|
|
Post by jh4db536 on Jul 20, 2021 20:43:20 GMT -5
Just wanted to share this project is one of many that i have created and it is still a work in process because it keeps yielding fruit. I'll try to keep it layman. It is my third DIY CDTransport (all analogue parts removed), and if you're wondering why i need three: each are of different philosophies ( Philips Arc single laser, Sony Heavy Platter, and the more modern light platter Sanyo. There are lots of variations of each of these over the decade, but that's the basic categories. Basically this stuff was invented jointly by Sony and Philips in the 80's ( SPDIF protocol as well) and basically went extinct in the late 90's or early 2000's. The single arc laser predates linear moving lasers which need 3 lasers to compensate for unnatural motion of straight line on a spinning disc. This is also light platter philosophy...Sony VRDS uses heavy platter with brushless motors to compensate for the torque needed to spin at different rpms required (not constant and weak like turntables). This is from an era where CD audio was not treated like data. Where mechanisms had to be built to actually read correctly with primative error correction algorithms. Where redbook audio and not random access time for CDROM was not a priority or even a thing yet. Only after modifications to remove "homogenizing and upsampling circuitry", there can be large performances variances between units. This is something you have to experience yourself to believe I found myself some treasure from 1994 at the junkyard and it was in sad shape.  The engine had good compression though, the servo board, and decoder board were still good. Everything else was smashed up pretty bad and i spent a month fixing it just to get it to turn on and read a disc and to know if this project even had a future. I took this to 5 repair shops to see if they could fix it for me, but i was declined by all of them because i never fixed damage this extensive before and i thought i was over my head    With overnight parts from Mouser and Russia, this will destroy everything on the streets. The CDM-9 single arc mechanism is still good (lots of other good vintage mechanisms in the family) but the CDM-9 is the final evolution before the whole thing was discontinued. The main reason this is worth the effort. Also wanted to mention, this player sucks before modifications. Sounds pretty much the same as a modern dvd player and there is no benefit to getting one of these unless you know what you're doing.  I restore and recap all the 25 year old parts im going to keep in the medium term. Eventually everything will be in the trash except the Mechanism and decoder board. The decoder board needs serious rework to undo the asynchronous reclocking scheme that was supposed to correct errors and also make the clocks more accurate. Worst, this player is not following the service manual schematics, so i had to do quite a bit of trial and error to make it work right. The two basic modifications once i get rid of the bad stuff is 1) Good Waveform clock directly to the decoder chip (SAA7310) and the SPDIF transmitter 2) and Good Digital Data (SPDIF) output directly from the transmitter chip (PCF2705) will lead to better things downstream when buffed by an EF11/12.   Now    This thing sounds great. |
|
|
|
Post by garbulky on Jul 20, 2021 21:30:56 GMT -5
Man, I bet it's a lot of fun when everything you use is made by you. Gives you a real pride of ownership in your work.
What's the digital connection you use? Tell us about your DAC and amp. Do you notice a difference in the sound of CD transports versus cheap DVD players? I notice no difference between PC BNC transport and my blu ray player's SPDIF.
What speakers do you like to use?
|
|
|
|
Post by jh4db536 on Jul 20, 2021 22:05:32 GMT -5
I didnt start with everything DIY, but that became the progression as i began to understand how to make things sound the way i imagine and also know and be confident about what type of sound i like. For that to occur, you need to be working with highly tunable or malleable designs (all classical topology, nothing new...no system on a chip stuff) and having lots of experience via meets, loaners, conventions helps.
Recently, i have started learning about antique radio methodologies and implementing those into my audio gear. This is another type of rabbit hole altogether, but it's all interrelated (guitar amps, antique radio, hifi audio stuff). Long time ago (haha im not that old), people did not have 8K UHD home theaters; people bought the latest radios and stereos and all the best technologies and efforts from Telefunken, RCA, Western electric, GE went into that because they also had military uses. Millions of tubes were made and only the best were selected to get into hifi audio use, in the Radar of a MIG, or the guidance of a ICBM. Today, everything is about silicon and software and audio is niche. You're plastic CD mechanism is out of tolerance? error correction will fix all imperfections which is why it all sounds the same because now it really is 1's and 0's philosophy down to the DVD mechanic. Rockna updating their DSP firmware every week (and radically changing the sound) by playing a special Wav into the USB like it was a windows10 update roflmao...jking i still think they they make a good product
I hit a wall with DIY chip sources as that reached beyond my paygrade (FPGA dacs) and i had LOTS of help with the DAC i own now which is a vintage Philips TDA1541. I had to take a break from sources for a while, and i familiarized myself with Analog Vinyl in the meantime (i dont own any). Specifically the idler tables ie. Garrards, Thorens, or Saskia. I want my chain to sound like that and it's very difficult to make digital sound like that. Philips dacs have a cult following and they have become extremely expensive. They usually sound bad in their default form, do not implement easily like a modern dac, and are extremely tunable. I am learning to play with this have a long way to go.
Use Coaxial connection myself, i notice a difference between my digital cd transports...i have 3 examples VRDS, Philips, and Sanyo and they are all very different. The more modern you go, the more similar they sound. I would probably say most of the modern ones ive heard sound approximately the same and i am pretty sure that's what most people have experienced. I could switch the connector on my DIY dacs and it likely wouldnt matter. The VRDS is the best out of the box by far. Arcam philips the worst out of the box. The Sanyo is the easiest and cheapest to DIY with satisfactory results.
Some DAC's prefer different outputs and inputs. Example, a Schiit Yggdrasil A1 or GS, it prefers it's XLR balanced outputs and AES digital input and if you owned one of these before you know exactly what voodoo im talking about. Yggy was my last commercially made dac that i owned and probably my favorite schiit product. My favorite DAC that doesnt cost more than a car (ie Rockna also of Soviet origins) is probably the Theta GenVA surprise was designed by Mike before he founded Schiit.
My speakers are DIY "Full range". My small room is limiting me, otherwise i would have a car size cabinet made for them. This is the best that can be done for my tiny closet. I mightve posted about building those before. I play in the high efficiency SET niche. I heard these ghetto rigged diy things at a at convention almost 7 years ago and fell in love with them on first sight. My mind still isnt changed today. They are not the best at everything (they do NOT throw the widest stage, crazy depth, "3d" hearing stuff behind u on 2ch, and they certainly dont rattle my windows outta the frames with thunderous 8hz bass), but they fulfill their niche.
There is a video i made with the Philips CDT before mods and a custom 300b amp. It's a cell phone S20 recording.
|
|
|
|
Post by audiobill on Jul 21, 2021 3:09:58 GMT -5
Neat!
I must say many of your preferences seem to align with AudioNote / AudioNote kits approaches.
Thanks for your writeup.
|
|
|
|
Post by Soup on Jul 21, 2021 5:26:39 GMT -5
All i can say is "wow"
Old school rules!
|
|
|
|
Post by jh4db536 on Jul 21, 2021 8:06:17 GMT -5
Neat! I must say many of your preferences seem to align with AudioNote / AudioNote kits approaches. Thanks for your writeup. interesting, there are elements of Kondo (Andy G founder of AN is disciple) in my chain. I do not own a single audio note product btw I might have a AN capacitor or resistor or two in my dac but that's about it. I do not sell anything, i am not a EE or have any formal education on this stuff. This is purely hobby and amatuer enthusiast purpose there is a commercially available modern version of my dac available (related through vintage Philips following) and I don't Avocate anyone buying it...it's full of audio note stuff. SW1x designs is very pricey equivalent to my dac from them costs as much as a car in £ (but you should check it out if you live in the UK i never heard one before). My DAC's PCB and design of the board were made in the Soviet bloc (Ukraine) and they follow Soviet audio philosophies...which are based upon antique German philosophies. The lore and legend behind this is wild and extensive, but it's pretty cool if you're interested in reading into it. There are actually books about it and i've read a few chapters before online to get the dL. my power Amp (originally a boutique American made Amp) is quite international: uses modern magnetics from Belgium, American modern output transformers, I prefer Chinese tubes on that one (normally you should avoid) but they successfully have reproduced vintage globe sound. I had to overcome my biases. Just to show you what my 300b power amp looks like (i was output transformer rolling). I ended up liking the Chinese (with Hitachi amorphous core) Double 8C design. In the back are also Chinese mixed metal core transformers. This is much cheaper than buying namebrand stuff because more $ isnt always better (sometimes)  been there and done that. This amp was originally designed to be an "objective" SET tube amp and i hated the sound. I have modified it more to my liking, but its hopeless. i guess i prefer lots of distortion and this doesnt have it. I keep this around to assess magnetics and to demo for people on what a 1% THD tube amp sounds like compared to something with 69SINAD and 2 digit noisefloor   My 2a3 amp. This is where all my effort goes.   I even make my own wires out of belden9497 and neutrik phono plugs. Attention to detail will catch that i mark the way i cut the wires. They are at least marked in the orientation that they were forged in...if you believe in that. It's not required, but you only get the opportunity when you actually cut it yourself.   |
|
|
|
Post by garbulky on Jul 21, 2021 17:13:04 GMT -5
Looks like you are living your best life!
Why did you pick the TDA chip over say a PCM 61 (or was it PCM 63)?
|
|
|
|
Post by audiobill on Jul 21, 2021 18:13:55 GMT -5
|
|
|
|
Post by garbulky on Jul 21, 2021 18:15:31 GMT -5
Recent ASR reports, for what they’re worth, found that certain popular transistor amps offer plenty of the wrong kind of distortion. Basically underdriven switches. “Accuracy” over great sonics is a huge mistake so many fall into. Experience generally corrects this, hopefully. It depends. People interested in neutrality or accuracy will not be satisfied with anything but that. However, some people, like me and you, I wager fall in to the camp that they want what sounds good to them...hence tube gear. Other than one spectacular tube amp, I haven't been in to tube gear but I can certainly see why people would be. |
|
|
|
Post by jh4db536 on Jul 21, 2021 21:49:36 GMT -5
Looks like you are living your best life! Why did you pick the TDA chip over say a PCM 61 (or was it PCM 63)? I signed up for and received a TDA1543 dac (i wasn't even told what it was) to evaluate and i liked the unique bold Philips sound although I did not like lower level technicalities of that specific dac. I decided to yolo a better TDA1541 because it showed me i have been going off tangent in my dac journey. I had strayed from good Timbre in pursuit of "special effects" : listening to how much tape hiss could be dug up, listening to where i perceived sounds were coming from and how it was presented instead of being engaged to listening to the music. I was warned at the time not to use it with PI2AES, but i thought it was the best thing since sliced bread since it worked so well in Yggdrasil. Anyway i gave this dac a terrible review, but it grew on me after a month and i realized that i been listening to music out of focus for a years...as in things were artificially widened the sound spread thin like a fish eye lens. It was this budget DAC that taught me what good timbre sounds like...that is frighteningly realistic sound. The TDA1541 ended up being a great buy as it had realistic sound and fixed all the technicality issues that the 11bit 1543 dac had. Another thing is, this DAC scales with the gear behind it in a way that i really like. It has been the most enjoyable dac that ive ever heard so far. When i went back to my other dacs, they just seemed to be missing some organic element to it. The burrbrown PCM58 interests me and maybe i'll get around to trying that chip next. |
|
KeithL
Administrator 
Posts: 10,273
|
Post by KeithL on Jul 22, 2021 10:46:31 GMT -5
I should point out that, sort of by definition, SET amplifiers are not EVER going to be objective. (It's sort of like trying to build "a really powerful, really smooth running, car - with a really good one-cylinder engine".) Push-pull amplifiers replaced single ended amplifiers for several reasons... - in a push-pull design a significant amount of the distortion is cancelled out by the balanced topology
(the distortion a SET design produces may be less unpleasant - but there is a LOT more of it) - in a push-pull design the magnetic flux in the output transformer produced by the DC plate current of the output tubes also cancels out (this prevents the DC plate current from the output tubes from making much contribution to saturating the core of the output transformer)
(in a single-ended design the DC plate current, with nothing to balance it out, drives the transformer into saturation much sooner)
(this is why you need a MUCH bigger transformer in a SET design to produce even halfway decent performance)
And triodes also replaced pentodes for several reasons [2022/05/13 - just noticed - that's reversed - pentodes replaced triodes...K]
- they have somewhat better distortion characteristics to begin with - one of the extra grids can be used for local feedback
The "Williamson ultralinear" design topopogy is considered to be one of the most significant advances in the performance of tube amps. An extra winding is added to the output transformer - which is then connected to one of those extra grids on each of the output tubes. This provides local feedback that serves to linearize the performance of both the output tubes and the output transformer. (And you can only do that with a pentode or a beam tetrode that has the extra grid.) (I'm sure you can find plenty of articles that explain how it works.)
If you REALLY wanted to try and optimize the design of a SET amplifier... You could try AC COUPLING the output tubes to the output transformer (using coupling capacitors)... Then provide the DC plate current for the output tubes via a separate high voltage current source...
Basically, the audio signal goes through the output transformers, while the (unbalanced) DC plate current bypasses them. (Thus eliminating ONE of the major problems with SET designs.)
This should theoretically give you much better performance with much smaller transformers. (With a normal SET design about 3/4 of the transformers "flux carrying ability" is used up by the flux created by the output tube's DC plate current.)
This would have been extremely complicated using only tubes... but isn't all that impractical with solid state power supply regulation...
I'm sure this has been done... and it probably worked... But it's still going to make much more distortion than a decent solid state design.  (It's like trying to design a really comfortable horse.)
The problem with a lot of recent Chinese SET designs seems to be pretty simple...
Rather than even pretend to attempt to deliver the best performance possible from a SET design...
They instead actually intentionally do their best to deliver "the goopy syrupy SET sound that many audiophiles seem to expect". I've heard one or two that were downright awful.
(They expend a lot of effort on pretty metalwork... usually heavy aluminum... but mix that with truly awful circuit design and parts choices.) It's also worth noting (for those not familiar with the subject) that SET amplifiers tend to have what is called a monotonic distortion curve... Most modern designs, and especially those with plenty of feedback, have very low distortion at low levels, which rises suddenly as their output power approaches clipping... In contrast, SET designs have an almost pure monotonic curve... This may literally mean a THD of 0.1% at 0.1 watts, 1% at 1 watt, and 10% at 10 watts... (So not only do they "clip more gracefully" - but they actually produce clearly audible distortion even at relatively low power levels.) One of the reason that people routinely report that such amplifiers "sound more powerful than their ratings" is because we humans tend to interpret distortion as loudness or power. (This is one reason why a table radio, with only a few watts of power, can seem to be able to play so loudly compared to a more powerful system with lower distortion.) Thus, because of the gradually increasing distortion at higher levels, the resulting audio sounds almost like it has passed through a dynamic range expander. (An increase in level sounds "bigger" than it actually is because it is accompanied by an increase in distortion... which makes the music superficially seem "more dynamic".)
Neat! I must say many of your preferences seem to align with AudioNote / AudioNote kits approaches. Thanks for your writeup. interesting, there are elements of Kondo (Andy G founder of AN is disciple) in my chain. I do not own a single audio note product btw I might have a AN capacitor or resistor or two in my dac but that's about it. I do not sell anything, i am not a EE or have any formal education on this stuff. This is purely hobby and amatuer enthusiast purpose there is a commercially available modern version of my dac available (related through vintage Philips following) and I don't Avocate anyone buying it...it's full of audio note stuff. SW1x designs is very pricey equivalent to my dac from them costs as much as a car in £ (but you should check it out if you live in the UK i never heard one before). My DAC's PCB and design of the board were made in the Soviet bloc (Ukraine) and they follow Soviet audio philosophies...which are based upon antique German philosophies. The lore and legend behind this is wild and extensive, but it's pretty cool if you're interested in reading into it. There are actually books about it and i've read a few chapters before online to get the dL. my power Amp (originally a boutique American made Amp) is quite international: uses modern magnetics from Belgium, American modern output transformers, I prefer Chinese tubes on that one (normally you should avoid) but they successfully have reproduced vintage globe sound. I had to overcome my biases. Just to show you what my 300b power amp looks like (i was output transformer rolling). I ended up liking the Chinese (with Hitachi amorphous core) Double 8C design. In the back are also Chinese mixed metal core transformers. This is much cheaper than buying namebrand stuff because more $ isnt always better (sometimes)  been there and done that. This amp was originally designed to be an "objective" SET tube amp and i hated the sound. I have modified it more to my liking, but its hopeless. i guess i prefer lots of distortion and this doesnt have it. I keep this around to assess magnetics and to demo for people on what a 1% THD tube amp sounds like compared to something with 69SINAD and 2 digit noisefloor  My 2a3 amp. This is where all my effort goes. I even make my own wires out of belden9497 and neutrik phono plugs. Attention to detail will catch that i mark the way i cut the wires. They are at least marked in the orientation that they were forged in...if you believe in that. It's not required, but you only get the opportunity when you actually cut it yourself. |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 22, 2021 11:00:55 GMT -5
BTW... since you used the word... a few times I feel obligated to ask you for a definition of it...
What is "timbre"?
A sound wave has an amplitude, and a frequency, and its phase can be described (although that's just a way of describing relative time)...
And, obviously, for example, frequency response can alter the way something sounds because it changes the balance between the principal frequency and its harmonics... And distortion can change the way something sounds because it alters that balance or downright adds new stuff (which could be either new harmonics or unrelated garbage)...
But what is "timbre"... and how do I measure it?
(Or is it really just a convenient term we use to explain some combination of other characteristics we can already measure?)
The old TDA154x DACs are older R2R DACs... And we know that their technical performance is relatively poor by modern standards...
(To be quite blunt the newer ones that replaced them measure better... and many people believe that they also sound better.)
So what difference would account for them "doing something better"... (Some folks suggest that it is their lack of oversampling filters...)
Yes I'm being both a Devil's advocate and an objectivist here... By my reckoning there is only one possible "technically exactly right"... (And it would be my goal to "reach absolute perfection and then stop".) (And, obviously, if two things were both perfect, then they would sound identical.)
Therefore if two things sound different... Then either one is right and one is wrong... Or both are wrong - but in different ways... But of course we are all free to NOT prefer the one that is technically the most correct.
And, accepting that two or more things are somewhat wrong, we are then free to decide which flaws we prefer... (or consider less objectionable).
Looks like you are living your best life! Why did you pick the TDA chip over say a PCM 61 (or was it PCM 63)? I signed up for and received a TDA1543 dac (i wasn't even told what it was) to evaluate and i liked the unique bold Philips sound although I did not like lower level technicalities of that specific dac. I decided to yolo a better TDA1541 because it showed me i have been going off tangent in my dac journey. I had strayed from good Timbre in pursuit of "special effects" : listening to how much tape hiss could be dug up, listening to where i perceived sounds were coming from and how it was presented instead of being engaged to listening to the music. I was warned at the time not to use it with PI2AES, but i thought it was the best thing since sliced bread since it worked so well in Yggdrasil. Anyway i gave this dac a terrible review, but it grew on me after a month and i realized that i been listening to music out of focus for a years...as in things were artificially widened the sound spread thin like a fish eye lens. It was this budget DAC that taught me what good timbre sounds like...that is frighteningly realistic sound. The TDA1541 ended up being a great buy as it had realistic sound and fixed all the technicality issues that the 11bit 1543 dac had. Another thing is, this DAC scales with the gear behind it in a way that i really like. It has been the most enjoyable dac that ive ever heard so far. When i went back to my other dacs, they just seemed to be missing some organic element to it. The burrbrown PCM58 interests me and maybe i'll get around to trying that chip next. |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 22, 2021 11:03:35 GMT -5
I would be remiss if I failed to compliment you on the layout and lead dress on that SET amp. It manages to be both attractive and technically superb... which is a rare combination these days.
Neat! I must say many of your preferences seem to align with AudioNote / AudioNote kits approaches. Thanks for your writeup. interesting, there are elements of Kondo (Andy G founder of AN is disciple) in my chain. I do not own a single audio note product btw I might have a AN capacitor or resistor or two in my dac but that's about it. I do not sell anything, i am not a EE or have any formal education on this stuff. This is purely hobby and amatuer enthusiast purpose there is a commercially available modern version of my dac available (related through vintage Philips following) and I don't Avocate anyone buying it...it's full of audio note stuff. SW1x designs is very pricey equivalent to my dac from them costs as much as a car in £ (but you should check it out if you live in the UK i never heard one before). My DAC's PCB and design of the board were made in the Soviet bloc (Ukraine) and they follow Soviet audio philosophies...which are based upon antique German philosophies. The lore and legend behind this is wild and extensive, but it's pretty cool if you're interested in reading into it. There are actually books about it and i've read a few chapters before online to get the dL. my power Amp (originally a boutique American made Amp) is quite international: uses modern magnetics from Belgium, American modern output transformers, I prefer Chinese tubes on that one (normally you should avoid) but they successfully have reproduced vintage globe sound. I had to overcome my biases. Just to show you what my 300b power amp looks like (i was output transformer rolling). I ended up liking the Chinese (with Hitachi amorphous core) Double 8C design. In the back are also Chinese mixed metal core transformers. This is much cheaper than buying namebrand stuff because more $ isnt always better (sometimes) been there and done that. This amp was originally designed to be an "objective" SET tube amp and i hated the sound. I have modified it more to my liking, but its hopeless. i guess i prefer lots of distortion and this doesnt have it. I keep this around to assess magnetics and to demo for people on what a 1% THD tube amp sounds like compared to something with 69SINAD and 2 digit noisefloor My 2a3 amp. This is where all my effort goes. I even make my own wires out of belden9497 and neutrik phono plugs. Attention to detail will catch that i mark the way i cut the wires. They are at least marked in the orientation that they were forged in...if you believe in that. It's not required, but you only get the opportunity when you actually cut it yourself. |
|
|
|
Post by jh4db536 on Jul 22, 2021 13:22:24 GMT -5
Thanks. Not trying to turn this into a debate. I've heard people describe "timbre" as a coagulation or summation of objective and subjective measures. I think of it as a Primarily Subjective measure of realism without 'redflag' objective issues. I guess it would be like trying to measure "clarity" of an amp objectively, which i bid is way different than lining up a output sine wave on a scope compared to the generator wave versus just hearing clarity. Measurements on the specsheet on singular piece of equipment would not really sway my decision to buy gear. Loaners and auditions contribute to lightening my wallet most of the time. My end goal though is mainly to know how to create and customize my own sound rather than order off the menu. That would be much easier. I could confidently say that i prefer certain types of distortion and i don't want it canceled out. The 300b amp pictured above (ironically is designed actually by one the top objectivists) supposedly measures 1% THD. I have not kept the original design because i do not like that type of sound, but i keep it around if anyone who is over wants to hear the difference. Funny how difficult it is to attempt to Add good distortion to that amp, it's easier to redesign the whole thing from scratch to have good distortion in the first place. The tubes amps that i keep around here do not sound syrupy and slow. They are tubey but highly energetic, clear and hyper realistic and that is likely not going to measure objectively. I have a $90 schiit magni heresy in my drawer for times when I'm feeling objective On TDA154x, they are from 1980's and an honest 16bits maybe the good binned ones. There are LOTS of chips that have way better specs. TDA1541 usually do not come in NOS implementation, but it is a popular modification to remove the OS chip. They are removed (or not included) on my DAC, so it is in fact operating in NOS mode. I like a lot of NOS and OS dacs of different technologies ie. Rockna Wave dacs, Yggy A1/GS, Theta, TDA1541. I have a high appreciation of Vinyl although i will never own one. Tying this back to the original topic, there is a LOT of information within 16 bits of redbook audio technology originally created in the 80's and still the same stuff. It is amazing how great a 30 year cd transport is Capable of sounding. You could even feed Modern XRCD and SHM-CD (backwards compatible unlike SACD) into a vintage player no problem. |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 22, 2021 15:29:41 GMT -5
That was an excellent reply - and an excellent definition. You obviously have an solid handle on what you're looking for and what you actually want. (Many "audiophiles" seem to lack this... and are merely reaching for something they imagine based on what they read somewhere in a review.) It should be pretty obvious that I personally am a fan of "maximum accuracy and concise detail with the least possible coloration" and so I don't listen to a lot of SET amplifiers . However, several years ago, we had a sample here from one of the Chinese off-brands that was popular on eBay at the time. The construction included nice meters, and lots of shiny aluminum, including heavy side plates, and a safety cage comprised of a row of round aluminum bars. (It looked sort of like the killer satellite in an old James Bond movie...)
As I recall it delivered about 30 watts/channel from three chassis - a preamp and two monoblocks.
But it also sounded remarkably syrupy and goopy... It was obvious that the manufacturer had gone far beyond the limitations of the technology to deliberately make that amp sound different. (I would have to say that it managed to sound impressively bad.)
I'm not a fan of NOS DACs... Without oversampling there is no way practical to avoid a significant roll-off with 44k content - and, to me, that's one of those deal breakers. I should also point out that, while Yggdrasil is an R2R DAC, it does use oversampling - using Schiit's proprietary oversampling filter.
I'm inclined to believe that the major problem with how CDs sound is that many modern CDs simply aren't mastered very well.
I've heard enough CDs that sound spectacularly good to have no doubt about how good the medium can sound if done correctly.
Thanks. Not trying to turn this into a debate. I've heard people describe "timbre" as a coagulation or summation of objective and subjective measures. I think of it as a Primarily Subjective measure of realism without 'redflag' objective issues. I guess it would be like trying to measure "clarity" of an amp objectively, which i bid is way different than lining up a output sine wave on a scope compared to the generator wave versus just hearing clarity. Measurements on the specsheet on singular piece of equipment would not really sway my decision to buy gear. Loaners and auditions contribute to lightening my wallet most of the time. My end goal though is mainly to know how to create and customize my own sound rather than order off the menu. That would be much easier. I could confidently say that i prefer certain types of distortion and i don't want it canceled out. The 300b amp pictured above (ironically is designed actually by one the top objectivists) supposedly measures 1% THD. I have not kept the original design because i do not like that type of sound, but i keep it around if anyone who is over wants to hear the difference. Funny how difficult it is to attempt to Add good distortion to that amp, it's easier to redesign the whole thing from scratch to have good distortion in the first place. The tubes amps that i keep around here do not sound syrupy and slow. They are tubey but highly energetic, clear and hyper realistic and that is likely not going to measure objectively. I have a $90 schiit magni heresy in my drawer for times when I'm feeling objective On TDA154x, they are from 1980's and an honest 16bits maybe the good binned ones. There are LOTS of chips that have way better specs. TDA1541 usually do not come in NOS implementation, but it is a popular modification to remove the OS chip. They are removed (or not included) on my DAC, so it is in fact operating in NOS mode. I like a lot of NOS and OS dacs of different technologies ie. Rockna Wave dacs, Yggy A1/GS, Theta, TDA1541. I have a high appreciation of Vinyl although i will never own one. Tying this back to the original topic, there is a LOT of information within 16 bits of redbook audio technology originally created in the 80's and still the same stuff. It is amazing how great a 30 year cd transport is Capable of sounding. You could even feed Modern XRCD and SHM-CD (backwards compatible unlike SACD) into a vintage player no problem. |
|
|
|
Post by jh4db536 on Jul 22, 2021 16:44:31 GMT -5
Other people have told me that my amp (dual purpose headphone and speakers...single secondary 8ohm output straight to the taps) does not sound like a Single ended triode; my amp can outslam/macro a comparable pentode amp easily while maintaining all the nuance of a triode. i know what they mean, but i dont know how to describe it either because all other amps that sound like this have been DIY so far. It's kind of one of those things that you have to hear with your own ears.
I thought i did not like NOS either, but this TDA1541 DAC changed my mind. i have heard other NOS TDA1541 like MDHT orchid and i didnt prefer it. Implementation really matters.
I am certain that i hear no roll off coming out of my headphones or speakers with this DAC (it might measure differently). In fact, this DAC made my yggdrasil sound like it was all mids it had so much extension in the lows and highs all else equal (i was shocked on how that was possible).
My initial hypothesis for why that is would normally point to my use of folded horn speakers. When you feed objectively perfect phase linear information into this type of speaker, it might not spit out full FR extension because it uses a lot of phase tricks to accomplish full range. Sometimes...feeding imperfect phase information into them might spit out more frequency ranges.
However, for that to be true, i should be hearing less extension on headphones and i dont. I am at a loss on how to put it in rational terms and i have accepted that it is what it is. It is tuned Specifically for HD580 or HD800 depending on how i set it up, but always works with my speakers. Not being a generalist use case though is the strength of these types of amps to be exploited
There really is a difference between CD digital transports and some of it can be objectively measured, but since it's digital data that has no correlation to how much difference that makes on the whole the chain and it varies alot depending on what is downstream. A lot of what i have done is documented by an infamous polish designer. I might agree with their datapoints observed, but maybe not their products or exact methodologies. I think these differences are as large or larger than different CD versions, which is caused by information losses as it flows down stream. Sins of omission are really that difficult to know unless you heard something that doesn't omit.
Some CDS sound better than others depending on the mastering just like a good or bad Vinyl pressing. A good amount of my CD collection is actually burned compilations that i bought from an estate sale 15 years ago. These somehow sound better than some of my real albums and they might be vinyl rips or just really well sorted first generation source material. It doesnt even have to be original always.
|
|
|
|
Post by groverh on May 10, 2022 14:14:45 GMT -5
Hi I'd like to modify my Arcam like you, could you give me a more detailed info as
to what you did? It's very difficult to make sense of the photo of the resistor to the decoder chip. Could you give me more information. I'm excited to hear what you accomplished.
|
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on May 13, 2022 15:13:01 GMT -5
Just by LOOKING at the photo it looks like both ends of the resistor are soldered to the same pad... which would do nothing.
I would assume that the end of the resistor that is soldered to the wire is actually above the pad and not touching it or soldered to it. (So, the end of the resistor towards the right goes to the pad, and the end to the left goes to the wire, and NOT the pad.) Hi I'd like to modify my Arcam like you, could you give me a more detailed info as to what you did? It's very difficult to make sense of the photo of the resistor to the decoder chip. Could you give me more information. I'm excited to hear what you accomplished. |
|
|
|
Post by jh4db536 on May 13, 2022 19:34:21 GMT -5
 there's a pad nearby and close. its basically a resistor across the clock (like stock circuit) since I removed the entire original smd section. this section is not described in the repair schematics. it's like redundant through hole section of the board that I had to figure out with dmm. as for the schematics, the one for Delta 250/270 are publicly available. cd880 cd650 are other popular units modded. as far as mods go, I would break it down into 3 sections: there is no step by step really this is really diy and subjective based. 1) 5v digital psu. on its own power transformer, no filtration other than 120hz (60hz), no ceramics, rectifier & caps chosen by ear. **if u don't believe digital section mods do anything, the entire concept of this cdt moot** my proof is my unit itself and especially how it interacts with non asrc/ uncorrected digital inputs of my downstream gear. modern gear that "fixes" signal input into it will diminish homogenize performance. i am not the only one who owns one of these, i heard someone else's first and i was sold...just copied it in diy 2) removal of all quartz, asrc, clock divider (logic gates), silicon buffer schemes from the QIP. replacement with external tube rectified, tube oscillator circuit, and pentode clock buffer. this was purchased as a mod kit, but its no longer for sale due to the war. ime this is where most of the performance gains come from. 3) digital spdif output is cut pure from QIP pin per the scheme. Replacement with tube rectified, and pentode spdif buffer. this was purchased as a mod kit, but its no longer for sale due to the war. all this effort leads to a cdt that sounds quite "analog and real" with none of the downsides. I've been into cdt sources long before discovering this stuff and I honestly never worried about usb nervousa, firmware, fpga or the latest upsampler 1.5314mhz and #taps. I cannot really explain why this works, but I've been able to reproduce it in 2 other units with the same mods. a few others have been successful as well. |
|
been there and done that. This amp was originally designed to be an "objective" SET tube amp and i hated the sound. I have modified it more to my liking, but its hopeless. i guess i prefer lots of distortion and this doesnt have it. I keep this around to assess magnetics and to demo for people on what a 1% THD tube amp sounds like compared to something with 69SINAD and 2 digit noisefloor 
been there and done that. This amp was originally designed to be an "objective" SET tube amp and i hated the sound. I have modified it more to my liking, but its hopeless. i guess i prefer lots of distortion and this doesnt have it. I keep this around to assess magnetics and to demo for people on what a 1% THD tube amp sounds like compared to something with 69SINAD and 2 digit noisefloor 
