency and detail. Eric's Hi-Fi 13:22
• 2023-03-26Watts says reducing RF noise makes sound warmer, softer and quieter, allowing louder playback, less impressive but more natural; audiophiles should use car batteries as a reference. Eric's Hi-Fi 14:07
• 2023-03-26Watts says homes and streets hold many switch mode supplies, but the main worry is random RF noise from fluorescent tubes and LEDs, while switchers produce a regular PWM waveform. Eric's Hi-Fi 14:59
• 2023-03-26Watts says SPICE simulations of his RF filters show a switching waveform is easily reduced to nanovolts with a couple of filter stages like those in his DACs. Eric's Hi-Fi 15:47
• 2023-03-26Watts says filtering random RF noise to sub-nanovolts needs a filter more than ten times more complicated, and random RF noise causes the sound quality problem. Eric's Hi-Fi 16:25
• 2023-03-26Watts says random RF intermodulates in analog electronics into noise down to DC that pumps with the signal, while a regular 500 kHz component's products stay outside the audio band. Eric's Hi-Fi 16:50
• 2023-03-26Watts says eliminating random RF noise needs really complicated RF filters that nobody else builds, and asks if that answered the question. Eric's Hi-Fi 17:39
• 2023-03-26Watts says Qutest and TT2 sound the same on a car battery as on the supplied switch mode supply, except TT2 with an M Scaler; better supplies would add RF. Eric's Hi-Fi 19:18
• 2023-03-26Watts says he prefers solving problems fundamentally to selling costly supplies; DAVE predates his power supply work but has more RF filtering than TT2, and he is satisfied with it. Eric's Hi-Fi 20:03
• 2023-03-26Watts says TT2 has supercaps and RF filters, but the supercaps do not help with RF noise. Eric's Hi-Fi 20:47
• 2023-03-26Watts says supercaps help low frequency noise but not RF noise because their impedance is too high, so specialist RF treatment is still needed. Eric's Hi-Fi 20:56
• 2023-03-26Watts says claiming a supercap supply is perfect is wrong; at gigahertz RF what matters is minimal inductance, tracking impedance and capacitor connections, using extremely small capacitors. Eric's Hi-Fi 22:02
• 2023-03-26Watts says capacitors must be about half a millimeter to filter gigahertz noise; 100 MHz to 10 GHz was hardest, and this work is for the Ultima DAC's separate supply. Eric's Hi-Fi 22:44
• 2023-05-27Watts says reducing RF noise lowers perceived volume by one or two dB, since noise floor modulation sounds brighter, and is convinced a linear supply on TT2 adds RF noise. Head-Fi, May 2023
• 2023-06-21Watts says the M Scaler passed EMC tests, the reported problem is likely 50 to 104 MHz rather than GHz, and suggests trying stock or Belden BNC cables. needs context Head-Fi, Jun 2023
• 2023-08-30Watts says DAVE has RF filters on its outputs and a pre-amp adds random RF noise and noise floor modulation, so the most transparent pre-amp is none. Head-Fi, Aug 2023
• 2023-11-21Watts doubts grounding boxes, says star grounding is useless for RF, and that good ground sounds warmer while bright, artificially transparent sound means more RF noise. Head-Fi, Nov 2023
• 2023-11-22Watts says RF via cables and mains matters more than screening, stays humble about unheard ground boxes, and repeats listening tests to guard against errors. Head-Fi, Nov 2023
• 2024-02-06Watts says optical makes no difference or sounds warmer with better depth than USB, since source jitter does not matter but RF ground currents can, making Toslink the reference. Head-Fi, Feb 2024
• 2024-02-29Watts says two new products took over 5 man-years, he had no involvement in Suzi, and his new PSUs isolate RF both ways and remove modulated mains ripple. Head-Fi, Feb 2024
• 2024-02-29Watts says optical is the reference input for Mojo 2, while USB from a quiet or battery-powered source such as a phone sounds identical, given bit-perfect data. Head-Fi, Feb 2024
• 2024-03-01Watts says the new PSUs must stay independent, as their high impedance from 1MHz to 10GHz blocks RF currents that shared supplies would couple between scalers and DACs. Head-Fi, Mar 2024
• 2024-03-20Watts names three ways a source can change a DAC's sound (jitter, random RF noise causing noise floor modulation, correlated RF noise) and defines jitter as clock-timing uncertainty. MFG Audio Reviews 2:33
• 2024-03-20Watts says random RF noise harms all analog electronics; in the early 1980s his system sounded best at 2 am and sometimes bright and aggressive. MFG Audio Reviews 5:02
• 2024-03-20Watts recounts that solid-core mains cables sounded warmer, darker and smoother than stranded, puzzling given three 70 dB regulator stages and op-amps with 140 dB power supply rejection. MFG Audio Reviews 5:50
• 2024-03-20Watts explains that power supply rejection applied only in the audio band, so he concluded RF noise was changing the sound and began developing a complex RF mains filter. MFG Audio Reviews 6:40
• 2024-03-20Watts describes cascading ever-smaller LC filters from 100 kHz to hundreds of MHz; with the filter perfected, solid-core and stranded mains cables sounded identical, so the difference was RF. MFG Audio Reviews 7:30
• 2024-03-20Watts explains that random RF noise intermodulates with the audio signal, producing random in-band noise that pumps up and down with the signal, which he calls noise floor modulation. MFG Audio Reviews 8:15
• 2024-03-20Watts says random RF noise must be removed, while fixed-frequency switching noise matters less because its products sit around the carrier; with his filters fitted his system always sounded good. MFG Audio Reviews 9:00
• 2024-03-20Watts says a digital cable's RF characteristics, especially if not impedance matched, change RF noise, and RF current through the DAC ground plane to the mains creates small voltage drops. MFG Audio Reviews 17:23
• 2024-03-20Watts says ground planes have finite impedance at GHz, so RF current creates voltage drops that cause intermodulation and noise floor modulation; this current flows in a mains-source-DAC loop. MFG Audio Reviews 18:03
• 2024-03-20Watts introduces correlated RF noise as another issue. MFG Audio Reviews 19:26
• 2024-03-20Watts explains two's complement coding: near zero, a slightly negative signal flips all bits to ones, as a processor moves music data from memory to the USB output. MFG Audio Reviews 19:32
• 2024-03-20Watts says processor state changes create GHz noise bursts correlated with the signal and worst for small signals; the DAC's analog circuitry demodulates this into audio-band noise corrupting small signals. MFG Audio Reviews 20:22
• 2024-03-20Watts says correlated noise makes small signals slightly bigger or smaller than they should be, which degrades the perception of depth. MFG Audio Reviews 21:06
• 2024-03-20Watts says noise signals correlated to small signals in the digital domain degrade the perception of depth. MFG Audio Reviews 22:00
• 2024-03-20Watts says a low-power processor reduces correlated RF noise, so a mobile phone can be the best source: battery operation stops current flow and low power means little GHz noise. MFG Audio Reviews 22:11
• 2024-03-20Watts says optical blocks correlated and random RF noise; it often has worse jitter, though his Audio Precision ATS2 measured less on optical than BNC, and his DPLL removes this. MFG Audio Reviews 23:47
• 2024-03-20Watts says USB cables are costly to design; good audiophile ones sound identical to generic cables, bad ones add RF noise and sound brighter, and optical fiber type is irrelevant. MFG Audio Reviews 26:16
• 2024-03-20Watts says the switching regulators sounded considerably better, warmer and smoother, because switchers force thorough RF filtering at input and output, and begins a noisy-current example. MFG Audio Reviews 28:35
• 2024-03-20Watts says 1 amp of noisy current through a linear regulator from 1 V to 5 V stays 1 amp, while a switcher outputs 200 milliamps, limiting noise propagation. MFG Audio Reviews 29:24
• 2024-03-20Watts says zero supply impedance avoids power-supply-induced distortion, and that linear power supplies have no RF filters, letting large amounts of random mains RF noise through. MFG Audio Reviews 31:36
• 2024-03-20Watts says switch mode supplies always have input and output RF filters for EMC, fixed-frequency switching matters little, and during COVID he SPICE-modeled filters for noise up to 10 GHz. MFG Audio Reviews 33:55
• 2024-03-20Watts says his SPICE models included parasitics; a two-stage LC filter cut fixed-frequency switching noise to picovolts, but random noise up to 10 GHz needed very complex filtering. MFG Audio Reviews 34:50
• 2024-03-20Watts says the modeling matched his 1980s mains filter, built by listening into a 12th-order design with hand-wound air-cored PTFE coils, agreeing with what he found subjectively. MFG Audio Reviews 35:41
• 2024-03-20Watts says any track length is inductance, roughly 1 nH per millimeter, so resistors have RF resonances and capacitance, and leaded resistors use squeezed-on push-fit tin contacts. MFG Audio Reviews 48:12
• 2024-03-20Watts says in every listening test they sound the same because current causes the problem; without current there are no ground-plane voltage drops, and perfect ground planes are impossible. MFG Audio Reviews 1:00:30
• 2024-03-20Watts says yes: inverters are not RF filtered, and his SPICE modeling showed that eliminating random RF noise requires extremely complex filters. MFG Audio Reviews 1:01:13
• 2024-04-16Watts says RF noise scales with circuit power, Toslink lowest and USB two orders worse; he still RF-filters optical, and more pulse array elements give lower noise, distortion and impedance. Head-Fi, Apr 2024
• 2024-05-10Watts says a power supply feeds Quartet and has a very sophisticated RF filter built into it. Reactcore 0
• 2024-05-10Watts says this must be isolated from that and from the rest of the base to stop currents flowing in a loop, using two cascaded RF filters of a couple hundred components each. Reactcore 23
• 2024-05-10Watts says this came to require RF filters; the clip begins mid-sentence. Reactcore 3
• 2024-05-10Watts continues that the RF filters are very compact (YouTube captions read 'very complex'). needs context Reactcore 6
• 2024-05-10Watts says the problem is random RF noise, which he says is not correlated; the end of the sentence is garbled in the transcript. Reactcore 9
• 2024-05-10Watts says a noise at 400 MHz is not going to cause any problems at all. needs context Reactcore 17
• 2024-05-10Watts says it is random RF noise that is the problem. Reactcore 21
• 2024-05-10Watts completes the sentence: they used random RF noise. needs context Reactcore 25
• 2024-05-10Watts says the simulation started with one volt of random RF noise up to 10 gigahertz, where noise is a problem, and led to the very complex RF filters. Reactcore 51
• 2024-05-16Watts says Quartet has a power supply delivering 75 amps to the FPGAs, plus a separate power supply of his own design whose key element is an RF filter covering DC to 10 gigahertz. ConvinceMeAudio 1:31
• 2024-05-16Watts says 200-plus-component RF filters at Quartet's input and supply output keep RF out of Quartet and DAVE; stopping circulating currents makes BNC cable quality irrelevant. ConvinceMeAudio 2:12
• 2024-05-16Watts says this isolation gives much better sound quality, and that isolation through the RF filters is fundamentally crucial. ConvinceMeAudio 2:55
• 2024-05-16Watts says optical would require changing the DAC and fails anyway with DAVE and scaler both on mains, so he kept dual BNC and contained the noise inside the product. ConvinceMeAudio 10:19
• 2024-05-16Watts lists five FPGAs, two million lines of code, 2,500 discrete components and a 75 amp supply of his own design, with RF filters from DC to 10 GHz. The Speaker Shack 4:00
• 2024-05-16Watts says the RF filter has over 200 components and the same filter sits at Quartet's input and at the power supply output, controlling noise in and out. The Speaker Shack 4:43
• 2024-05-17Watts says separating it reduces processing noise and isolates Quartet from the DAC: RF filters in the supply and Quartet's input stage make it behave like a battery. AVstore 1:38
• 2024-06-18Watts says Quartet will cost more than DAVE, and describes Spice-modeled 'pinch-off' RF filters in its separate supply and input, isolating DAVE from its noise and making the sound warmer and smoother. article, Jun 2024
• 2024-06-18Watts says the supply is switch-mode because the core needs 75 amperes at one volt; a linear supply would dissipate 1.8 kilowatts and add more RF noise. article, Jun 2024
• 2024-07-12Watts says with a perfect M Scaler and DAC every bit-perfect source would sound the same, so he hardens DACs and scalers against source noise, focusing on the power supply. HIFICLUB 47:23
• 2024-07-12Watts rejects criticism that audible source differences mean the M Scaler is badly designed: it passes EMC and has galvanic isolation; Quartet is isolated further. HIFICLUB 48:49
• 2024-08-12Watts says inputs differ only through source RF noise: optical is the reference, USB has ground ferrites, and coax is most sensitive, hence galvanic isolation on the Hugo M Scaler. Head-Fi, Aug 2024
• 2024-10-03Watts says PCB ground layers screen EMC, the new coefficient work needs a whole 200T FPGA and won't reach the Hugo M Scaler, and single-ended has the transparency edge. Head-Fi, Oct 2024
• 2024-10-25Watts says balanced DACs still use ground-referenced I to V converters, balanced complexity raises noise floor modulation, and common mode rejection fails at RF. Head-Fi, Oct 2024
• 2024-11-30Watts says substrate noise matters less in op-amps than DAC chips, and explains noise floor modulation as RF intermodulation in the differential input stage, reduced by local decoupling. Head-Fi, Nov 2024
• 2025-01-13Watts says Quartet's price is unknown but it will ship with its own PSU, the only one meeting its demands after extreme RF filtering work. Head-Fi, Jan 2025
• 2025-02-14Watts says an unnamed design running at 104 MHz must filter RF to cut noise floor modulation and aliasing above 52 MHz, using a simple first-order filter; it partially works. needs context Head-Fi, Feb 2025
• 2026-03-17Watts says she heard it as warmer; the preamp rejected poorly above 10 MHz, so he suspected RF, as stranded wire creates more RF noise than solid core. ecoustics 55:38
• 2026-03-17Watts says he then built RF filters, shrinking inductors and capacitors down to hand-wired solid-core PTFE inductors and tiny ceramic caps, each step sounding warmer. ecoustics 56:33
• 2026-03-17Watts says with the finished RF filter, solid-core and stranded mains cables sounded identical, proving RF noise caused the difference. ecoustics 57:16
• 2026-03-17Watts says his system sounded best at 2 a.m. and brighter and harder with a fluorescent light on, as plasma discharge injects RF noise up to gigahertz onto the mains. ecoustics 57:52
• 2026-03-17Watts says random RF intermodulates with the music in analog circuits, so the noise floor pumps up and down with the signal, and flat random noise sounds bright. ecoustics 58:44
• 2026-03-17Watts says reducing RF noise improves timbre and instrument separation, and that noise floor modulation however small damages separation because the brain uses correlation. ecoustics 1:01:20
• 2026-03-17Watts says yes, but it is very hard: the DAC's own processing creates RF that must be isolated from analog parts, needing very advanced filtering and isolation. ecoustics 1:02:33
• 2026-03-17Watts says RF noise reaching the analog parts is the main reason digital cables change the sound, at least in his DACs, which are completely jitter tolerant. ecoustics 1:03:17
• 2026-03-17Watts says RF cannot change the ones and zeros, and starts an example with a processor or FPGA. ecoustics 1:04:18
• 2026-03-17Watts says each logic gate a signal passes creates supply noise about 50 picoseconds apart, cascading to about 9.6 nanoseconds in his case. ecoustics 1:04:26
• 2026-03-17Watts says each switching gate disturbs ground and supply, with noise extending to ten gigahertz, and a PC's fast four-nanometer logic reaches 50 gigahertz. ecoustics 1:05:17
• 2026-03-17Watts says this produces a lot of broadband noise energy. needs context ecoustics 1:06:01
• 2026-03-17Watts says the noise does not interfere with the interpolation, which is perfect. needs context ecoustics 1:06:32
• 2026-03-17Watts says the digital processing is bit perfect, but DACs are analog devices, and trouble starts when supply noise gets into the ground plane or supplies feeding analog parts. ecoustics 1:06:37
• 2026-03-17Watts says noise enters electromagnetically or through supply and ground; a USB link forms a mains ground loop between computer and DAC, sending gigahertz RF currents into the ground plane. ecoustics 1:07:21
• 2026-03-17Watts says the expensive scaler for DAVE has 'five two hundred T' FPGAs and a 75 amp supply that can create a lot of noise, which he had to keep from corrupting the DAC. ecoustics 1:10:13
• 2026-03-17Watts says he designed a separate power supply for the unit with complex RF filtering and a new RF filter architecture. needs context ecoustics 1:10:38
• 2026-03-17Watts says he wanted RF rejection down to nanovolts, actually picovolts, given the noise levels inside a processor. needs context ecoustics 1:10:54
• 2026-03-17Watts says adding low shunt capacitors and special RF cables brought parity, the battery then slightly better than the mains supply. ecoustics 1:12:50
• 2026-03-24Watts says noise floor modulation, noise pumping with the signal and caused by random RF noise, is subjectively huge: rich, dark instruments become brighter, so RF must be filtered. Audio Force 24:57
• 2026-03-24Watts says he developed a complex new 'pinch-off RF filter' to keep RF in and out of Quartet, aiming at battery sound, since lithium-ion or car batteries are the ultimate source. Audio Force 25:45
• 2026-05-03Watts says mains power brings RF noise, which creates noise floor modulation, noise rising with signal; this bright white noise pumping with a saxophone cannot be separated by the brain. Audio Force 10:37
• 2026-10-11Watts says the battery versus charger difference is very low, depending on system RF sensitivity; if heard, use a power bank into Mojo 2; forums report no losses. Distorsiuni Interioare 42:28
• 2026-10-17Watts calls one chassis a nightmare because processing noise upsets the DAC; hence separate supplies and a new two-way pinch-off RF filter of about 200 components each. needs context article, Oct 2026