Update.
Just a quick update of where i'm at.
After finally receiving the remaining parts on backorder I fired up the regs and tested the Analogue side with dummy loads first. All three voltages came up on the money, with temperatures as calculated. A quick look on the scope revealed no obvious issues.
On the digital side; voltage and temps were also stable as calculated. Scope shot also appeared ok.
Input voltages from the SLA batteries were 25.5V and 6.5V respectively.
Sorry for the slow rate of my testing process... few delays but all on track.
Ryan.
Just a quick update of where i'm at.
After finally receiving the remaining parts on backorder I fired up the regs and tested the Analogue side with dummy loads first. All three voltages came up on the money, with temperatures as calculated. A quick look on the scope revealed no obvious issues.
On the digital side; voltage and temps were also stable as calculated. Scope shot also appeared ok.
Input voltages from the SLA batteries were 25.5V and 6.5V respectively.
Sorry for the slow rate of my testing process... few delays but all on track.
Ryan.
Interesting project!!
Haven't been thru this thread so I may be posing redundant queries, but, IAC ...
Will 1541-v2 incorporate DEM ckt, notably the one tubee designed? (I posted on my success and respect of this mod here).
What's that SMD part you're using to re-clock?
A full schematic with parts list would be helpful.
Continued motivation for the project ...!
Haven't been thru this thread so I may be posing redundant queries, but, IAC ...
Will 1541-v2 incorporate DEM ckt, notably the one tubee designed? (I posted on my success and respect of this mod here).
What's that SMD part you're using to re-clock?
A full schematic with parts list would be helpful.
Continued motivation for the project ...!
Hi hollowman,
Thanks, i'm trying to keep things interesting
So far the design i'm working on uses a simple negative bias to stabilize DEM oscillations - 6k8 to both pins from -15V, and a cap for the frequency.
Check out this comparison, take note of the frequency counter:
With 6k8 resistors
Without resistors (just timing cap)
The flip flop i'm trying is the Fairchild 74VHC175 in a TSSOP16.
Schematics - yes definitely post them but have to finalize the design first, still in testing.
Thanks for the link.
Ryan
Thanks, i'm trying to keep things interesting
So far the design i'm working on uses a simple negative bias to stabilize DEM oscillations - 6k8 to both pins from -15V, and a cap for the frequency.
Check out this comparison, take note of the frequency counter:
With 6k8 resistors
Without resistors (just timing cap)
The flip flop i'm trying is the Fairchild 74VHC175 in a TSSOP16.
Schematics - yes definitely post them but have to finalize the design first, still in testing.
Thanks for the link.
Ryan
DEM and re-clock
Unfortunately, tubee's schema JPGs are no longer active. If anyone reading this has those schemas, please re-post (I it/think they were simply hand-drawn on paper, and then tubee whipped out his dig. camera and snapped a shot! Hey, that works!)
EDIT:
Found this comment from ecdesigns:
"Never reclock all I2S signals with one chip, you will get massive crosstalk and highly increased jitter on both BCK and DEM signals. This is also easy to measure using an oscilloscope. The DATA signal causes most problems with crosstalk.
..."
In the DEM thread, with contributions by HtP, I tried a variety of DEM schemes and settled on tubee's version (which I found to a real improvement, even over Henk's original).
Unfortunately, tubee's schema JPGs are no longer active. If anyone reading this has those schemas, please re-post (I it/think they were simply hand-drawn on paper, and then tubee whipped out his dig. camera and snapped a shot! Hey, that works!)
For re-clocking the I2S lines, I think using one logic chip per line is ideal. E.g., WS and Data will have their own dedicated 74HC chip.
EDIT:
Found this comment from ecdesigns:
"Never reclock all I2S signals with one chip, you will get massive crosstalk and highly increased jitter on both BCK and DEM signals. This is also easy to measure using an oscilloscope. The DATA signal causes most problems with crosstalk.
..."
Last edited:
Hi hollowman,
Yes your right, crosstalk is an issue with a single chip re-clocking solution. Of Course the compromise going for multiple flip flops is clock loading, pick your poison. On Iancanadas PCM pcb you will notice a single chip flip flop - to my ears it sounds excellent. But i'll look into it, thanks for your input.
At this stage i'm quite happy with my current DEM solution, but i'd be interested in seeing the schematics from tubee.
Update
Hi Guys,
Testing is going well with the new pcb - i've been listening for the past week and i'm happy with the results so far. The filtering in the pre regulated supply has a big impact on SQ. Attention to detail here is important, but with only modest filtering from the SLA batteries good results were realised.
For the first few days I was listening in time multiplex mode (I2S), reclocked with the on board flip flops. Although for the whole time I felt that something was lacking. When I tried re-clocking PCM in simultaneous mode I couldn't get a lock on the I2S to PCM pcb (i'm thinking a latency issue from my clock output?). So I bypassed the re-clocker straight to the on board attenuators only to finally uncover some decent natural sounding music. I'm not sure I could ever go back to listening in I2S mode anymore.
So now it's apparent that a bit more thought needs to go into the digital side. Suggestions welcome, more research required. Perhaps re-clocking/ buffering on board is not totally necessary? Maybe just some jumpers to bypass the re-clocker if people don't want or need it?
At this stage I think it would be good to get some input from how people intend to use this pcb. I assume most will be using Ians gear?
Ryan
Hi Guys,
Testing is going well with the new pcb - i've been listening for the past week and i'm happy with the results so far. The filtering in the pre regulated supply has a big impact on SQ. Attention to detail here is important, but with only modest filtering from the SLA batteries good results were realised.
For the first few days I was listening in time multiplex mode (I2S), reclocked with the on board flip flops. Although for the whole time I felt that something was lacking. When I tried re-clocking PCM in simultaneous mode I couldn't get a lock on the I2S to PCM pcb (i'm thinking a latency issue from my clock output?). So I bypassed the re-clocker straight to the on board attenuators only to finally uncover some decent natural sounding music. I'm not sure I could ever go back to listening in I2S mode anymore.
So now it's apparent that a bit more thought needs to go into the digital side. Suggestions welcome, more research required. Perhaps re-clocking/ buffering on board is not totally necessary? Maybe just some jumpers to bypass the re-clocker if people don't want or need it?
At this stage I think it would be good to get some input from how people intend to use this pcb. I assume most will be using Ians gear?
Ryan
Hi Rayan,
i recently discovered the sound of the Philips tda 1541/1541A even the 1543
i managed to get my hands on 2 cd players with the tda1541 and another 2 players with the tda1541A
the sound produced by these players is truly outstanding i am looking to get this sound quality when playing music from my pc
i own a pro-ject dac with 4 tda1543, the sound is good but nothing like the 1541/A
the players i`m talking about are:
TDA1541
Sony cdp-710 (fantastic build and sound for a cheap player)
Philips cd-650 (not as well build but similar sound)
TDA1541A
Sony cdp-750 (sounds bad without modding)
Grundig 8400 MK 2 (outstanding sound)
i recently discovered the sound of the Philips tda 1541/1541A even the 1543
i managed to get my hands on 2 cd players with the tda1541 and another 2 players with the tda1541A
the sound produced by these players is truly outstanding i am looking to get this sound quality when playing music from my pc
i own a pro-ject dac with 4 tda1543, the sound is good but nothing like the 1541/A
the players i`m talking about are:
TDA1541
Sony cdp-710 (fantastic build and sound for a cheap player)
Philips cd-650 (not as well build but similar sound)
TDA1541A
Sony cdp-750 (sounds bad without modding)
Grundig 8400 MK 2 (outstanding sound)
Hi xedox,
Only a few years ago I did just the same as you. I bought a Philips CD-470, and Later a Rotel RCD-855 - Both ended up heavily modified. Although I was always impressed by the music the 1541 played I was still curious to see how much better it could be. It can get much better.
With the PCB i'm working on, the aim is to provide diyers with a good foundation for the TDA1541A and build up from there. With a good pre-filtered low noise supply for the analog rails, good output stage (discrete, Cen/Sen), low jitter clocking for the digital lines (LE/BCK), FIFO, I2S-PCM, Isolation from the source, and your choice of DEM solution - there's no reason why you shouldn't have a DAC that will compare very well against some great commercial DACs.
Presently i'm really enjoying tweaking the prototype PCB, at the moment i'm just trying to get the tonal balance right. Clocking the DEM at 705k with bias from -15 seems to have cleaned up the highs nicely (thanks james!). As for the the bass - I think I need a few more uF on the MSBs. One other thing to note is i'm probably still under 100 hours of listening with all the new caps, etc, so still may need more time to settle in.
Not much I can comment on with the 1543, unfortunately i've never heard one.
Sorry for the late reply, i've been too busy listening to music and modding.
Ryan
Only a few years ago I did just the same as you. I bought a Philips CD-470, and Later a Rotel RCD-855 - Both ended up heavily modified. Although I was always impressed by the music the 1541 played I was still curious to see how much better it could be. It can get much better.
With the PCB i'm working on, the aim is to provide diyers with a good foundation for the TDA1541A and build up from there. With a good pre-filtered low noise supply for the analog rails, good output stage (discrete, Cen/Sen), low jitter clocking for the digital lines (LE/BCK), FIFO, I2S-PCM, Isolation from the source, and your choice of DEM solution - there's no reason why you shouldn't have a DAC that will compare very well against some great commercial DACs.
Presently i'm really enjoying tweaking the prototype PCB, at the moment i'm just trying to get the tonal balance right. Clocking the DEM at 705k with bias from -15 seems to have cleaned up the highs nicely (thanks james!). As for the the bass - I think I need a few more uF on the MSBs. One other thing to note is i'm probably still under 100 hours of listening with all the new caps, etc, so still may need more time to settle in.
Not much I can comment on with the 1543, unfortunately i've never heard one.
Sorry for the late reply, i've been too busy listening to music and modding.
Ryan
Hi Ryan,
sorry for misspelling your name in my previous post
i am really optimistic and eager to try your board, i`m good at soldering and have the required tools, hope you get those uF sorted out real soon
will try to get it working in combination with my usb to spdif/coax board based on the xmos u8 chip
good luck with the testing and fine tuning
sorry for misspelling your name in my previous post
i am really optimistic and eager to try your board, i`m good at soldering and have the required tools, hope you get those uF sorted out real soon
will try to get it working in combination with my usb to spdif/coax board based on the xmos u8 chip
good luck with the testing and fine tuning
Hi xedox,
No problem for the typo, nice to know you're keen on my pcb. At the moment i'm quite happy with the prototype in my system, it has not been a subtle improvement from my previous pcb (v1) at all.
I've added an extra 1uF CD film to the 4 MSB pins (paralleled with 220nF C0Gs) which gave a substantial improvement on the impact of the music, including the bass. I will make extra room on the PCB for extra capacitance.
The DEM pins have 2, 50pF C0G/NP0 TDK ceramics in parallel - Providing about 960 kHz oscillation. A good quality cap here seems more important than the exact value of the oscillator frequency.
Is the "spdif/coax board based on the xmos u8 chip" pcb your own design? Did you program the xmos chip yourself? Simultaneous data output?
A design consideration for myself at the moment is whether to either remove the reclocking section on the PCB or provide a "backdoor" to bypass the reclocker. At the moment i've got it bypassed straight into the attenuators with no local buffering and the waveform looks quite alright considering the impedance mismatch with the jumper wires. Advice anyone?!
So far all the improvements in sound quality have been verified in my system using AKG 701 headphones through my F5 amp, although room acoustics in my case is prefered. I've never quite liked listening through headphones but at the moment it's never provided such a comfortable listening experience as it does currently.
Ryan
No problem for the typo, nice to know you're keen on my pcb. At the moment i'm quite happy with the prototype in my system, it has not been a subtle improvement from my previous pcb (v1) at all.
I've added an extra 1uF CD film to the 4 MSB pins (paralleled with 220nF C0Gs) which gave a substantial improvement on the impact of the music, including the bass. I will make extra room on the PCB for extra capacitance.
The DEM pins have 2, 50pF C0G/NP0 TDK ceramics in parallel - Providing about 960 kHz oscillation. A good quality cap here seems more important than the exact value of the oscillator frequency.
Is the "spdif/coax board based on the xmos u8 chip" pcb your own design? Did you program the xmos chip yourself? Simultaneous data output?
A design consideration for myself at the moment is whether to either remove the reclocking section on the PCB or provide a "backdoor" to bypass the reclocker. At the moment i've got it bypassed straight into the attenuators with no local buffering and the waveform looks quite alright considering the impedance mismatch with the jumper wires. Advice anyone?!
So far all the improvements in sound quality have been verified in my system using AKG 701 headphones through my F5 amp, although room acoustics in my case is prefered. I've never quite liked listening through headphones but at the moment it's never provided such a comfortable listening experience as it does currently.
Ryan
Hi Ryan, try getting one of the 8 X parallel tda1543 dacs on ebay some time. I got a used one for a hundred or so and I was quite impressed. I think its a chip I could imagine experimenting with. But I have a cue of things I want to build.
hi Xedox, I modded a Philips CD650 and the clock upgrade with its own little power supply was an amazing difference. Have you tried a clock upgrade?
Hi Luke,
I've always been curious about that little chip but have been too involved with other projects. I think one DAC project is enough for me at the moment. But thanks for the tip.
Ryan
Hello to all followers of this thread,
@Luke
I havent got so far with my 650, i got this silver unit in mint optical condition for little money, so far i recapped then added some capacitors and new op amps plus sockets also i removed 4 transistors before the rca output.
I am so impressed by this player i have no words. At the moment im listening to all my cd`s again the music sounds so alive the way i wanted to hear music.
The fact that this unit can get better just makes me smile, i`ll go that way but right now i`m just listening
I am happy to receive your tips on reclocking the 650 if you would like to share your experiences.
@Ryan
regarding that xmos U8 chip board, i purchased online at:
hxxp://diyhifishop.com/new-xmos-u8-decoder-usb-to-optical-coaxial-384k-p-140.html?zenid=7g4hcfv2e7i2uifujfdo2r03a1
i`m still in the testing phase, it`s good size without need of external power.
@Luke
I havent got so far with my 650, i got this silver unit in mint optical condition for little money, so far i recapped then added some capacitors and new op amps plus sockets also i removed 4 transistors before the rca output.
I am so impressed by this player i have no words. At the moment im listening to all my cd`s again the music sounds so alive the way i wanted to hear music.
The fact that this unit can get better just makes me smile, i`ll go that way but right now i`m just listening
I am happy to receive your tips on reclocking the 650 if you would like to share your experiences.
@Ryan
regarding that xmos U8 chip board, i purchased online at:
hxxp://diyhifishop.com/new-xmos-u8-decoder-usb-to-optical-coaxial-384k-p-140.html?zenid=7g4hcfv2e7i2uifujfdo2r03a1
i`m still in the testing phase, it`s good size without need of external power.
I know what you mean. I want to finish a couple more before starting a new project.
Hello to all followers of this thread,
@Luke
I havent got so far with my 650, i got this silver unit in mint optical condition for little money, so far i recapped then added some capacitors and new op amps plus sockets also i removed 4 transistors before the rca output.
I am so impressed by this player i have no words. At the moment im listening to all my cd`s again the music sounds so alive the way i wanted to hear music.
The fact that this unit can get better just makes me smile, i`ll go that way but right now i`m just listening
I am happy to receive your tips on reclocking the 650 if you would like to share your experiences.
@Ryan
regarding that xmos U8 chip board, i purchased online at:
hxxp://diyhifishop.com/new-xmos-u8-decoder-usb-to-optical-coaxial-384k-p-140.html?zenid=7g4hcfv2e7i2uifujfdo2r03a1
i`m still in the testing phase, it`s good size without need of external power.
use its a good player. It was a long time ago and the player was sold as I now use a PC and no CD.
The clock circuit was from here, I think Guido was the designers name. I used a finesse regulator and a 3va transformer. The clock and its power supply was on one board.
Hi Walter,
I'll try, but everyone - just remember that a lot has changed other than the layout and psu design-
-Powering the analog and digital supplies off batteries.
-The 14 decoupling caps are differently configured.
-DEM running at higher frequency and also better quality cap.
Having said that...
My subjective opinion presently, based on a completely unscientific approach (no level matching, no blindfolds).
Simply listening to familiar music with an as open mind as possible:
I've never enjoyed music as much as I am currently, and still under 100 hours of play time with all the new parts. I'd have it on all the time to speed up the process but I have no battery protection for the SLAs.
At the moment there is one main change that stands out the most - Clarity and detail in cymbals and percussion in general. Overall it feels like i'm getting closer to listening to a "performance".
One other thing that isn't necessarily always a good change is noticing characteristics of the recording itself. I'm noticing a large variation in the quality of the recordings more than i did previously. Some albums that I thought were really great recordings seemed to sound even better with more detail, better imaging and seemingly smoother dynamics. Then other albums that used to sound pretty good, are now exposing possibly poor recording quality in general, lack of detail, imaging is all over the place, etc.
So far it's been an enjoyable process.
After giving this subjective opinion I must say that what I am hearing currently is not what I was expecting in the beginning, not that i've got any complaints so far. For some reason I was expecting a more dynamic energetic sound, but instead I got smooth, realistic dynamics... Interesting stuff.
Ryan.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.