MDAC First Listen (Part 00101001)

[JohnW]

USB Detoxification

"Detoxification is one of the more widely used treatments and concepts in alternative medicine. It is based on the principle that illnesses can be caused by the accumulation of toxic substances (toxins) in the body. Eliminating existing toxins and avoiding new toxins are essential parts of the healing process. Detoxification utilizes a variety of tests and techniques."

I am sorry for my questions regards this Detox think. But I did not see the advantages. .....
What is the advantages to reclock the data within the transport to the fdac? The fdac will do the job anyway....

The idea of the Detox is to remove multiple layers of noise on the USB interface by various methods:-

First the incoming USB signal is RF filtered then passed to a USB Hub IC that is powered by its own Lowe noise PSU and feed by a low phase noise clock. This USB Hub IC buffer the output USB Data from the "jittery" input Data - the Phase noise of the output Data is an "attenuated" version of the input phase noise.

The process is repeated through two more USB hubs (x3 in total), at each stage the RF and input Phase noise is attenuated.

The Hub device used has two High Speed USB 2.0 480Mbps outputs. The first is directly RF Filtered and avaible as a USB 2.0 non isolated output.

The second output from the final USB Hub, RF filtered and fed to a USB 1.1 (12Mbps) Galvanic isolation device, RF filtered and then output'ed on the second USB connector (Galvanically isolated, but limited to a maximum supported data rate of 96KHz 24 Bits).

The Detox also provides a Very low noise 5V supply for Self powered DAC's and USB sections (as will be the case for the FDAC).

The internal circuits of the USB hub IC's are not designed for the lowest phase noise, so even when feed by a clean PSU and Clock they will still have an unwanted Jitter level whose phase noise will be related to the incoming Data / and the phase noise of the Host device - we will add a "Spread Spectrum" clock mode that adds Random noise to the clock to "Bury" individual spuire components.

Having in mind that the Detox offers no galvanic isolation.

The Detox offers a galvanically isolation USB 1.1 12Mbps output - perfect for the MDAC1 or any DAC if your OK with a limit of 96KHz sample rate.

And use anyone one of these Detox at the beginning and end of a USB cable?

You can do you will help improve our cash flow and it might be the final icing on the cake - but if you heard a difference by using a second Detox (which would not surprise me) it would also indicate that the Detox does not offer perfect Detox = "attenuation" of unwanted artefacts.

I've been careful to state "Attenuation" of the Detox and not 100% isolation... its not possible in the real world to have 100% attenuation as its the nature of "RF" to radiate

 

[JohnW]

I not sure I follow this as general principal- surely rise time is lower for higher frequency waves?

The faster the Edge the higher frequency bandwidth required to transmit this edge - if you add a LPF then you filter the Higher frequency response of the system and reduce the edge speed = slower edges.

That would at least make sense as a measurement artefact, but it would not speak highly of Paul miller's intellectual honesty.

Paul makes errors, but we all do - I've worked with Paul and like everyone has his own beliefs, but I've never had call to question his honesty.

I've stared at the eye pattern and I'm still not sure that there's that much difference between the 90% and 10% points. Did you see it on the link I posted; I'd be happy to email it to you as PDF if you'd like to look at it

I looked at the Eye patterns and at around the 10% portion there is a Flat step portion - its very easy for the scope to have trouble deciding the exact position to start the edge measurement - what's more important is the edge speeds around the logic level transition points...

 

[JohnW]

Will it have power on/off switch in front panel?

Tim, if time allows then we will add a standby switch (preferred), but its hard to meet the new European Standby current requirements.

These Standby requirements might not apply to the FDAC as for optimal performance we require to keep the internal clocked circuits operational at a stable temperature.

 

[JohnW]

I received an Uptone Regen on Friday, performed some listening yesterday and will continue today.

I wont talk about my impression so far, I'll post after some more listening.

BTW John, if it is of use I am quite happy to send the unit through to you to measure / evaluate. I dont know if that may help at all with the design of your unit, but it would be of interest to me to see what these items are doing (or not )

I'd be VERY interested to measure a ReGen unit - and for comparative purposes

What concerns me about the whole subject is that no one publishes real measurement, only wishy washy ideas - but no real means to back up there claims and ideas.

Also interested in borrowing a Jitter Bug so I can measure its RF attenuation performance....

 

[JohnW]

When trying to reduce RF generation and emission, wouldn't it be a sound engineering target to aim for the slowest possible rise times?

For high speed Data rates you need High speed edges...

When I talk RF am referring to unwanted "leakage" RF that's not related to the transmitted USB Data - CPU clock, PSU switching noise etc.

 

[radamel]

I'm in for the Detox too.
Will use it with a 8200CD.

 

[JohnW]

As said various software coding techniques and hardware design/choices claim to affect jitter, so people go to all sorts of lengths to minimize this. But isnt this all moot once we're dealing with async usb.

The incoming USB Data is varying in time around its expected "time slot" the frequency and amplitude of this time variance is the encoded phase noise. When the USB device in the DAC processes this data it radiates this noise into the system.

So am I understanding correctly that the main focus of the detox is the filtration of rf energy emitted from source components, over electrically coupled usb interfaces.

The Detox prime reason for exsistances is to attenuate the "unwanted" RF and Jitter pattern of the incoming USB Data, provide a low noise PSU and also offer a galvanically isolated USB 1.1 output.

Ive always had the idea of dual spdif with async capabilities. Would solve the jitter and radiation leakage.

Yes an optically isolated "Clock Locked" SPDIF source is the the perfect solution, but only for single sample rate applications as you are left with the problem of how to command the sample rate change / Master clock rate change. Worked well in the days of 44.1KHz CD only systems but not in our modern multi samplerate world.

 

[JohnW]

hi,

my mdac started to transmit current to my speakers and im getting shocked when i touch the speaker screws.

im using the mdac with krk g6active monitors. with either trs or rca i get an electrical shock when i touch the speaker screws. no shock with mdac detached.
i have an apple tv on optical and usb -lightning connector upstream.

any idea whats wrong?

best regards
bubba

Sounds like Earth leakage current.

Disconnect everything connected to your MDAC - but leave the MDAC connected to your Speakers - I suspect you will not have the problem (The MDAC's isolated from Earth).

Apple products have HORRID earth leakage currents - I suspect when you start connecting the inputs to the MDAC you will find the troublesome device.

 

[JohnW]

Must be a very subtle difference if you didn't notice/decide on first listen. This is why measurements by JohnW are crucial. Subjectivity vs objectivity. Please report back after further listening...especially if it 'blows your socks off!'

Really keen to measure a ReGen and Jitter Bug, hoping someone here can loan me there units - and I'll insure they will at the ahead of the queue for the first handbuilt Detox units

 

[PhilCTTE]

Is the detox unit to be a separate device for the FDAC also? If so then add me to the list John. Thank you. Things have moved in a little here, currently looking at a family home with room for hifi and av . House sold just need to find the room for proposed kit Some of the older Victorian properties around Worcester have some interesting cellars that could have potential . But the markets gone mad , as soon as they are on the market they get sold

 

[JohnW]

Is the detox unit to be a separate device for the FDAC also? If so then add me to the list John. Thank you. Things have moved in a little here, currently looking at a family home with room for hifi and av . House sold just need to find the room for proposed kit Some of the older Victorian properties around Worcester have some interesting cellars that could have potential . But the markets gone mad , as soon as they are on the market they get sold

Yes, separate unit device - optimised for the MDAC / FDAC With its external optical Clock-Lock link.

 

[sprint]

Originally Posted by JohnW View Post

Typically RF tends to harden the sound and add a "Glassy" sharpness and sibilance which we tend to associate these "signatures" as sounding "Digital".

The lower Jitter tends to widened the Left / Right sound stage while firming up the systems low end, but how USB Frame packet jitter impacts the sound is very dependent on the System / DAC.

What the Detox will do remains to be "heard" But at least I can be sure we have given the MDAC / FDAC the best quality USB signal - and in the case of the MDAC, a galvanically isolated USB port.

The USB 1.1 port with its galvanic isolation will reduce "background" noise and hum loops if present in the system when connected to the PC or when the Laptop is connected to its AC mains adaptor.

I believe that when the Spread spectrum decorrelator mode is enabled the system will lose an extra layer of hardness and sound more "Organic / Analogue".

Sorry if this sounds very vague and "HiFi babel" - but when you hear the effect for yourself then you will understand what I'm trying to describe.

John

I am using a Vortexbox connected directly to a MDAC via a USB cable. The Vortexbox is basically a headless computer so I assume it is likely to benefit from one of these recently developed gizmos in the form of a Jitterbug, Decrapifier, ReGen or your own proposed Detox units?

There are various reviews available for the Jitterbug, Decrapifier and ReGen promoting the benefits of these devices some claiming that using both the Jitterbug and Regen together can yield further improvements http://www.audiostream.com/content/...est-jitterbug-schiit-wyrd#hLzkTOrXbxqmIHdK.97.

You are proposing your own Detox unit, is it expected to provide the same benefits that seem to be achieved with the current crop of devices or will that only be known once it has been designed, built and tested/evaluated? Is the proposed design of your Detox unit expected to provided the combined benefits of both the Jitterbug and ReGen in one unit? Are there other improvements/benefits that you feel your Detox unit will have/bring?

Bump

 

[wintoid]

Ugh, so the Detox can't cope with the data stream for DSD?

Also, does the Detox remove the need for a USB hub with the Squeezebox? I assume so.

 

[JohnW]

The Detox will apply a multitude of techniques to cleanse the USB Data:-

1. An RF Filter on the USB input and Filtering between each USB Hub stage.

2. 3x Cascaded Hub devices, with each Hub attenuating the Jitter and RF from the Host Device

3. The USB data packets are synchronised to the MDAC / FDAC internal Audio clock via the optical clock-lock link (Optional, can be operated without the clock-link for other DAC's).

4. Ultra low noise PSU for a Clean 5V output for "USB" powered DAC's & USB powered USB input circuit (FDAC).

5. Spread Spectrum clock mode to decorrelate any remaining jitter spurie products

6. Galvanically isolated USB 1.1 port for upto 96KHz sample rates.

7. RF filtered USB 2.0 480Mbps port

 

[JohnW]

Ugh, so the Detox can't cope with the data stream for DSD?

Also, does the Detox remove the need for a USB hub with the Squeezebox? I assume so.

The Detox will work with DSD data (and any Data type) - its operates transparently (its indifferent) to the actual type of USB data - USB Data's just Data

 

[wintoid]

Great, OK so perhaps I haven't fully understood, but is 96kHz 24 bit enough bandwidth to cope with DSD, or is that limitation only for PCM?

 

[Ciunas Audio]

I believe this is what's happening in Paul Millers review of the Jitter Bug

Yes, I think so too & he just got the numbers the wrong way around - 16nS without Jitterbug & 20nS with (which is on the border of failing USB FS compliance - the device speed he was testing).

What this implies is that the reduction in RF noise is more likely to be the key to the SQ improvement & the slower edge rates are a secondary phenomena

I cannot see edges faster then 100pS causing data failure - in fact I can see it reducing the jitter on the logic level discussion points. I suspect the 100pS "Limit" is just for EMC reasons?

Perhaps - I don't know the thinking behind it

 

[Ciunas Audio]

The incoming USB Data is varying in time around its expected "time slot" the frequency and amplitude of this time variance is the encoded phase noise. When the USB device in the DAC processes this data it radiates this noise into the system.

Apart from the jitter embedded in the USB datastream, the packetised delivery system itself will result in noise spikes as these packets are handled by the receiver logic - the 125uS microframe timing in HS USB2 will give rise to a 8KHz noise spike - this has been measured

 

[JohnW]

Great, OK so perhaps I haven't fully understood, but is 96kHz 24 bit enough bandwidth to cope with DSD, or is that limitation only for PCM?

DSD has a sample size of 1 bit and a sample rate of 2.8224MHz data rate is 2.8224Mbits/sec so this is equivalent to 16 bit PCM at 176.4kHz.

The Detox will have two outputs, 1 Galvalically isolated USB1.1 which can handle sample rates upto 96KHz and the other non isolated port suports the full USB 2.0 480Mbps rate.

 

[JohnW]

Apart from the jitter embedded in the USB datastream, the packetised delivery system itself will result in noise spikes as these packets are handled by the receiver logic - the 125uS DSD has a sample size of 1 bit and a sample rate of 2.8224MHz. In other words the data rate is 2.8224Mbits/sec. This is equivalent to 16 bit PCM at 176.4kHz. timing in HS USB2 will give rise to a 8KHz noise spike - this has been measured

With the MDAC / FDAC the USB Microframes will be locked to the Audio Master clock so will not generate non synchronise noise.

I've only seen the 8KHz frames on USB devices with poor clock recovery (Non ASync) such as Tenor, TI PCM290X / 270x and Cmedia.