finesensations
pfm Member
Any pics please?
MDAC first listen thread
- Thread starter Interzone
- Start date
- Status
Any pics please?
JohnW
pfm member
Thank you, that was Dom and myself who came up with the concept, and watchfully guided our mechanical guy Peter who at times had some very odd ideas... Sometimes it’s better not to be too creative...
Imagine the fun I had during the redesign to squeeze the CROSS output stage into the confined space of the original circuit - it took almost one month to layout the PCB for that tiny circuit section - drove me mad...
The Black Caps. on the PSU board we had designed as "specials", they are along the lines of the ultra low ESR caps. from Panasonic / Nichicon. We spent much time with the capacitor vendor convincing them to sources their raw materials from Germany and Japan - and yet still be affordable (not Panasonic / Nichicon pricing)! This is the great advantage of working in China
Not sure if there would be much of an improvement using “Standard low ESR caps”, you would have to use high voltage / high capacitance Organic type capacitors from say Sanyo to gain a decent step change in quality.
Many Thanks, its nice when someone appreciates the hard work that’s gone on under the hood 
JTC
PFM Villager...
Point taken, but that would mean extension cable time, or replacing the HD800 cables with something longer - which, if done using a 'premium' custom cable, will run to 1/2 the price of an MDAC in itself......
john
I tried converting to FLAC, which did allow Sonos to play the track but the M-DAC reported test failed immediately upon playing the track.
0rangutan
pfm Member
Just got mine from Audio Emotion today (super service) and have it plugged into my iMac and Rega Brio-R. Will test other amps later, but have had great results so far against my existing DAC, a HRT Streamer II+. The HRT has been my favourite DAC to date, replacing and beating anything else I have tried, including a Benchmark DAC1, MF V-DAC, NuForce HDP and Lavry DA10.
I have been able to A/B test the HRT and MDAC by using the OSX app SoundSource to switch outputs and the Brio-R remote control to switch inputs. This allowed for instant comparison and I have been pleasantly surprised to hear a clear improvement in the MDAC over the HRT. I'm not great at describing sound, but would go for something like better separation and control of bass. And bigger smiles.
I have also compared the headphone output with that of a dedicated headphone amplifier - the Woo Audio WA3. I used Sennheiser HD650's and Beyer DT880's (600 Ohms). In this instance it was less clear which I preferred, although both were excellent. On this basis, given that the MDAC headphone output is built in, it is a no-brainer to now sell the WA3. This was a big relief as it was only the headphone output that I had any real concerns about in advance of buying the MDAC.
I have also been able to test the bit perfect WAVs as posted earlier today. These all pass running either in Decibel or iTunes (+ BitPerfect plug-in) on OSX.
All positive so far… however, there are a few issues with the hardware that are worth being aware of:
1. USB galvanic isolation is poor. The MDAC passes through electrical noise from the iMac that can be heard through the Brio-R. The HRT has galvanic isolation built in and does not do this, so this seems a strange omission on a larger and more sophisticated DAC. Switching to using a cheap optical cable instead fixes the problem (and even then, the USB cable must be disconnected too as just having it plugged in passes through the noise).
2. The power socket and thin pins on the lead look quite fragile. The power lead does not go in very far, nor feel like it has actually been inserted properly. I miss my IEC leads.
All in all, this will replace a £270 DAC and £400+ of headphone amp, while providing better sound - a "free" upgrade. Thanks John and Dominik!
I have been able to A/B test the HRT and MDAC by using the OSX app SoundSource to switch outputs and the Brio-R remote control to switch inputs. This allowed for instant comparison and I have been pleasantly surprised to hear a clear improvement in the MDAC over the HRT. I'm not great at describing sound, but would go for something like better separation and control of bass. And bigger smiles.
I have also compared the headphone output with that of a dedicated headphone amplifier - the Woo Audio WA3. I used Sennheiser HD650's and Beyer DT880's (600 Ohms). In this instance it was less clear which I preferred, although both were excellent. On this basis, given that the MDAC headphone output is built in, it is a no-brainer to now sell the WA3. This was a big relief as it was only the headphone output that I had any real concerns about in advance of buying the MDAC.
I have also been able to test the bit perfect WAVs as posted earlier today. These all pass running either in Decibel or iTunes (+ BitPerfect plug-in) on OSX.
All positive so far… however, there are a few issues with the hardware that are worth being aware of:
1. USB galvanic isolation is poor. The MDAC passes through electrical noise from the iMac that can be heard through the Brio-R. The HRT has galvanic isolation built in and does not do this, so this seems a strange omission on a larger and more sophisticated DAC. Switching to using a cheap optical cable instead fixes the problem (and even then, the USB cable must be disconnected too as just having it plugged in passes through the noise).
2. The power socket and thin pins on the lead look quite fragile. The power lead does not go in very far, nor feel like it has actually been inserted properly. I miss my IEC leads.
All in all, this will replace a £270 DAC and £400+ of headphone amp, while providing better sound - a "free" upgrade. Thanks John and Dominik!
Hi John,
I did try several times, including pausing the track, restart test, then hit the "go to beginning of track |<<" button. No difference. It fails immediately.
For my Sonos zoneplayer the volume is set to "fixed" output.
It does indeed imply the data from the Sonos is not bit accurate. It would be nice to have proven it wasn't the conversion from WAV to FLAC though, or the way the Sonos handles FLAC vs WAV.
I will be curious if other forum members have different results to me with Sonos. I am wondering though if the Sonos is not bit accurate just at the start, throwing off the test, or continually wrong throughout.
Also John, I noticed , from cold, switching on the M-DAC the reported frequency is more jittery for the first 10 seconds or so, showing 44.095, 44.100 or 44.105. After about 10 seconds it consistently reports 44.100 and doesn't change again.
sq225917
Bit of this, bit of that
Basil, i'm well aware of when stuff is worth leaving alone...
Steven Toy
Accuphase newbie
There is ALWAYS room for improvement.
JohnW
pfm member
Hi Jon Statt,
If you PM your Email address I can Email the WAV file - I'm not sure if its not going to have the same issue with its header information - but worth a try.
You could try playing the FLAC file from your PC to confirm the File is "Correct"
It would be a shame if the Sonos is not Bit Accurate - but its useful to know.
Early days, lets see if anyone else has success.
Did you say the test passes from your PC?
John
PhilCTTE
Senior mumbler
I'll try the test later tonight with my Sonos.
Regarding 44.105 readings etc I do not get this from my Sonos I think I've seen 44.101 once.
JohnW
pfm member
It could well be the problem, like a "Fade-in", as we gain more feedback from the Bit accuracy test "usage scenarios", we can try and make it more robust - such as ignore the first half second or so...
PhilCTTE
Senior mumbler
Johns reason was that the USB isolator does a better job when not attached to the chassis.
Still playing around with my laptop USB isolator > MDAC with hi-res files 96/24 Nirvana Nevermind download. Although there is bags of detail, there's also a little harshness at top end, but this could well be the recording. Does anyone have this remastered hi-res version to compare notes ?
0rangutan
pfm Member
Hi SQ - was this suggestion in relation to the USB isolation issue?
The problem electrical noise exists just by having the USB lead plugged in, even when using optical as the actual input. The choice of filter makes no difference.
I have swapped the Brio-R out for my Bel Canto S300i and the problem is no longer audible. I guess this means that, while the MDAC passes through the noise, some amps can cope with or reject this better than others.
JohnW
pfm member
From Power-up from Cold, this is normal, the MDAC's internal Clock is not temperature compensated - it needs to stabilise at its operating temperature, this will occur within a few minutes of power-up. Once it’s reached its operating temperature it "drifts" about its "Datum" frequency by about 1ppm a day - (1 part per Million) - this is extremely good.
The MDAC's clock is designed for best short-term phase noise performance (Jitter) which is important for Audio - there’s a compromise between achieving best short-term phase noise (Jitter) over best long-term accuracy.
For best short-term clock accuracy we bias the oscillator circuit with higher current (this reduces noise) - but this degrades the longer-term "aging" of the Crystal, long-term aging (frequency drift) does not effect the sound quality (“aging” is measured in days, weeks and years).
Also due to internal PSU filtering, the PSU rails take 10's of seconds to reach there "set" operating levels - this will also effect the clock.
The DPLL BW of the PLL circuits is in the sub Hz region (Sub Hz performance is the great advantage of Digital PLL's), so these will also take some seconds to "settle".
Basically, normal - don't worry about it
John
- Status