Silkscreen bottom left says 'DAC 2' - the J3 bare connections (top right) show less-than-production-quality wiring 😱
Did you read, it was not about just the measurements? As far as I'm concerned their listening impressions are no more valid or invalid than any of the other ones I see posted here. I guess we will just have to wait for a philanthropist audiophile with something to prove to finance something like a full set of DBT using a mutually acceptable protocol with controls and training, etc. I defer to Jacob on that.
The silk does say DAC 2 but the converter is marked ES9028. The real DAC2 used ES9018.
I am 85% sure it's this part based on the markings, size, and footprint:
https://www.mercury-crystal.com/upload/product/file1_147434932599.pdf
Just to clarify, I'm not knocking these small SMD XOs like the CTS. I have used a similar Kyocera part as the ref clock for a multi-Gbps SERDES link and they work fine.
I would bet all of these small 4-SMD XOs are similarly constructed and perform around the same, too. That includes the NDK everyone loves.
Agreed, they appear to be the same markings. Interesting.
However, DPLL bandwidth stability shows if the input stream is jittered with respect to the dac clock. Could be if the 50MHz clock also clocks data out of the FPGA interpolation filter by some means (and more than one possibility for doing that), then jitter may be less of an issue with that dac chip.
Well, regardless of how they are delivering the data to the DAC, the DAC still relies on this signal to clock out the data, so I don't see how it would be any less sensitive in terms of the actual D/A conversion process.
The truth is most likely that these standard garden variety oscillators are actually pretty good and more than enough to realize the full performance of the DAC. They certainly aren't using some BS rubidium clock or an exotic Wenzel part.
The truth is most likely that these standard garden variety oscillators are actually pretty good and more than enough to realize the full performance of the DAC. They certainly aren't using some BS rubidium clock or an exotic Wenzel part.
Dreamth, I also use Nichicon Muse and Elna caps. BUT I don't put them in where I can avoid them. It is not too hard to direct couple circuits, especially with an IC based servo that generally about the same as the cap. At least you are using good electrolytic caps. You should see some that I have measured over the years. '-) I think that I can string them together and create an 'effects' box.
Attachments
Last edited:
Not according to NDK claims for the 'A' suffix parts, although they do not publish much in the way of specs. A Japanese only doc does show a nice example phase plot, but no guarantees.
So far, I am doing fine with Crystek 575. Since the last round of changes, some people are starting to say my modded dac#2 has a preferable sound to DAC-3. Interesting, but I will wait and see. Maybe some other people will have a different opinion.
Last edited:
Do you ever use protection circuits, like DC detect, beyond the servo? I'd say a power amp should have AC coupled inputs but there are lots of fashionable designs that don't.
They are probably just binned. I wonder how many 1 dollar XOs you need to buy to get yourself an NDK "A" part.
Probably binned, agreed. They go for $8 or $10 in very small quantities. About $3 each on a reel, IIRC.
I've asked that numerous times never get an answer. What if you like something like a simplistic RIAA with Vs/2 DC on the output. Well I guess the solution is to use a coupling cap on it, Duh.
Chris, of course Parasound amps use active DC offset protection. Also overtemp,oscillation, short circuit, etc.
I remember seeing your both documents years ago.I have read Jung's book on op-amps and still have it on my computer, also saved all his articles before his site shut down.
I've read many interesting studies over the years, but there's also a good body of evidence pointing in the direction that good quality coupling capacitors don't affect the sound in a way that can be heard if the values are right or at least they don't affect it in an unpleasant way when they do.
Cyril Bateman spent a lot of time making his own measuring equipment, the other link i've provided you
Nichicon Muse ES bipolar caps measured: <-120dB THD, <-140dB IMD
was about measurement on AP equipment in all audio range , which i'm not ready to discard easily as music signals are sinus waveform , their slew rate is maximum in zero crossing region where the DA(dielectric absorption) is null...
I won't refer to slew rate in clipping or on high loads with a feedback network...That is a different problem and capacitors should be out of the way when ever this is possible.Even so, some include the capacitors in the feedback loop to cancel capacitor's distortions with appropriate compensation.
I knew that Nichicon Muse caps are good enough 4 years prior to that test as i had the chance to buy hundreds of them at a a very good price in 2010.All my friends agreed with me shortly after 🙂
Attachments
Last edited:
So you don't have a proof of "that" quoted. Thanks for confirming my suspicion.
I asked him about sound, not things or perception. Zung can stand his own without your assistance, no?
Sound perception is fickle. From here The strange case of the Bristol hum - BBC News :
"But somewhere between the simple cause of wind around high towers, and alien visitation, lies another explanation. "Your brain can often fill in a sound where there isn't one," says Ze Nunes, owner of MACH Acoustics in Bristol. "You hear your mobile phone ringing, but no-one is calling you. We get called to a lot of 'fake complaints' - but it's not bogus, the person hearing the noise really does think it's there.""
I've thought I've heard my phone on a number of occasions, I'm guessing this is quite common?
"But somewhere between the simple cause of wind around high towers, and alien visitation, lies another explanation. "Your brain can often fill in a sound where there isn't one," says Ze Nunes, owner of MACH Acoustics in Bristol. "You hear your mobile phone ringing, but no-one is calling you. We get called to a lot of 'fake complaints' - but it's not bogus, the person hearing the noise really does think it's there.""
I've thought I've heard my phone on a number of occasions, I'm guessing this is quite common?
When MEMS oscillators measure like that at a resonable cost I think they will be the preferred choice due to their immunity to mechanical feedback. Yes of course digital is sensitive in the same way as a TT when it comes to feedback...
//
Last edited:
Yes, one slight noise can mis-trigger the correlator in your head, and make you think you heard a phone ring, or a voice, or... Anyone who has listened to noise on radio channel (radio amateurs, radio operators...) can tell you that now and again you will hear a voice clear as day that isn't there.
Some years back someone tried claim it was voices of the dead. Became a fad, for a while, despite the effect being well know for ages.
At least, batteries and electrolytic capacitors are the less durable components on a electronic board. A reason why to avoid them as much as possible, don't you think ?
I embrace batteries. They are easily replacable and remove a whole host of other probelms.
As for electrolytics if used correctly their life expectancy is more than adequate for audiophiles who replace everything every few years! Our cherished music has already passed through many of them anyway (many hundreds in some cases).
As for electrolytics if used correctly their life expectancy is more than adequate for audiophiles who replace everything every few years! Our cherished music has already passed through many of them anyway (many hundreds in some cases).
- Status
- Not open for further replies.