Universal timing chain schematics
Hi all,
D-I DAC update,
I set up a schematic diagram for a universal timing chain that supports 2,4,6,8,16 and 32 DAC chips. It uses 16 X SN74HC164 or HCT64, highest resolution is 2BCK. It can be used for experimenting with higher virtual sample rates for addittional harmonics amplitude reduction.
I also did some experimenting with high speed optocouplers (HPCL2630) to replace the isolation transformer in the SPDIF audio interface. This opto-coupler has very low capacitance between input and output, this reduces crosstalk quite a bit. I still use the differential RS422 driver DS8922 in the CD player. The HPCL2630 contains a dual opto-coupler, series resistor (LED) at 5V : 220 Ohms, open collector pull-up resistor : 390 OHm. This opto-coupler could also be used for isolating a I2S interface (2 X HPCL2630 needed) avoiding ground loops.
I have also been quite busy reducing interference signals in the octal D-I DAC. Perhaps reducing interference is even more important than focussing on the DAC output signal at the moment. Clean power supplies are also very important, I now use cascaded voltage regulators and parallelled electrolytic capacitors. The modifications made to the prototype to reduce interference have made the sound cleaner and smoother. SPDIF audio interface has to be set up very carefully, as this can be a major source for HF interference signals in the DAC. The opto-coupler made quite an improvement by reducing cross-talk to ground, the reduction is already audible, the octal D-I DAC is becomming very quit now. Perhaps using these opto-couplers for DEM clock synchronizing is not a bad idea, I am going to try this soon. It is also possible to fully isolate the DAC's chip digital input signals by using these opto-couplers.
The side panels of the octal D-I DAC have already been milled out of 8mm thick aluminum plates, they are the main support for the housing. I want to complete the housing first, so I can test noise figures with full screening. I am also planning to put a screening plate between the analog mainboard and the modules. The transformers of the tube power supply are now replaced by torroidial types as the existing E-I plate transformers generated weak mechanical 50Hz hum.
Hi all,
D-I DAC update,
I set up a schematic diagram for a universal timing chain that supports 2,4,6,8,16 and 32 DAC chips. It uses 16 X SN74HC164 or HCT64, highest resolution is 2BCK. It can be used for experimenting with higher virtual sample rates for addittional harmonics amplitude reduction.
I also did some experimenting with high speed optocouplers (HPCL2630) to replace the isolation transformer in the SPDIF audio interface. This opto-coupler has very low capacitance between input and output, this reduces crosstalk quite a bit. I still use the differential RS422 driver DS8922 in the CD player. The HPCL2630 contains a dual opto-coupler, series resistor (LED) at 5V : 220 Ohms, open collector pull-up resistor : 390 OHm. This opto-coupler could also be used for isolating a I2S interface (2 X HPCL2630 needed) avoiding ground loops.
I have also been quite busy reducing interference signals in the octal D-I DAC. Perhaps reducing interference is even more important than focussing on the DAC output signal at the moment. Clean power supplies are also very important, I now use cascaded voltage regulators and parallelled electrolytic capacitors. The modifications made to the prototype to reduce interference have made the sound cleaner and smoother. SPDIF audio interface has to be set up very carefully, as this can be a major source for HF interference signals in the DAC. The opto-coupler made quite an improvement by reducing cross-talk to ground, the reduction is already audible, the octal D-I DAC is becomming very quit now. Perhaps using these opto-couplers for DEM clock synchronizing is not a bad idea, I am going to try this soon. It is also possible to fully isolate the DAC's chip digital input signals by using these opto-couplers.
The side panels of the octal D-I DAC have already been milled out of 8mm thick aluminum plates, they are the main support for the housing. I want to complete the housing first, so I can test noise figures with full screening. I am also planning to put a screening plate between the analog mainboard and the modules. The transformers of the tube power supply are now replaced by torroidial types as the existing E-I plate transformers generated weak mechanical 50Hz hum.
Attachments
Optocoupler speed
Hi analog_sa,
Thanks for your reply [post#443],
It was an experiment to see if ground crosstalk could be significantly reduced, as the pulse transformers still caused significant crosstalk (34pF capacitance between primary and secundary winding). The HCPL2630 I tested had 0.2pF capacitance between the LED and receiver circuit, See what I mean. The HCPL2630 has 10Mb datarate. I arranged the HPCL2630 in a way that the positive going edge uses the faster 20nS risetime. I still use the balanced SPDIF cable from the CD player to the DAC. So the opto-coupler is only used for isolation over a very short distance. The signal looked much cleaner and The crosstalk to ground was significantly reduced due to the much lower capacitance. Faster optocouplers (40Mb datarate) are available like the HCPL2400 from Agilent, that should be fast enough.
However the opto coupler, if used, should not affect the sound quality in a negative way.
Hi analog_sa,
Thanks for your reply [post#443],
It was an experiment to see if ground crosstalk could be significantly reduced, as the pulse transformers still caused significant crosstalk (34pF capacitance between primary and secundary winding). The HCPL2630 I tested had 0.2pF capacitance between the LED and receiver circuit, See what I mean. The HCPL2630 has 10Mb datarate. I arranged the HPCL2630 in a way that the positive going edge uses the faster 20nS risetime. I still use the balanced SPDIF cable from the CD player to the DAC. So the opto-coupler is only used for isolation over a very short distance. The signal looked much cleaner and The crosstalk to ground was significantly reduced due to the much lower capacitance. Faster optocouplers (40Mb datarate) are available like the HCPL2400 from Agilent, that should be fast enough.
However the opto coupler, if used, should not affect the sound quality in a negative way.
Re: Optocoupler speed
ECD,
You are, as usual, steaming along at a rapid pace!
Do you ever take time out to relax?
OK...
How are you measuring Xtalk to ground?
Also, another alternative is very low coupling C transformer.
There are pulse trannies available such as scientific conversion that
have only a few pF coupling C between P and S.
WRT power supplies, maybe eventually ditch the 3 terminal regs at
the dacs and do simple discrete EF or shunt type. The 3 terms are
good for pre regulation.
Keep up good work
Cheers
Terry
-ecdesigns- said:
ECD,
You are, as usual, steaming along at a rapid pace!
Do you ever take time out to relax?
OK...
How are you measuring Xtalk to ground?
Also, another alternative is very low coupling C transformer.
There are pulse trannies available such as scientific conversion that
have only a few pF coupling C between P and S.
WRT power supplies, maybe eventually ditch the 3 terminal regs at
the dacs and do simple discrete EF or shunt type. The 3 terms are
good for pre regulation.
Keep up good work
Cheers
Terry
Hi ecdesigns,
Would it be easy to implement an AES/EBU interface for digital input to the Octal DAC? Seems like this interface is measurably and audibly better than SPDIF. Most pro digital equalizers/cross overs use AES/EBU, which may be handy for some who may use digital EQ to tame their rooms.
Also, by using battery power, wouldn't you have less noise problems that you are currently trying to eradicate with AC power supply?
Thanks.
Would it be easy to implement an AES/EBU interface for digital input to the Octal DAC? Seems like this interface is measurably and audibly better than SPDIF. Most pro digital equalizers/cross overs use AES/EBU, which may be handy for some who may use digital EQ to tame their rooms.
Also, by using battery power, wouldn't you have less noise problems that you are currently trying to eradicate with AC power supply?
Thanks.
Interference source located
Hi Terry Demol,
Thanks for your reply [post#446]
> Seems the only time out is during listening sessions, and the few hours of "sleep" listening to the frogs, yes I have a pond (photograph [post#8]) and the green frogs found it.
> I have been busy trying to reduce interference. I measure interference to ground by shortcircuiting the oscilloscope probe and holding it at GND, it seems strange but it works. Another technique is to stop the CD player and just measure the audio output, while the SPDIF stream is still running without data. Then when disconnecting the SVHS plug, the difference in noise levels can be clearly seen.
I just had a quick listening session with the optocouplers, well jocko was right and the opto's are out. I am quite sure jitter was introduced. Then I compared the transformer I made myself (ringcore 1:1) with the one I got from Farnell. The self made transformer had 1.4pF! so now I finally located the problem. I have wasted a few days locating interference that was created by the pulse transformer I bought.
Then I tried several pulse transformers, measured the capacitance between primary and secondary windings and observed the generated waveform at the output. Best sound results so far were accomplished with Valor transformers....from a computer network card, but capacitance between windings is around 4pF. The ringcore transformer I made had some distortion on the transients. SPDIF sure seems critical, even with a differential interface, can't wait to test I2S.
> At the moment I am using both pre and post regulators. I used 3 terminal regs because I needed a lot of them, as each module has it's own stabilized power supply. By placing the regulators very close to the circuit, interference is low and the main power supply becomes less complex. Heat is spread out over a larger area and heatsinks are no longer necessary on the modules, as each regulator only needs to supply little current. The tube power supply uses a 240V pre regulator and 2 cascaded zener shunt regulators, followed by the 8 separate stabilizer transistors. Tubes seem to be very sensitive to both hum on the power supply and the filament supply.
Cheers,
John
Hi Terry Demol,
Thanks for your reply [post#446]
> Seems the only time out is during listening sessions, and the few hours of "sleep" listening to the frogs, yes I have a pond (photograph [post#8]) and the green frogs found it.
> I have been busy trying to reduce interference. I measure interference to ground by shortcircuiting the oscilloscope probe and holding it at GND, it seems strange but it works. Another technique is to stop the CD player and just measure the audio output, while the SPDIF stream is still running without data. Then when disconnecting the SVHS plug, the difference in noise levels can be clearly seen.
I just had a quick listening session with the optocouplers, well jocko was right and the opto's are out. I am quite sure jitter was introduced. Then I compared the transformer I made myself (ringcore 1:1) with the one I got from Farnell. The self made transformer had 1.4pF! so now I finally located the problem. I have wasted a few days locating interference that was created by the pulse transformer I bought.
Then I tried several pulse transformers, measured the capacitance between primary and secondary windings and observed the generated waveform at the output. Best sound results so far were accomplished with Valor transformers....from a computer network card, but capacitance between windings is around 4pF. The ringcore transformer I made had some distortion on the transients. SPDIF sure seems critical, even with a differential interface, can't wait to test I2S.
> At the moment I am using both pre and post regulators. I used 3 terminal regs because I needed a lot of them, as each module has it's own stabilized power supply. By placing the regulators very close to the circuit, interference is low and the main power supply becomes less complex. Heat is spread out over a larger area and heatsinks are no longer necessary on the modules, as each regulator only needs to supply little current. The tube power supply uses a 240V pre regulator and 2 cascaded zener shunt regulators, followed by the 8 separate stabilizer transistors. Tubes seem to be very sensitive to both hum on the power supply and the filament supply.
Cheers,
John
AES/EBU interface / room acoustics
Hi, MGH
Thanks for your reply [post#447]
> AES/EBU interface is a differential interface (110 Ohm) and uses 3 way connectors. The differential RS422 interface I am using is basically the same, but I used a modified SVHS cable with 4-way SVHS connectors.
> The main noise problem is solved now, the residual noise is generated by the digital circuitry. Power supplies seem to be OK, so first I have a look what improvement can be made with screening (aluminum housing) and addittional screening plates.
> I don't use any equalizing, when using the sonic resonators, room reflections don't affect sound like using standard speakers. I have 2 medium sized prototype sonic resonators in the cellar where I do all assembling and testing, with brick walls, tiled floor and a concrete ceiling. The sonic resonators still sound beautiful, but when volume is turned up, all kind of stuff laying around starts resonating at specific frequencies. I use a test CD with a sinewave sweep to locate objects in the living room that start resonating at a specific frequency, distorting sound.
Hi, MGH
Thanks for your reply [post#447]
> AES/EBU interface is a differential interface (110 Ohm) and uses 3 way connectors. The differential RS422 interface I am using is basically the same, but I used a modified SVHS cable with 4-way SVHS connectors.
> The main noise problem is solved now, the residual noise is generated by the digital circuitry. Power supplies seem to be OK, so first I have a look what improvement can be made with screening (aluminum housing) and addittional screening plates.
> I don't use any equalizing, when using the sonic resonators, room reflections don't affect sound like using standard speakers. I have 2 medium sized prototype sonic resonators in the cellar where I do all assembling and testing, with brick walls, tiled floor and a concrete ceiling. The sonic resonators still sound beautiful, but when volume is turned up, all kind of stuff laying around starts resonating at specific frequencies. I use a test CD with a sinewave sweep to locate objects in the living room that start resonating at a specific frequency, distorting sound.
Hi ecdesigns,
Thanks for the input. But I was wondering how easy it would be to implement an AES/EBU interface in the Octal DAC because I have a pro digital equalizer that uses this interface in between the CD transport and DAC. I suppose I can buy adapters, but I figure it's better to have AES/EBU input. I have room modes I'm trying to get rid of without spending a $1000 on room treatments.
Thanks for the input. But I was wondering how easy it would be to implement an AES/EBU interface in the Octal DAC because I have a pro digital equalizer that uses this interface in between the CD transport and DAC. I suppose I can buy adapters, but I figure it's better to have AES/EBU input. I have room modes I'm trying to get rid of without spending a $1000 on room treatments.
Dear Ecdesigns:
I bet you will be astonished by the result!
My amateur experience with a homemade, humble 4*TDA1543 filterless DAC, fed through I2S from CDPRO, is excellent (if you do not count a little HF noise 
)
The 10 dollar DAC has no rigth to sound that good!
You may well consider afterwards that all that effort tweaking SPDIF is not worth it. If so, let us know
Regards
M
I bet you will be astonished by the result!
My amateur experience with a homemade, humble 4*TDA1543 filterless DAC, fed through I2S from CDPRO, is excellent
(if you do not count a little HF noise )The 10 dollar DAC has no rigth to sound that good!
You may well consider afterwards that all that effort tweaking SPDIF is not worth it. If so, let us know
Regards
M
Hello ECDesigns,
Lets fall with the door in the house. (Dutch saying, and I like it very much).
I am very curious about the sound of the DAC. So, is it possible to listen to your DAC? I do live in the Netherlands, in the North.
If it is okay I would like to build it. All the DAC's I have do not sound as good as a turntable. (I once have heard a Novelle Platine Verdier via Audio Research and Danisch speakers, made of "soaprock"). All the DAC's I heared did sound thin compared to the turntable. (Of course perception of sound is personal).
You will say of course, why not buy a Novelle Platine? To expensive and I don't like scratches and noise.
Greetings,
Dutchman
Lets fall with the door in the house. (Dutch saying, and I like it very much).
I am very curious about the sound of the DAC. So, is it possible to listen to your DAC? I do live in the Netherlands, in the North.
If it is okay I would like to build it. All the DAC's I have do not sound as good as a turntable. (I once have heard a Novelle Platine Verdier via Audio Research and Danisch speakers, made of "soaprock"). All the DAC's I heared did sound thin compared to the turntable. (Of course perception of sound is personal).
You will say of course, why not buy a Novelle Platine? To expensive and I don't like scratches and noise.
Greetings,
Dutchman
Optimizing audio interface
Hi maxlorenz,
Thanks for your reply [post#451]
I am going to use a mac with the USB to I2S module (still waiting for the PCM2706 to arrive).
Optimizing the CS8412 audio receiver circuit:
I made a new low capacitance 1.4pF 1:1 pulse transformer using a small 10mm diameter ringcore. Using self-made enamelled stranded wire (12 very thin insulated strands twisted together). Then I lowered the impedance of the loop filter (270 Ohm with 680nF), 270 Ohm is a metal film 1206 size SMD resistor. Finally I put a screen from a old MF transformer around the 680nF BC foil capacitor and connected it to ground. I also put a screen plate on the solder side covering the area above the CS8412 loop filter PIN's. Now when observing BCK transients on maximun oscilloscope setting and X10 enhanced timebase, they look very clean, no haze at all. So the circuitry around the loop filter is very important to obtain low jitter, as any picked-up (HF) interference will increase PLL jitter.
The sound quality confirms jitter must be lower then before now, I am quite happy with the result. I will post a photograph of the modifications made.
Hi maxlorenz,
Thanks for your reply [post#451]
I am going to use a mac with the USB to I2S module (still waiting for the PCM2706 to arrive).
Optimizing the CS8412 audio receiver circuit:
I made a new low capacitance 1.4pF 1:1 pulse transformer using a small 10mm diameter ringcore. Using self-made enamelled stranded wire (12 very thin insulated strands twisted together). Then I lowered the impedance of the loop filter (270 Ohm with 680nF), 270 Ohm is a metal film 1206 size SMD resistor. Finally I put a screen from a old MF transformer around the 680nF BC foil capacitor and connected it to ground. I also put a screen plate on the solder side covering the area above the CS8412 loop filter PIN's. Now when observing BCK transients on maximun oscilloscope setting and X10 enhanced timebase, they look very clean, no haze at all. So the circuitry around the loop filter is very important to obtain low jitter, as any picked-up (HF) interference will increase PLL jitter.
The sound quality confirms jitter must be lower then before now, I am quite happy with the result. I will post a photograph of the modifications made.
LCL filters
Hi tubee,
Congratulations with the successful repair of the 304!
the LCL filters (post#452) basically consist of 2 ferrite beads and a capacitor. The ferrite beads are in series, centre tap is connected to the capacitor. The ferrite beads can be ordered separately. The coil is most important to isolate ground noise.
Hi tubee,
Congratulations with the successful repair of the 304!
the LCL filters (post#452) basically consist of 2 ferrite beads and a capacitor. The ferrite beads are in series, centre tap is connected to the capacitor. The ferrite beads can be ordered separately. The coil is most important to isolate ground noise.
Listening session
Hi Dutchman,
Thanks for your reply [post#453]
Yes of course you can come over to listen to the DAC, I will send you an invitation by email.
When you decide to come over and listen to the octal D-I DAC and sonic resonators, you can bring your favorite CD's for comparison.
If you like the octal D-I DAC sound, you can build it of course. The octal D-I DAC core (analog mainboard), tube output stage and SPDIF audio interface could already be used.
Hi Dutchman,
Thanks for your reply [post#453]
Yes of course you can come over to listen to the DAC, I will send you an invitation by email.
When you decide to come over and listen to the octal D-I DAC and sonic resonators, you can bring your favorite CD's for comparison.
If you like the octal D-I DAC sound, you can build it of course. The octal D-I DAC core (analog mainboard), tube output stage and SPDIF audio interface could already be used.
Thanks EC for LCL explaining, i thought allready it should be made in about that way. I am thinking to add a LCL in Dem circuit.
The 304 had a simple fault, cause removed a FF, this redirects subcode dat back to uP, besides the 2nd half regulates the spdif.
I have a question EC:
Your CD-engine, the Sony, have you modified it?(digital clock)
This i ask because i have been listening, between day tasks, to the 304mk2 and discovered a huge improvement a clock (and ofcourse the other upgrades, the opa2132 and more) can bring in overall performance of this player. Background chores can be followed word for word, (i could allready before) but now i hear every small vibrato in their voices. Bass is very nice now, every bass tune can be easily followed and is very natural, mids is excellent now, very relaxing sound, soundstage & depth is improved. The dac in 304 is one of 1st plain TDA1541's, no A or S1 or whatever. There still is some sibilance in highs, i think because of my coupling cap choice: 10uF MKT bypassed with 15N styro, and of the high sampling rate ofcourse from dig filter. Better choice for coupling cap would be a BlackGate. Will try this soon.
Non-os has other qualities though, but this player sound very convincing now.
I am sure, when adding a Kwak 7 clock-gererator into the Sony, this can improve the overall sound. Even when it's only delivering data via Spdif. And don't measure with a scope this time(not offending ) but sit back and listen to it with the D-I dac, and decide if it has to leave or stay.
So, if you want to try a KC7, i can etche a pcb for you i have designed here with simple french TCI software. A friend of my who has a CD880 wants a pcb too, so 1 more is no big hassle for me. If you don't like the clock upgrade, remove it again. If you like it but want to improve pcb layout yourself you can do it later if wanted.
Let me hear what you think of this proposal.
This is first prototype, it found its place in the 304.
[img=http://img348.imageshack.us/img348/6459/kwak72sw.th.jpg]
The 304 had a simple fault, cause removed a FF, this redirects subcode dat back to uP, besides the 2nd half regulates the spdif.
I have a question EC:
Your CD-engine, the Sony, have you modified it?(digital clock)
This i ask because i have been listening, between day tasks, to the 304mk2 and discovered a huge improvement a clock (and ofcourse the other upgrades, the opa2132 and more) can bring in overall performance of this player. Background chores can be followed word for word, (i could allready before) but now i hear every small vibrato in their voices. Bass is very nice now, every bass tune can be easily followed and is very natural, mids is excellent now, very relaxing sound, soundstage & depth is improved. The dac in 304 is one of 1st plain TDA1541's, no A or S1 or whatever. There still is some sibilance in highs, i think because of my coupling cap choice: 10uF MKT bypassed with 15N styro, and of the high sampling rate ofcourse from dig filter. Better choice for coupling cap would be a BlackGate. Will try this soon.
Non-os has other qualities though, but this player sound very convincing now.
I am sure, when adding a Kwak 7 clock-gererator into the Sony, this can improve the overall sound. Even when it's only delivering data via Spdif. And don't measure with a scope this time(not offending
) but sit back and listen to it with the D-I dac, and decide if it has to leave or stay.So, if you want to try a KC7, i can etche a pcb for you i have designed here with simple french TCI software. A friend of my who has a CD880 wants a pcb too, so 1 more is no big hassle for me. If you don't like the clock upgrade, remove it again. If you like it but want to improve pcb layout yourself you can do it later if wanted.
Let me hear what you think of this proposal.
This is first prototype, it found its place in the 304.
[img=http://img348.imageshack.us/img348/6459/kwak72sw.th.jpg]
I2S
Hi tubee,
thanks for your important tip [post#459]
>It resulted in a complete disassembly of my Sony 300CD player, just to have a look at the laser pic-up module that also contains servo control and D/A conversion circuitry, and the crystal oscillator of course. Turned out the crystal was floating 5mm above the PCB, wasn't connected to ground, and a very small noisy SMD oscillator resistor was used.
So I put the crystal flat to the PCB, connected the housing to GND and replaced the small SMD resistor with a bigger one (thicker conductive layer / less noise). Then I screened both digital chip and oscillator circuit with adhesive copper tape (with a insulating layer in between), and connected this screen to GND as well.
I also had a quick look at the digital chip used : CXD2587Q, looked up the datasheet, and guess what. The CXD2587Q has a I2S bus! ,WS on pin 61, DATA on pin 62, BCK on pin 63. So the sony player awaits another disassembly and some serious modding very soon. If the CD player survives this, I can finally listen to a direct I2S interface. I just have to figure-out how to construct the wiring between player and DAC. I think I will use 3 differential RS422 interfaces with a computer networking cable for the moment.
> You mentioned a huge improvement due to several modifications of your 304, have you also noticed that sound quality gradually improves the longer the CD player is switched on? I ask this, because I noticed this with the OPA2604, OPA2132 and OPA627 OP-amps. They seem to need time to settle / warm-up before optimal performance is obtained, they take about 10-15 minutes for this.
> Coupling cap, that's why I used a differential DAC setup. But you can try a 10uF/100V polypropylene cap from Epcos. Adding cap's in parallel with one that has a relatively high dissipation factor doesn't improve things.
> Kwak7 clock, the problem is that the laser pick-up module moves when the next of the 300 CD's is loaded, and there is very little room for a circuit board.
Hi tubee,
thanks for your important tip [post#459]
>It resulted in a complete disassembly of my Sony 300CD player
, just to have a look at the laser pic-up module that also contains servo control and D/A conversion circuitry, and the crystal oscillator of course. Turned out the crystal was floating 5mm above the PCB, wasn't connected to ground, and a very small noisy SMD oscillator resistor was used. So I put the crystal flat to the PCB, connected the housing to GND and replaced the small SMD resistor with a bigger one (thicker conductive layer / less noise). Then I screened both digital chip and oscillator circuit with adhesive copper tape (with a insulating layer in between), and connected this screen to GND as well.
I also had a quick look at the digital chip used : CXD2587Q, looked up the datasheet, and guess what. The CXD2587Q has a I2S bus!
,WS on pin 61, DATA on pin 62, BCK on pin 63. So the sony player awaits another disassembly and some serious modding very soon. If the CD player survives this, I can finally listen to a direct I2S interface. I just have to figure-out how to construct the wiring between player and DAC. I think I will use 3 differential RS422 interfaces with a computer networking cable for the moment.> You mentioned a huge improvement due to several modifications of your 304, have you also noticed that sound quality gradually improves the longer the CD player is switched on? I ask this, because I noticed this with the OPA2604, OPA2132 and OPA627 OP-amps. They seem to need time to settle / warm-up before optimal performance is obtained, they take about 10-15 minutes for this.
> Coupling cap, that's why I used a differential DAC setup. But you can try a 10uF/100V polypropylene cap from Epcos. Adding cap's in parallel with one that has a relatively high dissipation factor doesn't improve things.
> Kwak7 clock, the problem is that the laser pick-up module moves when the next of the 300 CD's is loaded, and there is very little room for a circuit board.
Photograps of modified SPDIF module
Hi all,
Operation I2S failed,
Seems there is a firmware error in the CX2587Q (things like that always happen to me). WS = 44.1KHz (ok), BCK = 2.11677MHz ? DATA is in sync with BCK, the frequency is way too low. Result: very weak extremely distorted sound. Data for both CX2587Q and TDA1541A is 2's complement MSB first. Good news is, the CD player still works.
I also added a photograph of the modified SPDIF interface. Description [post#455]
Upper photograph: screening of the loop filter film capacitor (metal square close to the red jumpers), on the right, the new 1:1 ringcore pulse transformer.
Lower photograph, solder side with the loop filter screening on the left. Below the circuit board is a piece of stranded wire I used for the transformer. This stranded wire has good high frequency properties.
Hi all,
Operation I2S failed,
Seems there is a firmware error in the CX2587Q (things like that always happen to me). WS = 44.1KHz (ok), BCK = 2.11677MHz ? DATA is in sync with BCK, the frequency is way too low. Result: very weak extremely distorted sound. Data for both CX2587Q and TDA1541A is 2's complement MSB first. Good news is, the CD player still works.
I also added a photograph of the modified SPDIF interface. Description [post#455]
Upper photograph: screening of the loop filter film capacitor (metal square close to the red jumpers), on the right, the new 1:1 ringcore pulse transformer.
Lower photograph, solder side with the loop filter screening on the left. Below the circuit board is a piece of stranded wire I used for the transformer. This stranded wire has good high frequency properties.
