hi,the kit is designed by my buddy.
I found in the TDA1541 NOS DAC(not TDA1543) that the de-couple cap of CS8412 may obviously change the sound,you can try....
http://www.hifidiy.net/dispbbs.asp?boardID=4&ID=14231&skin=0
I found in the TDA1541 NOS DAC(not TDA1543) that the de-couple cap of CS8412 may obviously change the sound,you can try....
http://www.hifidiy.net/dispbbs.asp?boardID=4&ID=14231&skin=0
An externally hosted image should be here but it was not working when we last tested it.
Hi Bgt,
I noticed two coupling SPDIF capacitors on board. Remove both and use pulse transformer to feed CS8412. Try S22083 by Newava Technology. This will improve the sound a LOT!
S22160 is also very good.
Once you tried this transformer, next step is to buffer the secondary windings' signal before it hits CS chip. Search the forums for diagram. Loading is very important, as is the type of coax cable, length and use of BNC connectors to preserve proper impedance.
I tried the coax cable used for oscilloscope probes (very low capacitance per meter, I think 4pF only!) It sounds very good.
Regards,
Extreme_Boky
I noticed two coupling SPDIF capacitors on board. Remove both and use pulse transformer to feed CS8412. Try S22083 by Newava Technology. This will improve the sound a LOT!
S22160 is also very good.
Once you tried this transformer, next step is to buffer the secondary windings' signal before it hits CS chip. Search the forums for diagram. Loading is very important, as is the type of coax cable, length and use of BNC connectors to preserve proper impedance.
I tried the coax cable used for oscilloscope probes (very low capacitance per meter, I think 4pF only!) It sounds very good.
Regards,
Extreme_Boky
Hi all
I am from HifiDiy. I am excited with our DAC is being discussed here. As this is a non-commercial website, I should not go about publicizing how value packed our product is for the price they sell. What we do want you know is we welcome any suggestion and comment with regards to our product. We view your feedback is crucial to our growth. Please feel free to send us your feedback if you have any issue to our service or product. We believe that in order to succeed, we have to learn to listen to our shortcoming. Here is some comment from one of our Japanese user on the modification he did for the DAC. I hope you will find it useful.
------------------------------------------------------------------------------------
29th Oct 2005
The DAC is finally completed. Here are some of the modifications from the stock set:
Rectifers are now OnSemi MBR1100
Digital section, power supply caps are now Nichicon Muse FX.
Analogue section, the power supply caps are now Elna ROD
The DAC has just begun running in. It would take some time. As it is, the DAC has potential; it sounds musical and a little lazy. In comparing with another NOS 1541 (not HifiDiy 1541), this 1543 does not have the same level of analytical details. It¡¯s strength lies in its deep and warm mid range. No too bad for the value to price ratio.
The aluminum plate that acts as the heatsink does not seems to have a close contact with the TDA1543 chips. To solve that, I just add a piece of 1mm thick copper plate between them. The improvement of the 1mm copper plate is the tightening of sound, it is now more solid.
I rather hope after long hours of running in this DAC, the high frequency would be smoother and more fluid.
After one night of running in, the sound changes are significant. There are things I don¡¯t like. I am going to switch some components:
The ripple filter caps at the DAC and the digital section are changed to KMG 6800uf/25V (sounds improve in transparency)
The e-caps at the output of the DAC section voltage regulator are now MUSE KZ 470uf/50V.
I have reinforced the DAC +8V power line using braided cables. The power caps has been given an additional 1.5uf film caps for bypass.
The output is now Muse KZ 100uf/50V with 0.01uf bypass
The caps for LPF 2200pf and 470 pf are all now MIWA( as according to the original article). The original has more high frequency noise.
The power ripple caps has added a 2.2uf film cap in parallel.
The I/V stage resistor are now RMG 4701W, carbon film resistor. In my personal opinion, this resistor sounds more transparent as compared to the metal film resistor.
With all the above changes, the high is beginning to emerge. But on the hand, the bass has weakened slightly. Maybe a EI or R core transformer would be better, this toraidol transformer sounds thin in comparison. Overall the sound is still warm, it is not very analytical nor does it has a extremely wide sound stage, but the female vocal sound great.
Currently the output stage power supply is at 18VDC and the DAC is at 8VDC. Both the voltage regulator and the DAC are running hot.
The sound characteristic is starting to stabilized, gradually mellowing towards my preference. High frequency is not too bright, mid range is warm and musical and bass is lean and having strength. It is like a pyramid; sharp highs follow by warm mid build on a strong bass.
The orchestral piece is very beautiful, just that the symbols and percussion are stiff. Although the sound stage is not definitive, it does has impressive depth..
Having said all the good things about this 1543, I find that it still does not have the same amount of details as compared to TDA1541 or PCM58. In general, it sound nostalgic and does not have the amount of details of new generation DAC. The speed and the dynamic are not outstanding. The sound is not crisp and detail like a modern DAC, it is like comparing a beautiful oil painting to a photograph. Maybe the OPA604 has a major part in the makeup.
Honestly, preference for such sound characteristic is a rather subjective matter. I personally enjoy the relaxing and tuby type of sound. This DAC has a lot that matches my taste. But if you were to use this DAC on rock and roll, hiphop or dance music, you may not be fully satisfied with its result. This DAC scores better on orchestral pieces than jazz for live music and instrumental pieces. The sound of double bass in a string quartet is simply addictive.
Conclusion
I am currently running the DAC in with some rock and roll tracks. The bass has improved in delivering convincing impact, the high is now more open.
The level of difficulty in building this DAC is the same as others. Just that for the surface mounted components, you will have to use thinner gauge of solder and apply soldering paste.
The stock elna caps makes the DAC sounds really flat and lacking in dimension. The LPF caps with lead lead does not sound too good, not recommenced.
If you plan for switching the I/V stage resistor, the best is use an adaptor, just plug the resistor on to the adaptor. Stock some IC adaptor before hand (8/14/16 pin type) for the clock and the components around that section.
On the whole, this DAC is warm and musical; it is very relaxing to listen to. On the other hand, it does not stands out for having analytical power, details and speed..
I am from HifiDiy. I am excited with our DAC is being discussed here. As this is a non-commercial website, I should not go about publicizing how value packed our product is for the price they sell. What we do want you know is we welcome any suggestion and comment with regards to our product. We view your feedback is crucial to our growth. Please feel free to send us your feedback if you have any issue to our service or product. We believe that in order to succeed, we have to learn to listen to our shortcoming. Here is some comment from one of our Japanese user on the modification he did for the DAC. I hope you will find it useful.
------------------------------------------------------------------------------------
29th Oct 2005
The DAC is finally completed. Here are some of the modifications from the stock set:
Rectifers are now OnSemi MBR1100
Digital section, power supply caps are now Nichicon Muse FX.
Analogue section, the power supply caps are now Elna ROD
The DAC has just begun running in. It would take some time. As it is, the DAC has potential; it sounds musical and a little lazy. In comparing with another NOS 1541 (not HifiDiy 1541), this 1543 does not have the same level of analytical details. It¡¯s strength lies in its deep and warm mid range. No too bad for the value to price ratio.
The aluminum plate that acts as the heatsink does not seems to have a close contact with the TDA1543 chips. To solve that, I just add a piece of 1mm thick copper plate between them. The improvement of the 1mm copper plate is the tightening of sound, it is now more solid.
I rather hope after long hours of running in this DAC, the high frequency would be smoother and more fluid.
After one night of running in, the sound changes are significant. There are things I don¡¯t like. I am going to switch some components:
The ripple filter caps at the DAC and the digital section are changed to KMG 6800uf/25V (sounds improve in transparency)
The e-caps at the output of the DAC section voltage regulator are now MUSE KZ 470uf/50V.
I have reinforced the DAC +8V power line using braided cables. The power caps has been given an additional 1.5uf film caps for bypass.
The output is now Muse KZ 100uf/50V with 0.01uf bypass
The caps for LPF 2200pf and 470 pf are all now MIWA( as according to the original article). The original has more high frequency noise.
The power ripple caps has added a 2.2uf film cap in parallel.
The I/V stage resistor are now RMG 4701W, carbon film resistor. In my personal opinion, this resistor sounds more transparent as compared to the metal film resistor.
With all the above changes, the high is beginning to emerge. But on the hand, the bass has weakened slightly. Maybe a EI or R core transformer would be better, this toraidol transformer sounds thin in comparison. Overall the sound is still warm, it is not very analytical nor does it has a extremely wide sound stage, but the female vocal sound great.
Currently the output stage power supply is at 18VDC and the DAC is at 8VDC. Both the voltage regulator and the DAC are running hot.
The sound characteristic is starting to stabilized, gradually mellowing towards my preference. High frequency is not too bright, mid range is warm and musical and bass is lean and having strength. It is like a pyramid; sharp highs follow by warm mid build on a strong bass.
The orchestral piece is very beautiful, just that the symbols and percussion are stiff. Although the sound stage is not definitive, it does has impressive depth..
Having said all the good things about this 1543, I find that it still does not have the same amount of details as compared to TDA1541 or PCM58. In general, it sound nostalgic and does not have the amount of details of new generation DAC. The speed and the dynamic are not outstanding. The sound is not crisp and detail like a modern DAC, it is like comparing a beautiful oil painting to a photograph. Maybe the OPA604 has a major part in the makeup.
Honestly, preference for such sound characteristic is a rather subjective matter. I personally enjoy the relaxing and tuby type of sound. This DAC has a lot that matches my taste. But if you were to use this DAC on rock and roll, hiphop or dance music, you may not be fully satisfied with its result. This DAC scores better on orchestral pieces than jazz for live music and instrumental pieces. The sound of double bass in a string quartet is simply addictive.
Conclusion
I am currently running the DAC in with some rock and roll tracks. The bass has improved in delivering convincing impact, the high is now more open.
The level of difficulty in building this DAC is the same as others. Just that for the surface mounted components, you will have to use thinner gauge of solder and apply soldering paste.
The stock elna caps makes the DAC sounds really flat and lacking in dimension. The LPF caps with lead lead does not sound too good, not recommenced.
If you plan for switching the I/V stage resistor, the best is use an adaptor, just plug the resistor on to the adaptor. Stock some IC adaptor before hand (8/14/16 pin type) for the clock and the components around that section.
On the whole, this DAC is warm and musical; it is very relaxing to listen to. On the other hand, it does not stands out for having analytical power, details and speed..
Continue
31st Oct 2005
My previous NOS 1543 dac, DAC-AH has quite a significant amount of noise, so I am measuring the 1543 DAC from HifiDiy to see if this problem is also eminent.
Test Equipment
Amp: SONY TA-F505ESD
Speaker: Sonas Faber MINIMA
Cable: ACROTEC 6N-1010
The system was warm up for three hours of continues play before testing begin. Playing the silent track of SONY test CD, no.4, the noise was almost inaudible. Maybe with a good headphone amp, it would be more accurate, but too bad I don¡¯t have on hand now. As it is, there would not be any noise problem for normal usage.
Now after the DAC has more or less stabilized, lets talk about its sound.
The uncomfortable sharpness is now gone. Female vocal intonation is not an issue. Mid bass are now more stable. The tonal balance is more balanced now, this could be due to the stabilizing of mid highs. The feeling of the music being moved forward is now gone. Both the left and right channel are now more open, the sound stage has improved in width.
To test the sound stage, I played an orchestral piece on the system. The violin was very meaty and fluid. This was a lot better than my expectation; originally I perceived this DAC is only capable of presenting sophisticated instrumental pieces, it is able to do a complex orchestral piece convincingly.
Maybe the DAC is not analytical enough, Japanese pop songs sounds not too bad. The effect of all the recording fault and compression problem are being masked by its strength.
This could be a problem for recordings with very high emphasis and volume of high frequency sound, it will sound a bit strange.
In contrast, the 80s AOR would be very suitable for this DAC, try ±ÈÀûÇÇ.
--------------------------------------------------------------------------------
3rd Nov 2005
Opamp selection
604 does not have any flaws; beautiful music with a unique shine in its highs.
797 is not as luscious as the 604, but 797 sounds clear and real
5534 has unique flaws in the highs; it is not as clear and transparent in comparison to the above two.
I like the 797 best with this DAC. (From translator: the author¡¯s final choice was 604, the author did not give any reason or explanation for his changes)
For the output stage power supply caps
The stock Panasonic 5600uf caps sounds funny in mid range. Female singer sounds if she had a slightly block nose.
DUOREX 4700, the whole frequency range sounds clear and uncoloured, a little unexciting.
DUOREX 6800, the bass has a small improvement, but its size is too big for the PCB
Great 11200, frank and lively music but size just don¡¯t fit.
Constraint by size, I could only use DUOREX 4700.
Digital section power supply capacitor selection
The Panasonic 5600 cap, the sound has lesser clarity and is flat. It is as if it has been squashed on a piece of paper.
KMG 6800, the sound is straight forward, forceful. The highs are slightly reduced.
Muse FX 3300, the sound is more mature; the high extension are better, the bass is more backward.
For now, my pick would be the KMG to go with this DAC. (the author switch to MUSE later on)
I/V resistor
KOA metal film, 0.5w 680R, sounds slight hard and rigid and obstructed.
KOA carbon film, very open sounding, not as much detail, but warmer than the KOA metal film
Riken RMG carbon film1W470, no as warm as the KOA carbon, but has lower noise level
Tokyo Ko-on RD carbon film, the sound is lower than RMG, a lot more detail than RMG. Highs are more mature and luscious.
--------------------------------------------------------------------------------
5th Nov 2005
The opamp was changed to AD797, the high extension did not improved, I have decided to switch back to 604. As for why could it be so, my guess would be the other components are focusing on the bass section, the highs are rolling off too fast.
Before switching back to 604, the highs always seem to be from a far and away distance. With 604, the balance between the high and the mid bass are restored onto the same plane. Overall, the mid and bass is still too closely tied up. There are not enough separation between the notes, the sound stage is not good enough. This DAC does not have sufficient analytical power for a clear sound stage. Even though the sound stage is width, but the music is chunky and lumpy. Within my 30m2 working space, the sound is full and wide, it does not sound strained on my system.
In terms of compatibility with others, this DAC would be more suitable for tube amps than the more analytical solid state amps. But if the tube amp were not of the same class, you might have a blurring of the sound stage.
In conclusion, if you want to have critical and analytical details in music, this is not the DAC for you. But if you want to have full and balanced kind of music, this would be a very good choice. In comparing to current high sampling rate type of player, I believe there would many who would have very different opinion about this DAC. Ultimately, I think it is a matter of choice.
-----------------------------------------------------------------------------
6th Nov 2005
In the end, I have the power supply caps at the digital section switched from KMG6800/25V to Muse FW 6800uf/25V. The high extension are better, with no change to the bass and the mids. I/V resistor is now switched to gold plated lead RMG 1/2W 680R, the whole spectrum of sound seems to have been lifted just a lttle. As compared to the normal RMG (without gold plated lead), it has improvement in the details.
Last suggestion, for resistor and caps, it would be better if you demagnetized before installation. It will sound smoother and speeden up the running in.
The End.
-----------------------------------------------------------------------------
Link to the author¡¯s comment over a Japanese DIY NOS DAC forum
http://hobby8.2ch.net/test/read.cgi/pav/1121339259/
31st Oct 2005
My previous NOS 1543 dac, DAC-AH has quite a significant amount of noise, so I am measuring the 1543 DAC from HifiDiy to see if this problem is also eminent.
Test Equipment
Amp: SONY TA-F505ESD
Speaker: Sonas Faber MINIMA
Cable: ACROTEC 6N-1010
The system was warm up for three hours of continues play before testing begin. Playing the silent track of SONY test CD, no.4, the noise was almost inaudible. Maybe with a good headphone amp, it would be more accurate, but too bad I don¡¯t have on hand now. As it is, there would not be any noise problem for normal usage.
Now after the DAC has more or less stabilized, lets talk about its sound.
The uncomfortable sharpness is now gone. Female vocal intonation is not an issue. Mid bass are now more stable. The tonal balance is more balanced now, this could be due to the stabilizing of mid highs. The feeling of the music being moved forward is now gone. Both the left and right channel are now more open, the sound stage has improved in width.
To test the sound stage, I played an orchestral piece on the system. The violin was very meaty and fluid. This was a lot better than my expectation; originally I perceived this DAC is only capable of presenting sophisticated instrumental pieces, it is able to do a complex orchestral piece convincingly.
Maybe the DAC is not analytical enough, Japanese pop songs sounds not too bad. The effect of all the recording fault and compression problem are being masked by its strength.
This could be a problem for recordings with very high emphasis and volume of high frequency sound, it will sound a bit strange.
In contrast, the 80s AOR would be very suitable for this DAC, try ±ÈÀûÇÇ.
--------------------------------------------------------------------------------
3rd Nov 2005
Opamp selection
604 does not have any flaws; beautiful music with a unique shine in its highs.
797 is not as luscious as the 604, but 797 sounds clear and real
5534 has unique flaws in the highs; it is not as clear and transparent in comparison to the above two.
I like the 797 best with this DAC. (From translator: the author¡¯s final choice was 604, the author did not give any reason or explanation for his changes)
For the output stage power supply caps
The stock Panasonic 5600uf caps sounds funny in mid range. Female singer sounds if she had a slightly block nose.
DUOREX 4700, the whole frequency range sounds clear and uncoloured, a little unexciting.
DUOREX 6800, the bass has a small improvement, but its size is too big for the PCB
Great 11200, frank and lively music but size just don¡¯t fit.
Constraint by size, I could only use DUOREX 4700.
Digital section power supply capacitor selection
The Panasonic 5600 cap, the sound has lesser clarity and is flat. It is as if it has been squashed on a piece of paper.
KMG 6800, the sound is straight forward, forceful. The highs are slightly reduced.
Muse FX 3300, the sound is more mature; the high extension are better, the bass is more backward.
For now, my pick would be the KMG to go with this DAC. (the author switch to MUSE later on)
I/V resistor
KOA metal film, 0.5w 680R, sounds slight hard and rigid and obstructed.
KOA carbon film, very open sounding, not as much detail, but warmer than the KOA metal film
Riken RMG carbon film1W470, no as warm as the KOA carbon, but has lower noise level
Tokyo Ko-on RD carbon film, the sound is lower than RMG, a lot more detail than RMG. Highs are more mature and luscious.
--------------------------------------------------------------------------------
5th Nov 2005
The opamp was changed to AD797, the high extension did not improved, I have decided to switch back to 604. As for why could it be so, my guess would be the other components are focusing on the bass section, the highs are rolling off too fast.
Before switching back to 604, the highs always seem to be from a far and away distance. With 604, the balance between the high and the mid bass are restored onto the same plane. Overall, the mid and bass is still too closely tied up. There are not enough separation between the notes, the sound stage is not good enough. This DAC does not have sufficient analytical power for a clear sound stage. Even though the sound stage is width, but the music is chunky and lumpy. Within my 30m2 working space, the sound is full and wide, it does not sound strained on my system.
In terms of compatibility with others, this DAC would be more suitable for tube amps than the more analytical solid state amps. But if the tube amp were not of the same class, you might have a blurring of the sound stage.
In conclusion, if you want to have critical and analytical details in music, this is not the DAC for you. But if you want to have full and balanced kind of music, this would be a very good choice. In comparing to current high sampling rate type of player, I believe there would many who would have very different opinion about this DAC. Ultimately, I think it is a matter of choice.
-----------------------------------------------------------------------------
6th Nov 2005
In the end, I have the power supply caps at the digital section switched from KMG6800/25V to Muse FW 6800uf/25V. The high extension are better, with no change to the bass and the mids. I/V resistor is now switched to gold plated lead RMG 1/2W 680R, the whole spectrum of sound seems to have been lifted just a lttle. As compared to the normal RMG (without gold plated lead), it has improvement in the details.
Last suggestion, for resistor and caps, it would be better if you demagnetized before installation. It will sound smoother and speeden up the running in.
The End.
-----------------------------------------------------------------------------
Link to the author¡¯s comment over a Japanese DIY NOS DAC forum
http://hobby8.2ch.net/test/read.cgi/pav/1121339259/
ahh, I've just read the documentation for the DAC2000 and it has an auto sampling circuit. The 1543 kit does not have this. I just measured my PC's output and it is 48khz indeed. My MD recorder also is an auto sampler thats why I never noticed the different sample rates. How to make the 1543 kit auto sampling?
Bgt said:
I don't have much information about your 1543kit, if there is a crystal not 6.144Mhz, the CS8414 should be running in slave mode.
you have to choose the following method to make you 1543 support 48Khz.
1./very simple,change the crystal frequency,it should be 11.2896Mhz,change it to 12.288Mhz,but it can't support 44.1Khz any more.
2./remove the any connection of pin 23 at CS8412/4,tied it to ground.It will put the receiver to master mode.after it,find pin11,12 at CS8412/4,find the connection which connected to the crystal clock/divider.there should be some logic IC,if everythins are correct,the DAC will support 32-96Khz automatically!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.