Re: cd94 cd960
I have a big grin because I know how wonderful your sound will be there. There's just something in the instrument texture and dynamics that is musical with the TDA1541... and the '94 seems fairly well-sorted (when modded with regs, caps, clock etc.) so that must apply to your Philips too.
I ordered a TDA1541A N2 chip today. I'll see if that has anything/much over the TDA1541A I'm using now.
Simon
UV101 said:Simon - your comments on the 63 mods thread about me still listening to my 63 were just a few hrs too early! you were right I'm converted!
I have a big grin because I know how wonderful your sound will be there. There's just something in the instrument texture and dynamics that is musical with the TDA1541... and the '94 seems fairly well-sorted (when modded with regs, caps, clock etc.) so that must apply to your Philips too.
I ordered a TDA1541A N2 chip today. I'll see if that has anything/much over the TDA1541A I'm using now.
Simon
It must be a good day for this DAC chip, with all these successes. I've sorted out my ground problems, and the dual TDA1541A & PMD100 now sounds awesome!
I'm sure its all 'frame of reference' stuff, but its the best sound I've had from anything I've built or modded - and there's tons of improvements that can be done (decoupling, regulation (especially for clock), reclocking, TDA1541A input rescaling and filtering, DEM reclocking, discrete IV stage, output buffer, it just goes on) .
TDA1541A N2? I'm not sure if that's particularly special. I recall R1 is a lesser part, and the S1 and S2 parts are the crowned parts, selected for quality.
Also, Simon, as you've upgraded to an A, did you fit the 470pF cap across pins 16&17? It should make an improvement.
I'm sure its all 'frame of reference' stuff, but its the best sound I've had from anything I've built or modded - and there's tons of improvements that can be done (decoupling, regulation (especially for clock), reclocking, TDA1541A input rescaling and filtering, DEM reclocking, discrete IV stage, output buffer, it just goes on) .
TDA1541A N2? I'm not sure if that's particularly special. I recall R1 is a lesser part, and the S1 and S2 parts are the crowned parts, selected for quality.
Also, Simon, as you've upgraded to an A, did you fit the 470pF cap across pins 16&17? It should make an improvement.
jnewbold said:
Joe do you have a scope? If you have look at the voltage on 5V PS of 7220. After shunt reg i only had some very fast small spikes left, about 8mV top-top.
Further nice mods can be some beads to different dig chips, and stopping R's of 47 or 100 ohms in I2S signals (smd, cut trace a little and solder it on) By doing that you help to short out ongoing HF signals through all chips. It keeps the noise on 1 particular chip, the (oscon) cap near it filters better then. A nice mod can be also: keep original lytic, and solder a small tantal SMD (computer) cap on the lytic pins extra.
I wanted to do that too (the pmd)
Well, I was being stupid essentially. I was assuming I should keep digital and analogue grounds seperate - which it appears was the exact opposite of what I should have been doing!
I doubt anyone else would have this problem, and it was a generic problem - rather than something to do with dual DACs.
I've just got some Mundorf caps off Lee (I think, off ebay anyway), which I shall use as output caps. That'll be the next mod, nice and simple. I also want to use some shielded audio cable for the analogue signals, and some nice neutrik phono sockets. I doubt the cabling will make much difference, but it will be nice to do for completeness.
I doubt anyone else would have this problem, and it was a generic problem - rather than something to do with dual DACs.
I've just got some Mundorf caps off Lee (I think, off ebay anyway), which I shall use as output caps. That'll be the next mod, nice and simple. I also want to use some shielded audio cable for the analogue signals, and some nice neutrik phono sockets. I doubt the cabling will make much difference, but it will be nice to do for completeness.
, thank you.Well, firstly, you don't need to control the PMD100 with a PIC. I don't and it works fine for TDA1541A with 8x oversampling. I have a set of DIP switches just to configure a few things. If there's some particular special feature you're after then you'll need to use a PIC to send commands via a serial data port.
You need to set the chip into standalone mode (as opposed to program mode) by setting the logic level on a pin.
The only tricky bit is the format conversion. To convert from and back to I2S, you need some TTL sequential logic. I essentially used that designed by Guido...
http://www.diyhifi.org/forums/viewt...=t&sd=a&hilit=pmd100+pmd100+tda1541a&start=15
(However, Guido went on further than me and developed it all with a PIC).
I won't post a schematic myself, because I didn't really design anything here. I gleaned all the information from the PMD100 datasheet, and the aforementioned thread.
You need to set the chip into standalone mode (as opposed to program mode) by setting the logic level on a pin.
The only tricky bit is the format conversion. To convert from and back to I2S, you need some TTL sequential logic. I essentially used that designed by Guido...
http://www.diyhifi.org/forums/viewt...=t&sd=a&hilit=pmd100+pmd100+tda1541a&start=15
(However, Guido went on further than me and developed it all with a PIC).
I won't post a schematic myself, because I didn't really design anything here. I gleaned all the information from the PMD100 datasheet, and the aforementioned thread.
philpoole said:
No, but you keep reminding me (it's appreciated) so I'll get to it. I might tack this onto the top so as to avoid removing the PCB. Is a cheap ceramic cap suitable for this job? I don't know what it's doing... the non-A datasheet says the pins are N/C and the A version datasheet has them down as "oscillator".
Congratulations on your success with the DIY DAC, that's wonderful. I'd dearly love to hear it. Maybe if/when this next meet occurs! Would you make it up as far as Chesterfield again?
Simon
I might try dropping in an extra transformer, power supply and S Power regulators (parts I have already) for the I/V / output op-amps this evening.
Another mod - to move the entire player onto a sheet of BB plywood - can happen soon too, as I have the plywood for my new speakers and got them to cut a piece for my CD player whilst I was at Arnold Laver. Lovely 24mm Latvian birch. For those who believe in vibration and tone etc. you will know this should be a positive thing for sound quality.
Simon
Another mod - to move the entire player onto a sheet of BB plywood - can happen soon too, as I have the plywood for my new speakers and got them to cut a piece for my CD player whilst I was at Arnold Laver. Lovely 24mm Latvian birch. For those who believe in vibration and tone etc. you will know this should be a positive thing for sound quality.
Simon
philpoole said:
Phil,
Doubt is healthy but try to keep an open mind about component quality. It will ultimately allow you to hear a very relaxed musical presentation IME. Things like connectors and wires all add up - a tiny bit each. If they're all rubbish changing one won't help much, but changing every piece of wire and connector for good ones will make every subsequent change easier to discern and appreciate.
With this in mind, perhaps think about buying some good RCAs, such as any from CMC - http://www.diyhi-fi.co.uk/html/connectors.html. One day I'll get in trouble for recommending Lee's parts too much
I once tried some Neutrik Pro-Fi phono plugs and, honestly, they're terrible. I can't see the sockets being all that much better. Speakons also sound poor - I know because I used to use some.
BTW - why do you want to use shielded wire? It's probably not needed and will only harm sound quality.
I hope I don't start an unpleasant debate here, I just want to help
I added a 11 Mhz kwak clock to SAA7220. And fed dac with 1/2 clock with a HCT163 divider. the 163 has some more divider outputs, took /8 and another HCT163 with /4 gives 352.8kHz for pin 16. With a HCT04 inverter a inverted dem signal is created for pin 17
In this thread it is explained further.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=11949&perpage=25&pagenumber=6
with all the bells and whistles a 1541 can sound very good. Very natural, slightly warm but not coloured as unmodded 1541 players can sound. Very musical, that's how it can perform.
(paypal is received Simon)
In this thread it is explained further.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=11949&perpage=25&pagenumber=6
with all the bells and whistles a 1541 can sound very good. Very natural, slightly warm but not coloured as unmodded 1541 players can sound. Very musical, that's how it can perform.
(paypal is received Simon)
Jan,
Superb information, thanks. I've used a "flip-flop" chip once before, for servo reclocking in my old CD63, but I did manage to make a hash of it, so I'll probably leave this idea on the back-burner until easy things are done. I've realised it's sometimes wiser to maintain a working player rather than a very upgraded dead player!!
Simon
ps - my Marantz MA500s greedily await their feedback capacitor upgrade
Superb information, thanks. I've used a "flip-flop" chip once before, for servo reclocking in my old CD63, but I did manage to make a hash of it, so I'll probably leave this idea on the back-burner until easy things are done. I've realised it's sometimes wiser to maintain a working player rather than a very upgraded dead player!!
Simon
ps - my Marantz MA500s greedily await their feedback capacitor upgrade
My DAC chip now has 470p across pins 16 & 17. It didn't make any noise at first, I think due to the chip not being in firmly enough.. but it works now.
This seems better after just a few seconds of listening, I'm surprised! Low level detail has come out, as if the noise floor has been lowered. Is this what I should expect with this cap in place??
edit:
After hearing some intimately familiar tracks (Carol Kidd - Debut) I also think a layer of grime has come away from the treble. This was needed, and is a nice little step in the right direction.
Anyone who's upgraded their DAC chip to an "A" version really must drop a 470p cap on pins 16 & 17.
Simon
This seems better after just a few seconds of listening, I'm surprised! Low level detail has come out, as if the noise floor has been lowered. Is this what I should expect with this cap in place??
edit:
After hearing some intimately familiar tracks (Carol Kidd - Debut) I also think a layer of grime has come away from the treble. This was needed, and is a nice little step in the right direction.
Anyone who's upgraded their DAC chip to an "A" version really must drop a 470p cap on pins 16 & 17.
Simon
Probably a styro cap will do allright there
Here is the demodulator signal with dem reclock. The 1541 demodulates with the caps around it the digital signal. A capacitor along p16-17 makes a floating frequency. Dem reclock locks the demodulator frequency with the clock signal: a Dem mod does as much as a better clock, but only if it is done right (with pos-neg signals to 16-17)
Not the best signal but could not get a good lock with the old scope.
Yet another removing of excessive warmth will it bring. I didn know a 1541 could be a hint analytic but still musical
Here is the demodulator signal with dem reclock. The 1541 demodulates with the caps around it the digital signal. A capacitor along p16-17 makes a floating frequency. Dem reclock locks the demodulator frequency with the clock signal: a Dem mod does as much as a better clock, but only if it is done right (with pos-neg signals to 16-17)
Not the best signal but could not get a good lock with the old scope.
An externally hosted image should be here but it was not working when we last tested it.
Yet another removing of excessive warmth will it bring. I didn know a 1541 could be a hint analytic but still musical