I am not familiar with the hat this was pulled from.
Certainly for the reference voltage for chips like the AK4490, you want a circuit that reacts blindingly fast to any perturbance. Interaction is of the essence. Hence the suggestion in the spec sheet to use AD797 as the basis for voltage regulators supplying this voltage.
You and Demian have of course hit the most bent nail on the thread squarely on the head. I would be willing to bet that, on most DACs there is more to be gained respinning the PCB than any amount of glomming stuff that doesn't fit on there.
What's the recommended layer count for a DAC these days, I know for precision work it was 6 layers back in the early 90s but not looked at things for a looong while?
Completely nonsensical. There are small perturbances, which need small quantities of energy to correct them, delivered fast, that may hamper the functioning of a DAC. Where would all that energy in a supercap have to go?
I used one of those in 1989 during the dev of a high resolution panel indicator with multiple user selectable inputs. The HP Journal did a big write up on it at the time. It used a multi-slope A-D and it’s using that meter that I had to quickly learn to get thermocouple effects under control when taking readings.
HP used a Josephson junction array to verify the performance during development - details in the journal article Josephson voltage standard - Wikipedia
As an aside, insofar as noise is concerned, the 7815 is quieter than a 317 at 15 V out - about 30 uV WB noise vs c. 450 for the 317. You can get the 317 noise down below 100 uV but you then need to decouple the adjust pin (‘heroically’ as Self says, so 100uF or something like that).
Last edited:
I’m using a similar layout but not with the output devices inter-digitated. I run the power ground and supply rail to the OP devices on top of each other - small loop area.
Are you saying you never looked at stability of power supplies in any of your designs?
I don’t know that stability is a critical issue in audio as long as it’s well <<< 1/f it should not be a concern. A good cap multiplier on the output of a 7815 gets you down to nV/rt Hz noise wise. The DAC will like that - any opamps will add their PSRR so RTI you are talking about << 1 uV wide band noise on a 2V FS OP so you are in -130 dB territory.
Have you considered that without the battery or supercap your a PSU may just have been sub-optimal (it happens - part of the dev process) and adding the battery or supercap simply cleaned it up? Might be an idea to redo the experiment but with additional controls.
Have you considered that without the battery or supercap your a PSU may just have been sub-optimal (it happens - part of the dev process) and adding the battery or supercap simply cleaned it up? Might be an idea to redo the experiment but with additional controls.
+1. And the problem with his followers, and other nonsense believers, is that they are not open to rational argument. Nothing you say can get them to think.
I think he even does not understand the terms used, their meaning. It is difficult, then, to continue in any kind of discussion with him, as it is reverted to expressions like not understanding that "dynamic nature of the music signals" is in fact a trivial stress and that a step change covers all the behavior of the tested regulator - just an understanding of control theory is needed, which can be hardly substituted by listening experience
I wonder if JN tunes his motion control devices by listening... Ah, the sound of tons of steel colliding - must need a tweak!
In other words, the wrong term. If you say "stability" to an engineer he jumps to the conclusion that you are talking about stability. I know this seems quite unreasonable to non-engineers, but that is how our brains are wired.mmerrill99 said:
Some people come here expecting worship; they rarely stay very long. Others simply want agreement to everything they say; maybe offline they have sufficient status or have induced sufficient fear in others that this is what they usually get? Some people complain about argument, but then spend most of their time arguing. Experience tells me that those who are the most persistent in spreading FUD are the most likely to be commercially involved in audio at the supposed 'high end'; their income depends on FUD being accepted, because good engineering is cheaper.
20 years ago (and for the 50 years prior to that) vaccines were seen as a saviour. not any more, regrettably.
Vaccines: Low trust in vaccination 'a global crisis' - BBC News
Vaccines: Low trust in vaccination 'a global crisis' - BBC News
You have lost me, where is speed impeded and what function would be hampered? And please define nonsense?
Also, tell it to an increasing number of DAC designers and manufacturers. They are not doing it for nothing.
I am sure that you are quite able to simply test it yourself. But it would require hearing rather than measurement, but I don't consider listening to be nonsense, just perfect sense.
OTH, if not willing to try it, then I have at least given you some 'entertainment' and you are welcome.
That comes as a surprise, as I try hard to avoid posting "nonsense" even "confounding nonsense". (not claiming infallibility, though).
Could you please quote a few recent examples? I'm interested and if true, I'll be more than happy to correct and to avoid something similar for the future.
Just two or three examples will be sufficient. Thank you very much in advance.
You have lost me, where is speed impeded and what function would be hampered? And please define nonsense?
Also, tell it to an increasing number of DAC designers and manufacturers. They are not doing it for nothing.
I am sure that you are quite able to simply test it yourself. But it would require hearing rather than measurement, but I don't consider listening to be nonsense, just perfect sense.
OTH, if not willing to try it, then I have at least given you some 'entertainment' and you are welcome.
What are some of the commercially available dacs that this is implemented?
Under <$1k .....preferably <$500
I’m in the market for standalone dac to mess around with.
I don’t have the abilities to build one.....maybe one that’s easily modified should be in the criteria. Or even loaded boards (is that the correct term?) to be assembled (with decent instructions!)
- Status
- Not open for further replies.