advertisement


MDAC First Listen (Part 00101001)

Status
Not open for further replies.
adamdea said:
Quite difficult to finally eradicate a problem which might well not exist, and certainly can't be demonstrated to exist. Difficult design brief.
This months sterophile contains an interesting report on the audio quest jitterbug. John Atkinson fails to identify any consistent measurable effect but thinks it makes his Mac mini /laptop combo sound better.
Now assuming John Atkinson is responding only to vibrating air molecules- how would you change the device to make the jitterbug redundant? It already has no consistent measurable effect and strangely appears to work with all the dacs he used. Whereas the much greater measurable effect of using battery or mains power made no sonic difference.

How to solve such an elusive problem?
Click to expand...

Haven't read the stereophile review you mention, but enjoyed the way you summed up the situation. What impresses me about this device (and many like it) is that it suggests you use it almost everywhere there is a usb connection to gain a benefit.....very much like 'snake oil' it promises to cure everything and you'll soon forget you spent £40 or have a laugh about it over a drink with friends...it's priced at just about the right level for this effect:).
 
rtrt said:
Hi John,

I don't know about typical combinations, as my limited observation is that most, (though not all) phono stages offer little to no adjustment.

Speaking for myself as a MM owner only, i'd like to see MM capacitance ajustability between 50pF to 250pF. MM Resistance I'd like to see adjustable between 30k to 75k

Context of these values as follows;

My turntable/arm cabling of just over 120pF - I believe this is not untypical.

My 2 current cartridges;

  • Goldring 1040 (old version of the 1042) which is specified for 150-200pF and 47k, but I've seen it suggested that it's better balanced with slightly higher resistance of somewhere just over 50k
  • Shure M97xe which is apparently significantly more balanced sounding at c370pF and 75k.

My reason for suggesting flexibility below the 'standard' 47k resistance, even though those cartridges don't need it, is that Ive seen reviews of another good MM, the Ortofon 2M black which iirc suggested the low 30k's for resistance was ideal.

For comparison my current stage is c53k resistance and I think 220pF from looking at a circuit diagram.

If it becomes a challenge then a lower capacitance and a higher resistance, at least allow the use of external plugs, to vary capacitance up and resistance down as far as I remember.
Click to expand...

I need to come up to speed on MM / MC interface... My concern is that your suggesting quite fine steps in both Resistance and Capacitance loaded, my first thoughts are that this requires quite a complex arrangement = Cost + size... Cost that might be better spent on an MC cartridge in the first instance... are there loading options for MC cartridges or just gain?
 
adamdea said:
Yes. I seem to remember there was talk of someone sending you a unit. How do you measure USB jitter- is it a variation in the timing of the packets of data sent in 8 kHz bursts? If so, Is it jitter within a burst ie in the individual bits or a variation in the overall 8khz bursts? Sorry if this is a dumb question.
Click to expand...

By using a differential probe and observing the the eye pattern as per the USB certification tests. My jitter analysis system can calculate the Jitter from statistical analysis of the Eye pattern zero crossing points

adamdea said:
Where would we look to find the problem in the dac caused by this jitter?
If I understand it correctly the theory is that the jitter could possibly cause little electrical glitches which might somehow pollute the conversion clock and/or output stage.
Click to expand...

First place I'd look is differences in RF components on the PCB - these always find there way into the analogue domain.

adamdea said:
Given the galvanic isolation between the USB receiver and the dac and the further (optical?) isolation between the digital and analog power supplies in the MDac2 then isn't one entitled to wonder how this rather indirect and (afaik) never actually demonstrated effect could be expected to change the sound of a Mdac2. Of course isolation is never complete , but how do we even know what we are aiming to isolate
Click to expand...

The isolation barriers will offer little isolation from RF energy once its within the FDAC enclosure.
 
John, I've seen some claim great things for galvanic isolation, ie it removes just about all noise. I've also seen others claim it's only a DC blocker and therefore mainly helping with groundloops. What is the situation; does GI do little for RF? Your comment above seems to suggest this.
 
JohnW said:
I need to come up to speed on MM / MC interface... My concern is that your suggesting quite fine steps in both Resistance and Capacitance loaded, my first thoughts are that this requires quite a complex arrangement = Cost + size... Cost that might be better spent on an MC cartridge in the first instance... are there loading options for MC cartridges or just gain?
Click to expand...

Cartridge loading changes the cartridge eq.

I keep wondering if all this can done as dsp eq
(Possibly not if you want to make the signal pass entirely as DSD)
 
JohnW said:
The Spec's call for 6 BNC Flexible I/O's who's function can be defined "on the fly" but I need to insure I can squeeze all the I/O onto the rear panel, if space becomes short then I'll have to drop a BNC or Two...
Click to expand...

Hi John,

Good to know you're back at home!

Now six inputs would really be great! If you're really out of space on the back panel, maybe replace 3 BNC's with 3 RCA's?
And what about Toslinks, BTW? Will there be no more left?

Keep up the good work!
Pierre
 
JohnW said:
I need to come up to speed on MM / MC interface... My concern is that your suggesting quite fine steps in both Resistance and Capacitance loaded, my first thoughts are that this requires quite a complex arrangement = Cost + size... Cost that might be better spent on an MC cartridge in the first instance... are there loading options for MC cartridges or just gain?
Click to expand...

I wasn't sure re the steps - maybe 50pF and 15kohms?

I understand what you mean re the cost/complexity, hence my other comments around adjustability not being all that common and some making use of external RCA plugs - hadn't thought of size.

Would it be possible to have something that combines the functionality of the external RCA plugs, but more internal. Something like tube rolling, or socketed op-amps, but instead for capacitance & resistance. Or maybe a better q - would that be realistic option?

MC I don't know enough about to comment.
 
rtrt said:
I wasn't sure re the steps - maybe 50pF and 15kohms?

I understand what you mean re the cost/complexity, hence my other comments around adjustability not being all that common and some making use of external RCA plugs - hadn't thought of size.

Would it be possible to have something that combines the functionality of the external RCA plugs, but more internal. Something like tube rolling, or socketed op-amps, but instead for capacitance & resistance. Or maybe a better q - would that be realistic option?

MC I don't know enough about to comment.
Click to expand...

Yes, good idea we can add a Pluging "Personality Module" of some kind where I can customises the input loading + Gain Something like a DIL14 in IC socket.

There is a slight additional complexity in that the Phono stages is truly balanced - so you will need to double up on the Resistor / Capacitor loading components.
 
JohnW said:
I need to come up to speed on MM / MC interface... My concern is that your suggesting quite fine steps in both Resistance and Capacitance loaded, my first thoughts are that this requires quite a complex arrangement = Cost + size... Cost that might be better spent on an MC cartridge in the first instance... are there loading options for MC cartridges or just gain?
Click to expand...

MC carts are indeed sensitive to input resistance most like 100R but my Koetsu likes 200 Linn carts like more ,470 I think, capacitance is more related to RF break through and doesn't need to be variable.
I would suggest the phono input has an option for a 5 pin balanced XLR this allows any mains hum and noise to be cancelled out.
Geoff
 
John, thank you for your thorough reply. I'm constantly amazed by your dedication!

JohnW said:
I'll try to include an RCA ADC input, BUT its always better to use a RCA to balanced cable as this offers a semi-balanced system.
Click to expand...

Great! RCA will be useful as an auxiliary analogue input (e.g. FM tuner)

JohnW said:
What are the most typical combinations?
Click to expand...

Check out what a few well-regarded phono-stage makers go for:

https://www.lehmannaudio.com/phono-stages/black-cube-se-ii.html#technische-daten
Above, note the hardwire slot for input impedance!

http://www.channld.com/seta/
http://www.channld.com/pure-vinyl_faq.html
http://www.aqvox.de/phono_de.html
http://www.project-audio.com/main.php?prod=tubeboxds&cat=boxes&lang=en
http://www.gspaudio.co.uk/revelation-phonostagepreamp.htm

Other reference:

http://www.hagtech.com/loading.html
https://sites.google.com/site/zevaudio/turt/digital-equalisation
 
John MC carts are just sensitive to resitive loading as MM are. They are far less sensitive to capacitive loading. Plug in using a dil socket will be ideal for both. I use a similar arrangement when I build a Paradise for clients.

You could offer a basic physical loading to damp electrical resonance and then skew the EQ in the digital domain to offer fine tuning, similar to doing a digital crossover EQ.
 
highfi said:
Check out what a few well-regarded phono-stage makers go for:
Click to expand...

Or look at the 4 box Paradise phono stage.

SQ of this parish has now built 14 of these, and they are insanely good.

If someone built a fully balanced version it might need 8 cases :D :)
 
clivem2 said:
I've also seen others claim it's only a DC blocker and therefore mainly helping with groundloops. What is the situation; does GI do little for RF? Your comment above seems to suggest this.
Click to expand...

Clive,

Yes, your second claim is correct, due to stray leakage capacitances, the Isolation (attenuation) becomes increasingly ineffective at higher frequencies - its best to consider GI as nothing more then LF / Ground loop isolation.
 
 
sq225917 said:
John MC carts are just sensitive to resitive loading as MM are. They are far less sensitive to capacitive loading. Plug in using a dil socket will be ideal for both. I use a similar arrangement when I build a Paradise for clients.

You could offer a basic physical loading to damp electrical resonance and then skew the EQ in the digital domain to offer fine tuning, similar to doing a digital crossover EQ.
Click to expand...

Simon,

What would you recommend as "Basic" MC / MM loadings?

For MC I see:-

30R

100R

470R

and for MM I typically see:-

47K / 100pF....
 
John, with all the talk of integrated MC compatibility in the FWC Fusion DAC, does this make your plans for an überphonostagein redundant? Not that I'm in any rush at all, but I was rather looking forward to the Lakewest MC preamp.....
 
John,

The initial plan was to have the phono stage built into the FWC... the Phono stage will be a pair of user add-on PCB's to the main FDAC PCB.
 
Hi John

I understand you are not a fan of the design, you mentioned mods can be made. What is your opinion of what can be done and how will it improve the performance?

I might be interested in 2 black pairs

Many thanks
 
Mission75 said:
Hi John

I understand you are not a fan of the design, you mentioned mods can be made. What is your opinion of what can be done and how will it improve the performance?

I might be interested in 2 black pairs

Many thanks
Click to expand...

What system do you want to use the amps with? I can modify them so they are truly balanced input rather then using a horrid 5534 for Balanced to SE internal operation.

If you only use SE (RCA inputs) then I can remove the 5534.... it only degrades the SQ, the AMP itself has a nice JFET input stage that's only degraded by the unnecessary 5534 opamp in the front end. LM5534 don't like leakage RF from digital products they add a hardness / brightness to the SQ.
 
Status
Not open for further replies.


advertisement


Back
Top