PC USB better than CD?

[rickcr42]

not only a "surprise" but has for the first time in a while got me actually "THINKING" and later Building.

the thinking part is the power supplies and what/how to get them on the board so I'm pouring through the LDO data sheets to find "The One".

Plans being what they are (easily FUBARed) I'm going towards 3.3VDC LDO at the PCM2707,KWAK Clock 7 for re-clocking,CCS shunt regulated TDA

not yet sure if I want to go with passive SINC correction,transformer step up and tube buffer or the full blown TL SINC Filter/tube stage but I figure even with tubes the footprint will be tiny as long as I keep the power supply in another "box".

the passive option would allow for a more fair head to head with the USB-SPDIF-DITB so I can actually hear the chips/USB topology and not the output stage once gain matched.Maybe even use the exact same output stage for the test

DAC 1-USB to SPDIF to 8X OS DITB
DAC 2-USB to I2S to 1X OS TDA1543 with SINC Filter

Totally different topologies,both having strengths/weaknesses so a head to head a must though chances are will sound more alike than different one the interfaces (spdif one,I2S the other) and filter methods (8X digital,single pole passive analog provided by the Sowter/zero digital,SINC Correction for 1X treble rolloff,single pole passive provided by Sowter) are made optimal......or at least as close to optimal as my lame ass can get them

rickmonster

[rickcr42]

OK so I was weeding through my "stash" of Digital Audio chips and what did i come across ?

A good selection of LDO chips at the bottom that I had forgotten about along with a little "baggy" of parts for a pair of Jung "Super Regs".

The opamps in the baggy are AD797s though I may toss them in favor of some newer OA devices I have though will take a bit of pouring through the data to find which.

anyway a mini rant :

Older digital chips had actual PINS and did not seem so damn fumbly/girlie to work with as do the new parts plus the voltage standards have sgifted away from 5V (at worst +/- 5VDC) to 3.3 VDC and all because the main market is battery operated portable devices.Many of the older parts not only preclude a teeny tiny high desnity pcb footprint but would snack on batteries as if no more than finger food to pass time while waiting for a serious meal (real POWER).

cool.I know the "why" of it and ralise where consumer electronics are headed but stilll makes things no easy task when you try to meld the two into one and a high SQ the goal.

firstly mixed back DIP/SOIC devices makes for a very schizoid layout,one loaded to heavy/bulky on the DIP end,tiny and "busy" on the other.

Second.the standards change from +5 VDC to +3.3 VDC means you MUST use more than one regulator : one for the 5 Volt section,one for the 3.3 volt section and they both must fire up at exactly the same time or be sequenced properly for proper circuit operation (why there are "control" pins on the newer V-Regs ) rather than supplying the entire circuit with a single clean power source that is bypassed an decoupled right at each devices Vdd and Vcc pins.Yes the analog section usualy needs an even higher voltage for good performance and high output levels but to be honest if the initial DC buss is really clean and the Op Amp properly bypassed and again decoupled right at the power pins regulation is not essential so can be fed with the "raw" DC source before going to any regulators so in essence will get the highest current feed being first in line and will be isolated as well as can be all things considered from the digital power buss through the combination of local decoupling and the digital section v-reg providing a "dam" to digital power buss.At least it sounds good in my head and until someone shoots holes in the theory i'm sticking with it

Those who are queesy about unregulated analog stages need to think about how many unregulated battery powered audio devices there are so the only difference would be clean battery DC vs. possibly dirty AC and if the preceeding power buss is a proper Pi-filter and the AC feeding it clean,if the rectifiers have hash suppression i see no reason why the PSRR of modern op amps is not at LEAST the equal of your average discrete stage WITH a regulated power supply !

Anyhoo..........

Have no choice but to bit the bullet and go with two V-Regs so it looks like right now it is TI REG101 Fixed 3.3 VDC @ 100mA for the USB section and the Jung Reg at +/- 5 VDC for the DAC chip digital and analog sections all fed from my external "Power Station" (stole that one from Audio alchemy but it fits dammit ) that pushes out +/- 16 VDC, and +12 VDC,all way too high without more CRCRC filter stages to hit the regulators with the differential between the "Raw" DC and the "Regulated" DC being so wide the teeny tiny chips may eat themselves while trying to cast off the exess voltage which would be be converted to heat yet with no heat sucking area large enough to wick it off (i.e. a heat sink big enough translated from rickanese to english )

Hate to do it,means larger final caps than I would like to keep the raw DC feed at a lowish impedance but I don't make the rules (rarely follow them either ) so now it is a matter of placing the order for the MIA parts then putting theory to rest so I can nail down the basic architecture and begin the long tweaking process.Means "plain brown wrapper" passive resistor IV to start (will have an output in the millivolts range so will need me to use one of my gain stages to up it),then the "all passive" SINC correction,then various active IV stages then.....


Once I get the sound "ballpark" satisfactory I will fk around with trying various V-Reg methods on the TDA side and reclocking to see if it is just over complicating the already complicated,changes just to do them (FOTM thingy) or just change,or if the additions truly are an upgrade in actual SQ.

One of my problems with something like this is i never seem to stop "fiddling" with it instead of just closing up the GD lid and bolting it down,accepting the "thing" for what it is and enjoying it on that level.

Anyone who followed my years long Szekeres Amp optimizing path,a path that began at Headwize then on to Head-
fi when it came into being five or so years ago knows i went full circle from fast and light simple to comlicated enough to need a road map to back on a diet with only modest changes in parts selection and power supply upgrading (SE Class-A has shitty PSRR so if it is on the DC feed it is in the output of the amp )


more to come as my mind drifts into new horizons of audio whackery rick style

[PinkFloyd]

Haven't a clue what this thread is all about as I haven't read it yet but Rick seems very happy so it must be worth reading

I glanced at 5V power supply which is interesting as I was just looking at these little jobbies last night www.rapidonline.com part number: 85-2251 don't know if they'd be any use but, sizewise, they're pretty hard to beat at only 55 x 42.8 x 20.3mm and 2.3A too, smaller than a box of matches. Not sure why I was looking at them but I was and seeing as 5V has been mentioned here I thought I'd share.......... any use?

[jelosno]

In the true manner of this forum this thread started of discussing not what we've ended up with ;D

don't know if they'd be any use

Well me being the most unqualified to answer will do first then. To mee the PS looks pretty fine. From all the schematics and boards I've seen lately without understanding them in full - the usual suspect parts are there. The quality and cleanness of the supplied power though???

Not sure why I was looking at them but I was

That seems to fit into that thread. Rick had parts in his drawers he did not remember and all of a sudden they fit into a project.

Maye its because we are nearing Xmas. Comets fly around, Kings roam through deserts. A special time with lotsa history. With Xmans being the replacment event for one of the most important Roman festivities: The three days celebration following the winter solstice it is carrying a very old tradition to honour a special spot in the calenar. It was a special time way before they introduced that arbitrarily implemented AD year counting. If it has survived that long it can't be bad

As for the USB DAC kit: I have bought it because IT IS a kit and I can not design something like it. Soldering the parts onto it seems ok to me. Now with Rick this 'kit' will be a small part of a complete new development. Wish I could do that

Stefan

[xerxes]

Looks like a switching PSU to me, the sort that are sometimes used with class D amps. Apparently they chuck out loads of RFI, so much that they can interfere with FM radio and TV reception.

Take a look at some of the pictures: www.nuforce.com/products-monoamps.htm

[jelosno]

being the most unqualified

Just as I said...

[rickcr42]

Haven't a clue what this thread is all about

join the club dude.I wrote a good portion of it an I am also a bit baffled

Rick seems very happy so it must be worth reading

I could just as happily write about the color of the dirt in connecticut then go on to explain why the soil is a different color in various regions and i highly doubt anyone but an amatuer mineralogist would give a rats ass but I'm happy that your happy that I'm happy

I glanced at 5V power supply which is interesting as I was just looking at these little jobbies last night

Pictures are for the "looks good must sound good too" crew Mikester.I need to see a SCHEMATIC before I can get a woody

Now with Rick this 'kit' will be a small part of a complete new development. Wish I could do that

Baby steps Stefan.First you gotta crawl,then learn to stand up and take those first wobbly steps then when you learn to sterady up walk around comfortable and only THEN can you learn to run !

You have hit the crawls real good/can stand up for a bit before falling stage with your EAR+ build with the rest just taking time if you have the will to stick with it and keep an open mind.

Me ? Still at the "I can walk really really fast" stage with jog hopefully in my near future and run only a dream I doubt I ever reach but will have fun aiming for

Looks like a switching PSU to me, the sort that are sometimes used with class D amps. Apparently they chuck out loads of RFI, so much that they can interfere with FM radio and TV reception.

coud be Owen.That coil seems to say it is along with the multiple devices with the SOT23 part being the switcher and the power transistors the pass transistors to up the output current

Owen just may be at the "I can walk a bit" stage guys ;D

cheers.

rick out

[PinkFloyd]

xerxes said:

Looks like a switching PSU to me, the sort that are sometimes used with class D amps. Apparently they chuck out loads of RFI, so much that they can interfere with FM radio and TV reception.

Take a look at some of the pictures: www.nuforce.com/products-monoamps.htm

Yeh it is a switcher but at £5 a relatively cheap one and the size of a matchbox too.

[rickcr42]

Yeh it is a switcher but at £5 a relatively cheap one

Yeah,we wouldn't want to pay mucho deniro for a sonic butcher,far better to pay just a wee bit

BTW-switchers seem to be OK with tube gear,can be 50/50 with analog audio depending on application yet are a death march with digital audio gear with the reasons being all about frequency of operation and what the actual audio circuit will/will not respond to.

the very RF garbage we try to stamp out using all manner of EMI/RFI filtration devices is actually injected into the circuitry by the nature of the switching supply operating frequency and while steps can be taken to clean up these power line crap riders by the time you do you could just as easily ( and cheaper in the long run) have done a proper linear power supply.

So why does tube gear work with switchers ?

Self filtration guys.Normally the best you can expect at the top end is 50khz so all those artifacts have no where to go when they are looking to be amplified along with the audio signal and if a zero feedback design all you have to worry about on the RFI front is the input

[jelosno]

Baby steps Stefan

JAJAJA!

I would be happy to be able to just follow/understand all of your mods. I don't even think of the design side of things. Not sure wheter that is what I want anyway. Just to follow and understand would do I guess. Enough to do the one or other little mod here and there.... ;D

[rickcr42]

Mod #1 : Power Supply-get rid of the DC-DC converter,use an LDO for the 3.3 volt USB then a +/- 5V "Jung Super Regulator" to supply the TDA1543 and the poitive voltage to the 3.3 VDC LDO (pretty much a super reg itself)

MOD #2 : lose the entire analog section and replace it with a passive IV/first stage LP filter/SINC-compensation which broken down means :

use a simple passive resistor to convert the current output of the DAC to a voltage (hence "I-V" stage) with the passive low pass filter to roll off the spurious out of band signals AND to compensate for the rolloff in the treble region that takes place in a Non-OS type DAC.The only indecison so far is how much gain do I want at the resistor and how much made up by the next stage and if I can get suitable gain at the resistor without comprimising performance if maybe the output stage will end up being a simple buffer,maybe my fav buffer the LT1010.

The above mods are no more than tossing out what I do not need-the headphone amp-while at the same time losing the power supply that while maybe fine for a "travel" amp is not fine for home use IMO (and for the road I use the simple PCM2704 with AD825 output stage anyway me not being anal about my travel system and especially so considering the source equipment involved )

That is the big shit.The easy shit.The the part where I get out the framing hammer and bang those sucker into submission making them conform to the new digs (pc board )


the rest ? Choosing the right part for the right spot,designong a proper enclosure on the tecnical audio merits first then on style points and finally attacking all sources of vibration be they internal or external which usually has some form or another of wood and cork involved ;D



,