Fierce-Freak, it didn't for me. Strange there can be so much differences in experiencing different module combinations.fierce_freak said:
Bgt said:
It's not so strange, there might be thousand technical reasons why the Opus with ASRC would have to sound better, but in the end, our ears decide.
I keep the metronome a hand to play with, but I end up removing it every time I listen longer.
Without the ASRC, the system simply sounds more natural to me.
A flea powered Tent oscillator in the 8804 receiver turned out to be the biggest improvement in my setup. (up to now)
I ordered two more DAC modules to be able to try out dual mono...
Let's hear what that brings, it should be at least equally good as the stereo config. Hope the difference will really be audible.
I just ordered the single USB DAC kit for now. Later I may add another module and a Metronome. Just depends if I decide this is good enough or if I want to wait it out and grab the Sabre.
Now hopefully the wife doesn't kill me 🙂 Better start my obituary!
Now hopefully the wife doesn't kill me 🙂 Better start my obituary!
neb001 said:
Pretty easy, TwistedPearAudio board are top quality, so they can take a fair bit of soldering heat.
No pictures yet, will try to make some after the weekend...
An externally hosted image should be here but it was not working when we last tested it.
That's the way I did it.
Since I use a 2x15V transformer, I can take the input of the voltage regulator (pin 3) to supply the voltage to the flea. (has to be at least 18V DC).
On the receiver, I removed the X4, C7 and C8.
Bottom of C7 is connected to XIN (pin 11 of WM8804), so there goes the output of the Tent XO.
Top of C7 is ground, same ground as I need for the Flea+Tent.
Glued both boards to eachother with thick double sided tape.
Total hight is still ok to stack an DAC-board...
btw, the output resistor of the XO has to be omitted when using the Fleatent with the 8804...
Nice home-made Flea pcb! It's a shame they are no longer making them, as people are always asking after them.
I use a modified Flea + Tent XO for my AD1896 ASRC, and would suggest anyone using a metronome upgrades the clock with a high performance clock.
I use a modified Flea + Tent XO for my AD1896 ASRC, and would suggest anyone using a metronome upgrades the clock with a high performance clock.
The clock in the metronome is already a high performance low jitter clock, so just be sure if your going to bother with other clocks, that it is "actually better". 🙂
Cheers!
Russ
Cheers!
Russ
Russ White said:
Yups, I knew that, but I'm not using the metronome anymore (at least not in the S/Pdiff-CD-player config), so I needed to give the 8804 a little upgrade...
Spartacus said:
No offense meant here, but you will need to back up that assertion to be taken seriously. 🙂
Russ White said:
The active part of the XO is on-chip, therefore it's sharing it's supply - which is critical - plus there may be issues with cross-conduction.
A good XO needs care to isolate it from any possible source of interference.
Spartacus said:
The jitter performance of the XO we use is well documented, and it cannot be called anything other than excellent. 🙂 If you have some proof that its not living up to its specs, I would very much like to see it. Seriously.
I have used it as the manufacturer recommends, so I expect it to perform as it is specified. If it is performing as designed, it is far more than good enough for the application. if it is not, I really want to know, and I will need some hard evidence. All tests here have shown it to be very much up to spec.
Cheers!
Russ
The XO shares it's supply with the rest of the circuitry. Anyone with experience in this area will recommend that you supply the clock with a dedicated supply.
The Crystek part is specified for low jitter only above 12KHz. Do you have any evidence regarding how it performs below that - in the audio band? The Text XO is specified from 10Hz upwards.
The Crystek part is specified for low jitter only above 12KHz. Do you have any evidence regarding how it performs below that - in the audio band? The Text XO is specified from 10Hz upwards.
Spartacus said:
This may be a valid point when you have mixed analog and digital supplies on the same board, but here we do not as it is all digital, and there is sufficient decoupling (via a ferrite and cap) to work very nicely (as has been tested).
Please explain why (probably random) jitter performance < 12khz is important for a 24mhz clock, especially given the operation (jitter rejection) of the SRC4192 and the DACs (Opus or COD). They are all pretty well immune to random jitter (if there is any).
In any case it is up to you to prove that its performance is poor in that regard. If you do, then I will give you full credit.
You made the assertion, not I. You will not to provide more than the anecdotal evidence so far presented.
Cheers!
Russ
Russ White said:
I can't prove that the supply in this instance is less than ideal without testing, I'm rather talking from experience in similar scenarios where I have tested.
Because that's the range we hear in.
The are most certainly not immune to jitter. Where did you get that idea from?
You yourself are claiming good jitter performance for your design, yet have provided no proof at all. I'm talking from experience and understanding of underlying principals.