Distinction-(TDA)1541 Build Thread

Status
Not open for further replies.
Hi Wlowes,

Is that one of Elibs S2 chips '97 that were available 10-12 years ago?..

Nice work BTW :)


Shane

Thanks guys.
I do not know where the S2 came from originally. I got mine, thanks to the generosity of a member in Alberta Canada. He had collected 2, and used one and sold me the other for what he had paid for it earlier. I consider it a treasure. I will debug my system with a lessor chip, and then plug in the S2. I have no doubt th2 S2 is authentic. Earlier I had purchased an S1 from an unknown source in europe. I may sell it off or use it in my Rotel CDP if its good.
 
Wlowes,

Reading you I see you prefer the non isolated output or remove the isolator from the WaveI/O, : better subjective sounding bypassing this isolator ?

Damien
True confessions. To date I have not tried WaveIO isolated. Mostly just lazy.
My current DAC is really a mess. I got WaveIO working with u-fl cables as a temporary measure and never went back to experiment. Cables were much longer than necessary, and just soldered to the pins of my tda1541a. The whole thing was way too crowded and poorly exectuted and once it worked, it made beautiful music and I did not want to mess with it.

I always expected to build this new music machine with a set of choices that the collective wisdom of this community indicates.

I believe most on this site have said WaveIO is at its best with the isolated output, so in this new configuration I am starting out this way. It is also a little bit me being lazy again because Ian supplies a cable with his I2S -> PCM board that I easily reconfigured to fit the WaveIO isolated output. This saves me buying the u-fl sockets and cables and then soldering those tiny things onto Ian's board. I know I am not being a good engineer and doing the thorough sequence of comparisons. This thing sounds so good, most of the time i just want to listen to it.

One surprising comparison did pop out however. I needed to test the new power supply for the WaveIO and BBB. The old one was a reasonable self built linear power supply with Toroid, silicon bridge, clcrc filtering. first 2 caps were very cheap electrolitics. Final is blackgate. The new linear supply is a repurpose of Peter Danial usb dac supply. The sound improvement was surprising. Looking at the 2 supplies, I think the material difference would be the boutique diodes first, and perhaps the hammond 229 style transformer vs toroid. I had never given much attention to diodes. Now I think I will be experimenting. My new music player used tube diodes for the tube output stage, and ultra soft recovery diodes in the digital stage. Since there are so many changes at once with the new player, I won't really know the impact of just the diodes. Given they realtively very inexpensive, I think I will experiment with them in my amps. I never cease to be surprised how importent the power supply is.
 
Another Distinction springs to life

Big milestone today. I fired up the new music server for the first time. All is working great.

BBB -> WaveIO (isolated I2S) -> Ian I2S 2 PCM (simultaneous mode) -> Distinction

It is in my basement listening room burning in. I will post my impressions of sound after a week of running in. Initial impression hints at greatness. With tons of new caps, (including Blackgates) it has that usual tightly wired sound with very little base or air. I know at 10 hrs that will all start to wash away and at 100 it will be time to comment. Clearly it has a lot more micro detail than the WaveIO to 1541A DAC it replaces.
Also very happy with the tube output stage. The old one had a lot of hum and noise. This one is dead quiet. My amps will now be able to show what they are really capable of with a lower sound floor.

Stay tuned for first impressions. A couple of pics with the finished version from the top side and the lower side. I kept the digital on the bottom and the power supplies are all isolated with their own copper or brass sections.
 

Attachments

  • reference music server analogue section.jpg
    reference music server analogue section.jpg
    237.4 KB · Views: 491
  • reference music server digital section.jpg
    reference music server digital section.jpg
    303.8 KB · Views: 478
  • reference music server top side.jpg
    reference music server top side.jpg
    291.9 KB · Views: 468
  • reference music server build bottom.jpg
    reference music server build bottom.jpg
    356.6 KB · Views: 470
  • Tube power supply.jpg
    Tube power supply.jpg
    239.1 KB · Views: 456
Last edited:
Thanks,

So the Ian's I2StoPCM is a huge improvement. Good !

Tell me more please about your BBB :

which distrib, whichstreamer soft ? Player or Server+streamer
why not using it as I2S output directly to the I2StoPCM ? because clock ?
what about an external clock to slave both the BBB and the distinction-1541 (external clock near the distinction board stacked !)

But sure because the NDK choice for XO and the good sounding reviews Wave/IO still a great choice and good Q/P SQ ratio.

When I look at your DAC and the one of Shane (Ceglar) also, I sleep like a baby... in the middle of the night I wake up & sudendly....I cry :bawling:
 
So the Ian's I2StoPCM is a huge improvement.
:
All I can say is that the combination of new items are trending to great. Positive changes likely from:
-First and foremost the Distinction board and its choices. It uses film decoupling caps vs my former polymer, free running DEM with 6k8 resistors vs Grundig, much better lay out, ground plane etc
-Ian's I2S with simultaneous mode
-The S2 double crown likely puts a little icing on the cake
-much better tube section with all film caps, dual mono, tube rectifiers etc
-separate power supplies with tons of filtering, salas shunts and lots of filtering (although the prototype was not a slouch in this area)
-I think the physical layout is important. I believe in both electrical and mechanical isolation, and vibration dampening. This is attached via AVC to 110W OTL monoblocks. On one channel with ear to tweeter, I cannot tell the system is powered up. On the other a small hiss and minute hum only audible at 2". I'll see if I can track that down over time.

Tell me more please about your BBB :
which distrib, whichstreamer soft ? Player or Server+streamer
why not using it as I2S output directly to the I2StoPCM ? because clock ?
what about an external clock to slave both the BBB and the distinction-1541 (external clock near the distinction board stacked !)
:
I am using current RuneAudio B2. (look forward to B4). It is MPD based. I expect at some point to go direct BBB to Ian once I have a reclocker/isolator. I have ordered Acko S03 board, and watching TP. Mean time I will use WaveIO happily as it is very good, especially when SPIDF is removed

When I look at your DAC and the one of Shane (Ceglar) also, I sleep like a baby... in the middle of the night I wake up & sudendly....I cry :bawling:

Thanks... I really kept this as a background hobby so it got a lot of time. My goal is to have a platform that can beat any commercial offering at any price so I figured I may as well build it strong as it will be with me for a while. Actually found myself surfing for atomic clocks ...
 
Last edited:
Hi Wlowes,

I have been beginning some reads about those nano-computers with the hope to replace a SB Duet but still with a good layout with the I2S because the number of pcbs interlinked involved.

You're right I believe : you can not swap the Wave I/O because the good role of the isolator chip !

Well, I scratch my head a little : BBB seems have some limitations as well and not so easy to re clock from a MCLK without loss due to the layout (my simple understanding.

I looked at Runaudio, great but don't understand if you can Stream into from an another player (in a pc) which manage a good upsampling (HQR player, AUdinirvana, sox extension in Foobar, Jriver, etc).

I have a particular attention to the new Raspi because the upper ram and the W10 capability (+ free ! = limited OS?) which maybe will permit to include those upsampling involved player !

It seems than an upsampling (176/192) played NOS on the TDA 1541 and/or an oversampling x4 with an upper upsampling at 352/384 give outstanding result.... We must re read Pedja, Thosten !

Yes I'm very Jealous about your tubes output, really think odds harmonics on some tones permitt (mixed with good others harmonics like a 300B shows) a pleasant listenings. It is my simple understanding acoustic instruments have those same mix of harmonics ! And some odds harmonics are involved in the identification of some notes. Maybe a too simple view from me. As I'm poorely skilled I will prefer to acheive that with discrete SE + good little output traffo (less expensive than the ones involved with tubes !)... but now would like as the life is too short and have a too huge Library going towards a DHT outputs (which be dangerous with my por skill again !) !

Well if I could avoid some expensive streamer like HQR, could be better, but some prioritary digital filter seems to be a good job like here ! Despite all the good job maid by some diyers for upgrade the server and the good x64 chips we have today, nothing seems to be better than a good isolator before the DAC and a FIFO... Ian's one for the TDA ... waiting the furure generation of Amanero, Wave I/O, JLSounds with embeded FIFO ! Well I give the idea :D... all those interlink connections are a nightmare and space consuming like EMC trap :headbash: !


Pffff must write to synology : the are always at 7.73 revision of Squeezebox server which is limited to 48 or 98 Khz ! It's 192 Khz with upper revision : 8.xxx. Well my Synology NAS and SB Duet have defintly some limitations. Luckily I believe I can Stream 384 K hz from the USB output of my NAS SB server) (with another streamer than the SB server package via a Wave I/O (not received this last yet : shipped last friday by Lucian, just went out of Romania yesterday :eek:... Romania Post :rolleyes:)

Well I 'm looking for a small cheap or nano computer to Stream via an usb output as isolators is a must go !

Have a poor understanding for the moment of the limitation of the outport drivers and the best to choose (WASAPI VS ASIO on Windows, the one used in the different linux distrib betweens all those nano computers !)

Julf, members, if you read me, any idea please ? Sure a good GUI remote solution could not be avoid.
 
Last edited:
Hi Eldam
I think we are straying a little off topic for the build thread, so it you want to PM me with questions about what player or PC to choose. Reading your post, I think we have all at one time struggled with the same issue... so many options and so many choices.

For me I got anchored at this wonderful 20 year old TDA1541a chip and that led me on a path to the Distinction board. Much thanks to Ryanj and the contributors for this.

Other forks in the road lead me to use minimalist nano PC as a source with music on NAS. But that is a story for a different thread.
 
:grouphug: Not sure it is totally off topic, I will ask to the man bellow this thread and which was also one of the initiator of the Distinction-1541 ;) !

But as you like ! Have to take care with the off topic !

Ceglar has been baned for life about an off topic spam and rough attitude?! ?! ?! which is a total uncomprehension for me as I know the man and find him a gentleman and very gentle nice & smart guy (as a serious one in vendor/consumer relations: look at him eBay profil !).

I believe sometimes second degree is not understood when it's writed and the man has a strong surealistic humour I like: jumping from serious to surealistic which has to be read between the lines : you can not be rude when you had just before a smart explanation and show after a freedom cartoons words description ! (here they kill for that not just baning ). I surmise a poor man with no brain to have delation to the mods because unable to defense oneself with words.... pffff !

Shane (ex Ceglar member) is really helpful for tda1541 tubes lovers and give a lot to the others ! Hope strongly the mods let him in again... holydays can be good for the two party as well :)

So Wlowes I will ask you if you want by PM but in my head all who go to the DIStinction-1541 and comes out of it has a place here ! It's my philosophy : feel free to write and to ask as myself not a Genius I feel we are more intteligent togethers :hug:. It was my motivation to be one of the main driver of the project which the distinction board was born : i GAVE TIME AND POSITIVE ENERGY AS i HAD NO KNOWLEDGE TO GIVE TO THE COMMUNAUTY :) and the main technichal contributors will answer you that without this attitude of me to drive against some real rude guys (the word real has a sense here !) like e.g. the OP of the "is there any good TDA 1541 kit" : you will never had the Distinction-1541 in the hands ! BTW I prefer from far member with free attitudes than hypocrite one ! IMHO.

It was a common effort ! And I paid my board !
 
Last edited:
Eldam
With respect, I thought the Distinction Build Thread would be about building the distinction board. As I said, I am happy to share my thoughts on total music server and the choices, but for those who simply want info on building the Distinction, they might not want to wade through the choices upstream or downstream of the Distinction. Happy to take this offline. I also have updated my blog to discuss my personal choices wrt the entire music server of which Distinction is the star player.

As before I remain grateful to you, RyanJ and the entire community for creating Distinction. I can assure you, it is making wonderful music in my room.
Walter
 
Updates to the Distinction 1.0 Build

I see its been 6 months since last post. But for the record, I wanted to make a couple of updates as this thing has just gone from great to outstanding.
First a couple of sort of off topic items. I think it is relevant as the DAC board is only as good as its transport. I had been running bbb-WaveIO-Ian I2S to PCM-Distinction to lampizator Tube output.
First removed the USB stage and replaced with Acko S03 isolator. Of course, linux distro is now by Miero. As good as WaveIO is, BBB direct plus the S03 as master clock and isolator is major step up. It too needed some tuning, and TPS4700 low noise regs and BlackGate NX HiQ caps on the S03 took system to a new level. Point is the transport was already getting pretty good.

So that sets the stage for the Distinction build topic. Having scored a new supply of Blackgate FX,N and Nx HiQ caps I was intrigued to see what the impact would be on Distinction. Eldam had intrigued me with posts suggesting choice of caps on TDA1541 board makes a significant difference. I replaced OSCON SEPF 160uF caps on each of the PS rails. The +5 and -5 got BG Nx HiQ 33uF 6v. The -15v got a BG N bipassed with a .1uF NX HiQ. At the edge of the PCB where power lines come in, the +-5v lines each got a BG FK 460uF and the -15V got 2x1000uF BG N in super E configuration.

I have been listening now for past 3 hours with jaw dropped. It is one of those got to relisten to everything changes. If you know BG you know that they will continue to bloom for another 100hrs so who knows where this is going. I almost didn't want to post this because I know these old BG caps are not easy to find and sometimes cost way too much. If you get lucky and see the chance to pick up 3 HiQ for the Distinction board, grab em. With that, I'm done. Thanks Ryanj.

All I can say is thanks Eldam for causing this itch.
 
Last edited:
hand holding please

I started to populate the boards and need some help.

1) Can someone post a picture of the completed board ?
2 ) Help me understand the parts to populate in the areas high lighted in red in the picture attached ?
3) Any simple op amp based IV for me to test the board after populating ?

thanks


kp93300
 

Attachments

  • tda 1541 board v1.jpg
    tda 1541 board v1.jpg
    664.4 KB · Views: 341
kp93300,

See pic of how I did mine. The jumpers in the area you highlight are basically there to allow you to pick in the future if you want the 1541a wired to run from an I2S signal, or from PCM. I use IAN's I2S to PCM card therefore if you look at the 1541a data sheet, (refer to Table 1) I need -5v to pin 27, so you tie pin 26 to 27 giving the -5v selection. When I ran direct from a WaveIO board, I would have wired it for +5v so tie pin 27 to pin 28.

In most cases 27 will need to be +5v. Most designers just short 27 and 28. the Distinction board is really elegant here. Ideally you should put a row of pins in your area highlighted in red so you can put a jumper making it flexible to start with a USB to I2S converter such as WaveIO and later add in Ian's I2S to PCM board and compare the effect for yourself.

I hope this helps.
 

Attachments

  • Distinction wiring.jpg
    Distinction wiring.jpg
    187.3 KB · Views: 328
  • tda 1541A.pdf
    186.6 KB · Views: 109
Wlowes. Very helpful picture. What are the values of the two resistors and the capacitor of the DEM circuit? Is the type and value of this capacitor critical?
resistors are optimal at 6.8k - choose reasonable quality low TCR.
SMT silver mica 430pF cap is considered by many to be tops. I used 100pF

If you go back through the thread, Ryanj and Eldam comment on cap values and make. I started with Oscon 160u, and migrated to Blackgate 33u. all good.
 
YOU NEED A GOOD POWER SUPPLY

@ KP93300 & Luke

Outputstage : http://www.diyaudio.com/forums/digital-source/227677-using-ad844-i-v-104.html#post4406885 :p

from J. Rasmussen (thank you to him) and as far I believe to know : from P. Rogic's original idea :) ! (notice P. Rogic prefers the opa861 (?) to the AD844 he used to populate at the beginning)


About the "3" PS caps for the -/+5V & -16V AND if you have three standalone shunt Power Supply only ! In your shoes I would only populate the smd decoupling traces : 0.047 to 0.100 uF .... and only on the - 16V trace (so Nothing for the 5V rails).

PPS cap is a premium choice (Ceramic NPO size is too big for this board). Avoid ceramic XR7.

Standalone 0.1 uF BG N or NX near the chip PS pins & ground is a premium choice as well.

KP93300, due to the ground layout bad design of the Distinction pcb : If the main caps holes of this board are "really" needed for you to PS the chip ( and with something different than a shunt PS design), you should let float the gnd lead of each PS decoupling cap to wire it with an independant wire to pin 5 star ground (and of course also from the same floating pin towards the regulator PS gnd before the board (just wrap the floating gnd lead of the cap on a non insulated section of the wire). So the only way is to populate the caps below the pcb (top of the cap towards the floor of the case to solder the current lead to each supply hole/via and allow the ground lead of the cap to float and wire it to the pin 5 below the pcb !).

Of course values of those caps have to be choosed in relation of the design choosed for the regulated PS boards ! So If not a shunt PS board or specific OEM regulator chip, you can experiment with several values from 100 uF to 470 uF. There is no universal value which feets for all the supply and value is different for each type of cap with a same PS design (so try & error method with different values and caps is the only good way : your ears will appreciate the needed type of the last decoupling cap with its specific capacitance value in relation to the PS you choosed before ).

But really, if you have a good shunt power supply for each voltage and fast enough, please just try my tip with the -16 V only (0.047 to 0.1 uF PPS smd cap) ;) :D

Hope this helps
 
Last edited:
Status
Not open for further replies.