Hi,
i was unsatisfied with the muffled, airless and closed sound of my original sony xa50es too. i still haven't touched the clocks so far, but i got some nice results anyway.
i firstly changed the electrolytic caps in the digital area of the da-board. i used rubycon MBZ 16V 1800uF for C625/626/613 to suppress noise from the supply chain. i used oscon sp 180uF 16V locally. i replaced ic605 against a MC7805ACT, which is a little less noisy.
i added a 22 Ohm resistor in series to the 3.3uH choke for the 45 mhz clock. i removed the 0.1uf ceramic and placed a 220uF 6.3V black gate nx at this place. at the bottom side of the pcb i placed 3 x 10nf npo smd ceramic chip capacitors very close to the smd inverter clock ic. the 16mhz clock was powered from the digital voltage chain, i changed that to the analog chain, in the same way as the 45 MHz clock.
but that kind of tuning was not deliver much better quality for the analog cinch-out.
now i removed the c 531,532-535, 431,432-435. they are 3% film caps, but i doubt that they are high-quality audio parts, cause they are very small. btw: strange construction, cause filtering should normally take place after i/v-conversion?!
i removed ic 501,401 and soldered ic-sockets at this place. cause the ics are dealing with some hf and heavely capacitive loads, i used 2 x ad 826.
for testing purpose i did a small hearing session, but the sound quality was not good, quite noisy and disquietingly.
now i placed wima fkp 2.5% 100V PP film capacitors:
C531, 431 10nF
C532,533,432,433 4.7nF
C534,535,434,435 2.2nF
(round about 50% of original values)
further i added two MKP 22nF PP film caps parallel at C621,622.
now i got the sound-quality, i expected from a player in that class. more air, more agile and powerful!
(the ad826 ics deliver a nice improvement, but the mkp caps are more important)
compared to my tuned parasound d/ac-1600 the sony sounded very nice, someway darker, but it still lacks some dynamics. at least i can say now, that the d/a-board of the xa50es is NOT crap or something like that ! (greetings at elso ;-)
some things left to do:
- adding a buffer for the cinch-out (i plan to change the opa2132 against ad8620+buf634)
- higher grade nichicon kz for C926,936 on the power board
- new clock
- the head phone amp is a impertinence
greetings
i was unsatisfied with the muffled, airless and closed sound of my original sony xa50es too. i still haven't touched the clocks so far, but i got some nice results anyway.
i firstly changed the electrolytic caps in the digital area of the da-board. i used rubycon MBZ 16V 1800uF for C625/626/613 to suppress noise from the supply chain. i used oscon sp 180uF 16V locally. i replaced ic605 against a MC7805ACT, which is a little less noisy.
i added a 22 Ohm resistor in series to the 3.3uH choke for the 45 mhz clock. i removed the 0.1uf ceramic and placed a 220uF 6.3V black gate nx at this place. at the bottom side of the pcb i placed 3 x 10nf npo smd ceramic chip capacitors very close to the smd inverter clock ic. the 16mhz clock was powered from the digital voltage chain, i changed that to the analog chain, in the same way as the 45 MHz clock.
but that kind of tuning was not deliver much better quality for the analog cinch-out.
now i removed the c 531,532-535, 431,432-435. they are 3% film caps, but i doubt that they are high-quality audio parts, cause they are very small. btw: strange construction, cause filtering should normally take place after i/v-conversion?!
i removed ic 501,401 and soldered ic-sockets at this place. cause the ics are dealing with some hf and heavely capacitive loads, i used 2 x ad 826.
for testing purpose i did a small hearing session, but the sound quality was not good, quite noisy and disquietingly.
now i placed wima fkp 2.5% 100V PP film capacitors:
C531, 431 10nF
C532,533,432,433 4.7nF
C534,535,434,435 2.2nF
(round about 50% of original values)
further i added two MKP 22nF PP film caps parallel at C621,622.
now i got the sound-quality, i expected from a player in that class. more air, more agile and powerful!
(the ad826 ics deliver a nice improvement, but the mkp caps are more important)
compared to my tuned parasound d/ac-1600 the sony sounded very nice, someway darker, but it still lacks some dynamics. at least i can say now, that the d/a-board of the xa50es is NOT crap or something like that ! (greetings at elso ;-)
some things left to do:
- adding a buffer for the cinch-out (i plan to change the opa2132 against ad8620+buf634)
- higher grade nichicon kz for C926,936 on the power board
- new clock
- the head phone amp is a impertinence
greetings
Attachments
the filter: finally ;-)
Hi folks,
i suppose that the only purpose for the filter caps after the current pulse ic's is to limit the V/t speed for the i-v-converter op-amps to prevent distortion by oversteering. the original opa2604 has only 25V/us , the new ad826 has 350V/us.
now i placed the following wima fkp 2.5% 100V PP film capacitors:
C531, 431 2.2nF
C532,533,432,433 1.0nF
C534,535,434,435 0.47nF
(round about 10% of original values)
the lack of dynamics and air disappeared completely! but the sound was still somewhat flimsy, not as compact and good defined as from my parasound.
i added further 10nF PP film caps (WIMA FKP2) directly from the AD862 supply pins to ground.
it is hard to detect the small differences in the sound from my tuned parasound and the sony player now.
my impression is, the parasound has a little better definition and sound stage, but the sony has nice colors. i still prefer my parasound dac even in asymmetrical mode, but i would say the play in the same class now!
the originally dark but tired trait of the sony is completely gone! it seems to me, that was only a problem of limiting speed for the cheap audio-opamps as i/v-converter.
greetings.
Hi folks,
i suppose that the only purpose for the filter caps after the current pulse ic's is to limit the V/t speed for the i-v-converter op-amps to prevent distortion by oversteering. the original opa2604 has only 25V/us , the new ad826 has 350V/us.
now i placed the following wima fkp 2.5% 100V PP film capacitors:
C531, 431 2.2nF
C532,533,432,433 1.0nF
C534,535,434,435 0.47nF
(round about 10% of original values)
the lack of dynamics and air disappeared completely! but the sound was still somewhat flimsy, not as compact and good defined as from my parasound.
i added further 10nF PP film caps (WIMA FKP2) directly from the AD862 supply pins to ground.
it is hard to detect the small differences in the sound from my tuned parasound and the sony player now.
my impression is, the parasound has a little better definition and sound stage, but the sony has nice colors. i still prefer my parasound dac even in asymmetrical mode, but i would say the play in the same class now!
the originally dark but tired trait of the sony is completely gone! it seems to me, that was only a problem of limiting speed for the cheap audio-opamps as i/v-converter.
greetings.
Lately I started using my CDPXA50ES as a transport, because my Denon 3560 needed service. Originally I bought the 50ES back in 98, but stopped using it around 2000.
Put into my system the 50ES was easily recognizable, that was actually the reason it went out earlier. I simply could not accept its way to emphasize transients, and it also has a certain way of boomyness in the the low end, and that seems to cause that there not really is any low end at all.
Well that was not good, but I came to think of the experience I´d gathered from an ongoing DAC project. Because this is actually some of the symptoms on jitter.
Used as a transport only, you are not really in a position of fighting jitter, but it is possible to help a little. To fight jitter at its very origin, you can try to create a datastream as readable as possible at the reciever end. This cals for good powersupplies, mechanical stability amongst a lot of other things that cannot be modified upon.
But I went for a try, first I paid my attention to the PSU´s.
The digital PSU is a 4 voltage regulated type, it supplies +5V, +-7V and -15V. The +-7V supplies drives the spindle and as I remember also the BSL, the -15V supplies amongs others the display. So the 5V supply is the important one, it supplies the digital processors, the laser and some funktions that goes of when the disc is turning, ie. the drawer.
The 5V supply is build so that the positive side of the digital +-7V supply has an extra regulator besides the +7V. So it is relatively easy to se and follow the whole idea behind the construction of this supply.
Looking at the components in the 50ES, Sony chose to use nice components in the audio part, the digital is made of GP components from the cheap end. The supply is regulated (7805) and smoothed with caps from Nippon Chemicom and a few Elnas as well, then decoupled at the processors with ceramic caps. This is the common way to that, but the Nippons are not good caps, they have high ESR - at least 1 Ohm. They can with good results be replaced by Panasonic FM paralleled with a Wima film cap, best of all from their pulse caps.
The relevant caps are C954, C951,C953 on the powersupply board. You can leave the C951 if you can´t find anything really low ESR to replace with, because this cap is parallel to C954.
On the servoboard -the one above the drive - you have to replace C94 and C101 and C102,
Finally replace C707 on the coaxboard - the small one just inside the RCA socket.
None of the values are critical, but stay as close to the original as possible, and don´t use lower values on the supply board.
The C951 is a 6800µF 16V type, Panasonic does not have this value in the very low ESR FM series, so you´l have to think about something else here ie FC series or maybe Sanyo could be relevant.
The idea behind this mod is, that the processors in the servosection and on the coaxboard are very fast, and they need powersupply in very small pulses at a speedy rate, this calls for very low impedance supply rails. Since 98 when this machine was build, a lot has happened to capacitors, especially to ESR values, they are lowered a lot during this period. So it will suit most transports and other digital gear to pay sufficient attention to this.
It will help create a clearer electrical image of the stream, which can be easier read and understod correctly by the reciever in the external DAC. The sound improves a lot by this, both smoother and cleare at the same time, resolution in bass and midrange improves, and imaging becomes a lot more realistic and stable. If it also the the player as standalone, I don´t know, but I don´t think the new caps costs more than 10$, so it is a cheap trick to try it out.
As an aditional, you can also make some of the players covers a bit more stable. I´ve earlier tried to use bitumen to get rid of wibrations, but I don´t really feel that this a good idea. I´d rather prevent vibration than to dampen them. So you could place a few peaces of whatever material you prefer, on top of the drive, this will make contact with the top cover, which then is prevented from vibrating. The buttom cover can benefit from the same, here the 2 separtion walls can be used to stiffen the buttom plate. This rweally leaves the player as a stif mechanical part, which behaves as if it was made out of just one piece.
Some prefer also to mount an IEC socket for the powercord, I´ve tried diffeent powercords on my player, and they all sound different, but Iit is just arround impossible to decide which is the better. This stunt I feel is more about systemintegration than actual sound quality, as such I would leave the original in place.
If your hands are carefull and smooth, I´d dismount the cover on tha laser, and clean the laserlens on both sides. If carefull this will prolong the life of the laser, as well as improve the sound.
Enjoy!
Put into my system the 50ES was easily recognizable, that was actually the reason it went out earlier. I simply could not accept its way to emphasize transients, and it also has a certain way of boomyness in the the low end, and that seems to cause that there not really is any low end at all.
Well that was not good, but I came to think of the experience I´d gathered from an ongoing DAC project. Because this is actually some of the symptoms on jitter.
Used as a transport only, you are not really in a position of fighting jitter, but it is possible to help a little. To fight jitter at its very origin, you can try to create a datastream as readable as possible at the reciever end. This cals for good powersupplies, mechanical stability amongst a lot of other things that cannot be modified upon.
But I went for a try, first I paid my attention to the PSU´s.
The digital PSU is a 4 voltage regulated type, it supplies +5V, +-7V and -15V. The +-7V supplies drives the spindle and as I remember also the BSL, the -15V supplies amongs others the display. So the 5V supply is the important one, it supplies the digital processors, the laser and some funktions that goes of when the disc is turning, ie. the drawer.
The 5V supply is build so that the positive side of the digital +-7V supply has an extra regulator besides the +7V. So it is relatively easy to se and follow the whole idea behind the construction of this supply.
Looking at the components in the 50ES, Sony chose to use nice components in the audio part, the digital is made of GP components from the cheap end. The supply is regulated (7805) and smoothed with caps from Nippon Chemicom and a few Elnas as well, then decoupled at the processors with ceramic caps. This is the common way to that, but the Nippons are not good caps, they have high ESR - at least 1 Ohm. They can with good results be replaced by Panasonic FM paralleled with a Wima film cap, best of all from their pulse caps.
The relevant caps are C954, C951,C953 on the powersupply board. You can leave the C951 if you can´t find anything really low ESR to replace with, because this cap is parallel to C954.
On the servoboard -the one above the drive - you have to replace C94 and C101 and C102,
Finally replace C707 on the coaxboard - the small one just inside the RCA socket.
None of the values are critical, but stay as close to the original as possible, and don´t use lower values on the supply board.
The C951 is a 6800µF 16V type, Panasonic does not have this value in the very low ESR FM series, so you´l have to think about something else here ie FC series or maybe Sanyo could be relevant.
The idea behind this mod is, that the processors in the servosection and on the coaxboard are very fast, and they need powersupply in very small pulses at a speedy rate, this calls for very low impedance supply rails. Since 98 when this machine was build, a lot has happened to capacitors, especially to ESR values, they are lowered a lot during this period. So it will suit most transports and other digital gear to pay sufficient attention to this.
It will help create a clearer electrical image of the stream, which can be easier read and understod correctly by the reciever in the external DAC. The sound improves a lot by this, both smoother and cleare at the same time, resolution in bass and midrange improves, and imaging becomes a lot more realistic and stable. If it also the the player as standalone, I don´t know, but I don´t think the new caps costs more than 10$, so it is a cheap trick to try it out.
As an aditional, you can also make some of the players covers a bit more stable. I´ve earlier tried to use bitumen to get rid of wibrations, but I don´t really feel that this a good idea. I´d rather prevent vibration than to dampen them. So you could place a few peaces of whatever material you prefer, on top of the drive, this will make contact with the top cover, which then is prevented from vibrating. The buttom cover can benefit from the same, here the 2 separtion walls can be used to stiffen the buttom plate. This rweally leaves the player as a stif mechanical part, which behaves as if it was made out of just one piece.
Some prefer also to mount an IEC socket for the powercord, I´ve tried diffeent powercords on my player, and they all sound different, but Iit is just arround impossible to decide which is the better. This stunt I feel is more about systemintegration than actual sound quality, as such I would leave the original in place.
If your hands are carefull and smooth, I´d dismount the cover on tha laser, and clean the laserlens on both sides. If carefull this will prolong the life of the laser, as well as improve the sound.
Enjoy!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.