Will a USB purifier, JitterBug or USB isolator improve my DAVE, TT2 or Hugo?

What he says such devices can and cannot change

From 2014 to 2020 Watts separated jitter from noise when discussing USB add-ons. In June 2014, on the original Hugo, he wrote: "Jitter problem is solved within Hugo, so a separate USB unit will have no affect on jitter", adding that "adding extra components will always add more RF noise" Head-Fi, Jun 2014. In October 2015, in the same DAVE-thread post as his advice on purifiers, he said the issue is "absolutely nothing due to jitter - its about RF noise and signal correlated noise" Head-Fi, Oct 2015. In May 2020 he described the same two mechanisms, RF noise entering the DAC and signal correlated noise that degrades depth perception, and said both are "due to currents flowing into the DAC" Head-Fi, May 2020. From 2018 to 2022 he attributed improvements people hear from USB add-ons to extra RF noise that makes the sound brighter (Hugo 2 thread in 2018; Watts Up in 2020 and 2022) Head-Fi, Mar 2018Head-Fi, Oct 2018Head-Fi, Jan 2020Head-Fi, Dec 2022.

2014: original Hugo, isolator helped with a noisy PC

In June 2014 he said there "may be some benefit in isolating Hugo from the computer. It's pros and cons" Head-Fi, Jun 2014. Two weeks later he reported listening through an Olimex USB-ISO isolator (£33 in the UK, working only with Hugo's SD USB input, so only 44.1 or 48 kHz) from his Dell and said "it sounds very good, better depth, instrument focus and separation" Head-Fi, Jun 2014Head-Fi, Oct 2014. In September 2014 he wrote "Yes isolating the USB really helps", adding that two USB cables would not work because the grounds stay connected Head-Fi, Sep 2014. He placed the Olimex improvement in his "1 week category" of how long an improvement stays noticeable Head-Fi, Oct 2014. In October 2014 he said his Olimex tests used a Dell i7 Windows 8 machine, much noisier than an iPhone 5, and that "its very good news that the iPhone does not need isolation" Head-Fi, Oct 2014. The Olimex did not work on Hugo's HD USB input Head-Fi, Oct 2014Head-Fi, Oct 2014.

2015 to 2016: DAVE, with built-in galvanic isolation

In October 2015 he wrote: "As to USB purifiers, for Dave, Hugo TT, 2 Qute don't bother as they are galvanically isolated" Head-Fi, Oct 2015. In August 2016 he confirmed that "2 Qute, Hugo TT and Dave have the same USB decoder and galvanic isolation" Head-Fi, Aug 2016. (In those years Hugo TT means the first-generation TT; his TT2 statements are below.)

In June 2016 he set out to test whether residual source noise reached DAVE via the mains or through the 2pF coupling capacitance of its galvanic isolation, using an AudioQuest JitterBug as an RF filter, and stressed "we are talking quite small differences here" Head-Fi, Jun 2016. In the same post he said another layer of galvanic isolation at the source "would not eliminate the problem - only half it", and that the extra circuitry would create a bit more noise Head-Fi, Jun 2016. His report, from a Win 10 MSI i7 laptop, DAVE and headphones in a hotel, opened with "very much - YMWV": on mains power "adding the Audioquest Jitter Bug did indeed improve SQ", sounding smoother and warmer with better separation and focus, while with the laptop on battery "it became impossible to reliably hear the effect of the jitter bug" Head-Fi, Jun 2016. His scores were 100% for battery with or without the JitterBug, 75% for mains with it, and 0% for mains without it, and his advice was: "Use the jitter bug for mains powered USB sources - it is worth the small cost", with the best sound still a laptop on battery Head-Fi, Jun 2016.

In July 2016, in the DAVE thread, he wrote: "Using another galvanic isolator won't help and will probably make things worse." He added that "99% of the RF noise problem is removed currently by Dave" and "a single jitter-bug gets you too 99.75%", and that he was enjoying DAVE from a powered laptop without having put the JitterBug in Head-Fi, Jul 2016. The same month he said that before galvanic isolation he got "huge differences from sources if they were mains powered", whereas the remaining source effect on DAVE was "a small change that is only audible on AB tests" Head-Fi, Jul 2016.

2017: Hugo 2, little or no benefit from further isolation

In February 2017 he said Hugo 2's USB 5V line is "very heavily RF filtered", and is just a flag to the FPGA, so it "can't have an influence on the SQ" and owners need not feed it clean lines Head-Fi, Feb 2017. In March 2017 he recalled that with Hugo 1 he "heard an absolutely enormous difference when galvanically isolating it from my Dell PC; but then I used active loudspeakers and the test I did earlier with Hugo 2 was with headphones", said he could not be sure the same held for Hugo 2, and expected reports from other users about the Intona with Hugo 2 Head-Fi, Mar 2017. Later that month, in the Hugo 2 thread, he answered "Short answer: no", based on optical and USB sounding the same from a noisy MSI laptop on headphones, and said "further isolating the USB will have no effect; but of course different sources, different results are possible" Head-Fi, Mar 2017. In April 2017 he found that grounding the outputs made a small audible difference and concluded that "galvanic isolation does not seem to be necessary for headphone use if there are no ground loops", while "In desktop mode driving power amps you should get an improvement - but not by a lot" Head-Fi, Apr 2017. In June 2017 he wrote that with headphones "it is a non-issue; if you are feeding power-amps then it can make a small improvement", and proposed a test: "You can test this out by listening to USB or optical - if they sound the same, then you have no problems, and do not need further treatment at all" Head-Fi, Jun 2017.

2018 to 2019: advice to avoid USB add-ons

In March 2018 (Hugo 2 thread) he wrote "My experience is that more complex USB gear makes the sound brighter" and recommended optical as the reference benchmark Head-Fi, Mar 2018, and six days later, in the same thread, "don't bother with USB tweaks as these, in my experience, only increase RF noise making it superficially better" Head-Fi, Mar 2018. In May 2018, in the TT2 thread, he said USB sounded "very close to optical" from his mains-powered MSI laptop on headphones Head-Fi, May 2018. In September 2018, in the DAVE thread, he said the servers and reclockers he had heard "all sounded brighter with poorer depth" and "Adding extra boxes, will always increase overall noise levels" Head-Fi, Sep 2018. In October 2018, for Hugo 2, with a YMMV caveat about sources: "My advice is to avoid all source gadgets - at best they will have no improvement, at worst they will make things sound brighter" Head-Fi, Oct 2018.

In March 2019, on Qutest, he wrote that "USB devices will not improve the isolation, but will actually degrade overall performance" because overall system RF noise rises with the added circuitry Head-Fi, Mar 2019. He described how a mains-powered USB reclocker can form a ground/mains loop that impairs a galvanically isolated DAC, calling it "an example of a reclocker actually raising system noise" Head-Fi, Mar 2019, and said "adding extra complexity to the source will actually make matters worse, as you are creating ground loops that didn't exist" Head-Fi, Mar 2019.

2020 onward: current Chord DACs, TT2 and converters

In May 2020, in Watts Up, he replied to a question (not in this archive) that also involved the M Scaler: "I doubt very much it will have much of an effect. After treating the USB grounds with ferrites (started doing this about 5 years ago on all designs) the jitterbug no longer made a difference" Head-Fi, May 2020. Three days later he said he could not comment on the JitterBug directly and that "all current Chord DACs employ extensive RF filtering or galvanic isolation on the USB, so further treatment is unnecessary" Head-Fi, May 2020, then clarified that he had been referring to "the Jitterbug really there, which benefited with Hugo 1" Head-Fi, May 2020.

For TT2 specifically, in October 2020 he wrote that "TT2 is galvanically isolated and the isolation is after the USB decoder", and that "adding extra devices in the source path to my mind can only increase RF noise within your local environment" Head-Fi, Oct 2020. In October 2022, in the TT2 thread, he said that when USB sounds different to optical it is because of RF noise currents injected into TT2 "via the 2pF of coupling capacitance in the TT2 galvanic isolator", and that he would stick with optical if you can hear a difference Head-Fi, Oct 2022. In December 2022 he advised: "Don't be tempted with USB re-clockers etc, as they again increase RF noise and then give the false impression of better transparency" Head-Fi, Dec 2022.

On USB-to-optical converters his statements differ by context. In January 2020, answering a poster, he advised keeping the source simple: "Adding an extra stage (USB to optical) adds more complexity, more capability to increase RF noise, and more problems with ground (or mains) loops" Head-Fi, Jan 2020. In October 2022, in the TT2 thread, he said that if you try one, "only use one that does not require an external PSU", and "I haven't tried it, but Douk Audio U2 USB Converter looks like a good bet to try" Head-Fi, Oct 2022. At CanJam NYC in March 2024 (check the recording) he said he was "not 100% convinced about fiber over USB because they often transmit power down the cable", and that converters powered from the computer rather than a separate supply "work perfectly fine" MFG Audio Reviews 58:32MFG Audio Reviews 59:13.

What he points to instead

His repeated alternative is a battery-powered source, with optical as the reference. That preference has a dated history. In May 2014, on the original Hugo (before USB galvanic isolation), he wrote "I prefer optical, as it normally sounds smoother and warmer" Head-Fi, May 2014. On DAVE in 2016 he favored USB: in March he spoke of wanting to "close the small gap from optical to the slightly better sounding USB input" Head-Fi, Mar 2016, and in July he wrote that optical is not as good as the USB "as this has zero jitter/timing issues as the clock comes from Dave", with "Detail resolution and flow is slightly better than optical" (the scraped text drops a word in this post) Head-Fi, Jul 2016. In January 2018 he reversed: "I used to think that USB had the edge on optical - it's slightly brighter - but I was wrong about that" Head-Fi, Jan 2018; in March 2018 he called optical the reference benchmark Head-Fi, Mar 2018. In February 2019 he still wrote "Technically USB is better from the clock POV" while saying he had heard optical sound identical to USB Head-Fi, Feb 2019. In March 2022 he called his 2016 preference a mistake and wrote "Optical should always be considered the reference (best) input" Head-Fi, Mar 2022; in October 2022: "I too used to prefer USB than optical. But I realised later that I was making a mistake." and "there is no technical benefit to using USB over optical when using my DACs" Head-Fi, Oct 2022. His 2019 and 2022 statements differ on whether USB has any technical clocking advantage; they are reported side by side, not reconciled. On batteries, in 2016 he said battery operation plus DAVE's galvanic isolation gives "effectively perfect RF noise isolation from the source" Head-Fi, Jun 2016; in 2018 that "A battery powered bit perfect source supplies the ultimate digital source" Head-Fi, Sep 2018; and in February 2023 that with a battery-powered DAC and battery-powered source "you've got perfect isolation" Dusty Audio 1:06:27. In April 2021, in the TT2 thread, he reported a noisy desktop PC where optical sounded clearly better than USB (he did not name the DAC in that post), saying "with nosier sources, optical will win out" Head-Fi, Apr 2021.

What he has not said: this archive has no Watts statement on specific named purifiers such as the iFi iPurifier or UpTone ISO REGEN, no report of his own test of a JitterBug or isolator on TT2, and no comment on the Intona beyond expecting others to report on it with Hugo 2 Head-Fi, Mar 2017.

Quotes

“As to USB purifiers, for Dave, Hugo TT, 2 Qute don't bother as they are galvanically isolated.”2015-10-03 Head-Fi, Oct 2015

“Use the jitter bug for mains powered USB sources - it is worth the small cost.”2016-06-27 Head-Fi, Jun 2016

“My advice is to avoid all source gadgets - at best they will have no improvement, at worst they will make things sound brighter”2018-10-14 Head-Fi, Oct 2018

“Note that all current Chord DACs employ extensive RF filtering or galvanic isolation on the USB, so further treatment is unnecessary.”2020-05-04 Head-Fi, May 2020

Dated positions

Caveats and open points

• His June 2016 DAVE JitterBug test found an audible improvement with a mains-powered laptop, eight months after his 2015 'don't bother' advice for DAVE; in 2020 he wrote that after adding USB ground ferrites 'about 5 years ago on all designs' the JitterBug no longer made a difference, and that it had benefited Hugo 1. This archive does not reconcile these.
• The Olimex results apply only to the original Hugo's SD USB input (44.1/48 kHz) with a noisy Dell PC; he said an iPhone source did not need isolation.
• The 2015 'don't bother' statement names the first-generation Hugo TT; his TT2 statements date from 2018 onward.
• Nearly every listening report is qualified with YMWV/YMMV and was made with his own laptops and systems, often on headphones; he says results can differ with noisier sources or when driving power amps.
• The May 2020 'I doubt very much it will have much of an effect' reply answers a question that is not in this archive.
• No statements on specific named purifiers (iFi iPurifier, UpTone ISO REGEN) or a Watts test of an add-on with TT2 were found.

Topics: Add-on tweaks and modifications, Ground loops, galvanic isolation, earthing and mains