MDAC First Listen (part 00100101)

[davidjt]

I finally managed to open my MDAC and can also confirm that the sound quality with the case removed is superior...warmer,more seductive sound.I'm very pleased and will not replace the case until later in the year when I receive the MDAC2 from John.

+1! After listening for a while, there's definitely no going back.

Which means there'll be a significant decision to be made further down the road.

Meanwhile, I'm enjoying my MDAC1.5.

 

[Stunsworth]

John, 4th instalment just sent.

Thanks for all the hard work!

 

[ti33er]

...soooo just thinking, if it is indeed RFI reflections from the aluminium case, won't we just have the same problem if Mr Westlake gets alloy chassis cases engineered in China?

Really hoping that certain circuits/chips can be shielded somehow to prevent this from happening regardless of casement.

 

[davidjt]

I trust John to come up with the best solution, and presumably in a full-size case he'll have more of a fighting chance than in a tightly packaged (if well engineered) little box.

The more I listen, the more I wonder how much of the so-called 'digital' sound that can be tiring or unnatural is actually due to this bleed-over (if that's what it is) into the analogue domain.
I'd always found the amp I'm using (temporarily) too bright and harsh when cranked up, which is why I replaced it. This morning it's sounding fabulous - so clearly it's 'crime' was to amplify every bit of the signal fed to it, warts and all.
Not for the first time, I've been looking for improvement in the wrong place.

So I'll leave it to the expert.

 

[Bigbobby666]

Hey John... I PMed you. Did you get?

 

[erglis]

Hi John, 4th instalment paid. I suppose I will go for L3.
Please ask Renata to change the color of my MDAC2 to gray. Thanks!

 

[LittleToast]

Thanks John for the reply. All the best for the projects and look forward to the email!

Btw I am trying a new lossless streaming service called Tidal. It is much better than Spotify Premium.

LittleToast,

The MDAC2 PCB has been issued to the vendor today so the MDAC2 project is progressing.

I'll compose information on the latest project status combined in the Email requesting the L2 payments.

Over the weekend I started work on the VFET amplifier so I'd just like to spend a few days working on the design and then once I'm happy with the design progress I can then temporally put it to oneside and work on the MDAC2 L2 write up / L2 payment request.

After the marathon effort to complete the MDAC2 PCB I just need to clear my head alittle - The VFET amplifier is a refreshing break - refreshes the creative portions of my brain

 

[JohnW]

Hey John... I PMed you. Did you get?

Rob,

Yes I received your PM , Renata's replied to both of your Email address we have on our records - maybe if you check your spam folder

 

[JohnW]

Hi John, 4th instalment paid. I suppose I will go for L3.
Please ask Renata to change the color of my MDAC2 to gray. Thanks!

erglis,

Thank you and noted, I passed your post to Renata to update our records - we will also confirm ALL details once we proceed to production.

 

[JohnW]

I trust John to come up with the best solution, and presumably in a full-size case he'll have more of a fighting chance than in a tightly packaged (if well engineered) little box.

The more I listen, the more I wonder how much of the so-called 'digital' sound that can be tiring or unnatural is actually due to this bleed-over (if that's what it is) into the analogue domain.
I'd always found the amp I'm using (temporarily) too bright and harsh when cranked up, which is why I replaced it. This morning it's sounding fabulous - so clearly it's 'crime' was to amplify every bit of the signal fed to it, warts and all.
Not for the first time, I've been looking for improvement in the wrong place.

So I'll leave it to the expert.

David,

I suspect the problem is the amplifiers "tolerance" to RF products - I'm pretty sure the MDAC's level / pattern of RF leakage changes when the chassis is removed - its the most logical explanation to the differences we've heard.

https://dl.dropboxusercontent.com/u/86116171/Resonances and Repercussions.pdf

Above is a LANDMARK article on the effects of RF on power amplifiers - read Paul Millers description of the "Sound" of RF on amplifiers performance After your experiences with removing the MDAC chassis - does this "Sound" familiar (Pun intended ).

By its nature, there will always be a certain level of RF leakage from a Digital device and its why I intend to use a Tube on the VEFT amplifiers Front end. As Paul notes in his article, tube's tolerate "unwanted" RF (unlike transistors) without in my experience sounding hard and bright / the words "Smooth & natural" spring to mind

I've always found that RF leakage from "Digital Devices" result in the Fatiguing "Digital" harness I so dislike!!! Roll on Tubes!

I'd bet you that this reason some have reported better results with an external preamplifier - due I suspect to the "filtering effect" of the Pre-amplifier.

 

[JohnW]

Okay, I'm a little confused - is the analogue board now done so I can send in my combined 4 payments?

ljmac,

As I explained in earlier posts, the PCB has been issued on Monday - but we have not sent official "L2" payment requests as I've dived into the design of the VFET amplifier - I'm more concerned about progressing the VFET design then requesting funds as many PFM'ers have very kindly sent the L2 installment without waiting for an "Official" Email (Once again a Big Thank You as this removes pressure and I can instead work on the VFET amplifier design)

I promised that the VFET amplifier project would not impact the MDAC2 timeline, so I have to squeeze the design of the VFET amplifier whenever the MDAC2 project allows.

You can join the MDAC2 L2/L3 project by selecting the combined x4 development installment payment option via our very crude webpage

Once I've progressed the design of the VFET amplifiers output stage, then I'll feel less pressured and will spend time writing up the last MDAC2 progress and developments and "officially" request the L2 - it would be nice to have the MDAC2 PCB back from the vendor so I can include pictures of the bare board in the project update

I'd be happy to see you join the project I just dread writing - developing the design the VFET amplifiers is easier for me then having to write... and VFET amplifier owners are just as keen as I am to get there amps (I plan to use the "Reference grade" VFET amps to optimise the MDAC2 sonic performance and develop the Fusion bulk foil resistor "upgrade" options).

 

[davidjt]

David,

I suspect the problem is the amplifiers "tolerance" to RF products - I'm pretty sure the MDAC's level / pattern of RF leakage changes when the chassis is removed - its the most logical explanation to the differences we've heard.

Thanks for that, John. All interesting stuff. I would pass on your suspicions to Linn, but when I made a few helpful(?) suggestions about two shambolic 'Linn Lounge' events I'd attended, they didn't take them that well, so maybe I won't.

 

[adamdea]

David,

I suspect the problem is the amplifiers "tolerance" to RF products - I'm pretty sure the MDAC's level / pattern of RF leakage changes when the chassis is removed - its the most logical explanation to the differences we've heard.

If this is rf in the signal output (not radiated from the M-Dac via the air, surely) will it be common mode or differential mode? I was wondering whether it would be reduced using the differential output from dac chip down to the power amp via XLR.

Also, if it is rf in the output couldn't it be filtered? Surely this would be easy if we could identify the rf on the output.

I'm finding the notion that rf on the output is improved by not having a metal rf shield a bit difficult to grasp.

 

[tull_hifi]

Hi John, 4th instalment just sent.

I follow the thread nearly everyday. Thanks for this great project. L2 ordered, but i'm thinking about L3 for the analog converter now

 

[JohnW]

Adam,

I believe is conducted emission via the RCA / XLR audio interconnects.

I can measure the levels with and without the chassis - this would be a good start.

I will not say I'm an expert on such matters, only enough to pass CE tests.

Filtering can always be added, but bead inductors are VERY bad for sound quality, and R/C filer would only increase the output impedance...

RF filtering is a double edged sward, while its possible to filter RF, I've never seen it not impact the sound in a negative sense due to the extra components added to the signal path... I insure we comply to CE, but add no more RF filtering then required - due to this catch 22 situation.

Filtering is only attenuation, at what level and across what B/W do you require?... I cannot answer that as there are endless system possibilities.

As MDAC conforms to CE certification requirements, RF emission are low* - considering this, maybe more thought should be given the Amplifiers tolerance to RF, most designs are poor in this respect.

* But within the context of an HiFi Audio system, what's to be considered "Low"?

An EMC "pre complacence" test house will just add inductors and capacitors to the outputs (etc) with no consideration to the effect on Sound quality, while these will filter the RF, the audio performance will be negatively effected, you need to strike a balance between RF filtering and "Simplest" signal path without really knowing where the datum is as its so system dependent.

Adding a metal plate near any radiating circuit will the capacitively couple this energy into other parts of the circuit - it will act as a "Bridge". the Chassis can never be "Grounded" with a satisfactory low impedance to all circuit nodes - at RF the chassis will have a high impedance to local circuit Grounds and then just act as a radiating surface...

What is required is that each circuit block is shielded within its own screened box section which insure a very low "localised" Ground impedance shunting the circuits radiated energy. This then dramatically reduces RF energy that can then be coupled into the effectively floating (at RF frequency's) Chassis.

 

[adamdea]

Adam,

I believe is conducted emission via the RCA / XLR audio interconnects.

I can measure the levels with and without the chassis - this would be a good start.
.....

What is required is that each circuit block is shielded within its own screened box section which insure a very low "localised" Ground impedance shunting the circuits radiated energy. This then dramatically reduces RF energy that can then be coupled into the effectively floating (at RF frequency's) Chassis.

Thanks, John. I hope you managed to find a solution with the existing boxes.

 

[JohnW]

Thanks, John. I hope you managed to find a solution with the existing boxes.

MDAC2 has several advantages - for starters the "Digital" Digital sections are on a separated PCB - the "Analogue" PCB is very clean when compared to the original MDAC design - if we start on the presumption of conducted RF via the Audio interconnects, then Lab measurement will be the real judge

We can use the conducted RF emission spectrum of the MDAC as a baseline and see where MDAC2 is by comparison.

The MDAC uses a 84MHz Clock, while the MDAC2 56MHz so the main spurie components will be in different locations, but we can still compare energy levels.

 

[Bigbobby666]

Rob,

Yes I received your PM , Renata's replied to both of your Email address we have on our records - maybe if you check your spam folder

Cheers John.

 

[ljmac]

Hi John,

Just sent my payment (feeling the pain of the current $AU exchange rate though ). I've really become quite fond of my MDAC - not just for the excellent sound per dollar, but also the wonderful flexibility and ergonomics, especially the excellent volume control and selectable filters (I know you recommend your own DSP filters, but the slow roll-off filter sounds just right in my system). It would be great to turn this little baby into a state-of-the-art DAC for a fraction of the cost of any alternative.

I'm also quite partial to "home-made" hi-fi (as long as I don't have to build it myself!): my speakers are custom built to my specs using insane components and materials that would be unaffordable on commercial speakers e.g. Duelund CAST caps and resistors, external crossovers, and a cabinet that weighs 32.5kg, even though it's only 42cm tall! Also, while my amp was commercially available at the time I bought it (a Pass Labs Aleph 3), its descendants are now Nelson Pass' "kitchen table" effort, First Watt.

Regards,
Lee

 

[Rjpcardoso]

Hi John,
Can you please let me know when is the deadline for people to join the mdac2 and help to support the project?
Thanks,
Ricardo