I have OPPO DV970 which I have bought from ebay because of very positive review.
The first impression was very good. I have liked the fat and musical sound, the bottom and middle frequency seems to be fine but the top end was not so good. After one-two hour of listening the sound of the Oppo is tiring me.
I have spend many hours with oscilloscope to measure and many hours to listen my oppo to improve the sound. Because I have only one Oppo DV970 player I have use the other player for reference which is with the some DAC CS4360 inside but with tube out section. However here is what I have done mostly on the power supply section.
Here is schematic and board of the SMPS.
The first impression was very good. I have liked the fat and musical sound, the bottom and middle frequency seems to be fine but the top end was not so good. After one-two hour of listening the sound of the Oppo is tiring me.
I have spend many hours with oscilloscope to measure and many hours to listen my oppo to improve the sound. Because I have only one Oppo DV970 player I have use the other player for reference which is with the some DAC CS4360 inside but with tube out section. However here is what I have done mostly on the power supply section.
Here is schematic and board of the SMPS.
Here is the power supply and current for the different line:
- 3.3v digital section for MediaTek take 0.75 A and are well regulated with very low noise.
- 5V digital and analog power supply of the DAC take 0.05A and is very bad regulated with a lot of noise. After checking what cause this noise I have noted that the motor drive servo control circuit is connected together with digital/analog DAC power supply.
- 5V power supply of the motor servo control take 0.5A
- (+9V and -9V) for opamp's take 0.01A and are very well regulated with low noise especially at -9V.
So my purpose was to separate the 5V DAC supply from BA5954 motor servo control.
The PCB is very well made with a lot of holes and jumpers so the work is easy.
I have try many way to made well controlled power supply line with low noise but the most proper manner is to force SMPS feedback circuit to control 5V line not 3.3V. This is done by removing jumper JP502 and place jumper at JP506 and change resistor R509 with 3.3K. Now the 3.3V line will be post regulated from 5V line via UZ1084-3.3 device. For this purpose TC505 and TC506 was increased to 1500uF (I have use Rubicon MBZ 1500uF/16V) and L507 was increased to 3A. For the diode D510 I have try MSR860, 1N5822, FR307, SF36, HER307, SR360 but the best working diode is SR580. For UZ1084 I have use 470uF/16V for input and output. D513, TC514, L506 and TC510 should be removed if you want to free the space. In every case D513 must be removed. Install UZ1084 in place of JP507 or JP514 (I cannot remember) because there is close +5V input, GND and 3.3V output. Now we have excellent 5V and 3.3V power supply.
- 3.3v digital section for MediaTek take 0.75 A and are well regulated with very low noise.
- 5V digital and analog power supply of the DAC take 0.05A and is very bad regulated with a lot of noise. After checking what cause this noise I have noted that the motor drive servo control circuit is connected together with digital/analog DAC power supply.
- 5V power supply of the motor servo control take 0.5A
- (+9V and -9V) for opamp's take 0.01A and are very well regulated with low noise especially at -9V.
So my purpose was to separate the 5V DAC supply from BA5954 motor servo control.
The PCB is very well made with a lot of holes and jumpers so the work is easy.
I have try many way to made well controlled power supply line with low noise but the most proper manner is to force SMPS feedback circuit to control 5V line not 3.3V. This is done by removing jumper JP502 and place jumper at JP506 and change resistor R509 with 3.3K. Now the 3.3V line will be post regulated from 5V line via UZ1084-3.3 device. For this purpose TC505 and TC506 was increased to 1500uF (I have use Rubicon MBZ 1500uF/16V) and L507 was increased to 3A. For the diode D510 I have try MSR860, 1N5822, FR307, SF36, HER307, SR360 but the best working diode is SR580. For UZ1084 I have use 470uF/16V for input and output. D513, TC514, L506 and TC510 should be removed if you want to free the space. In every case D513 must be removed. Install UZ1084 in place of JP507 or JP514 (I cannot remember) because there is close +5V input, GND and 3.3V output. Now we have excellent 5V and 3.3V power supply.
For the 5v analog/digital DAC power supply remove jumper JP501 which connect the two 5v line and place D507, TC515, L509 and TC508 to build separate 5V line. For D507 I have use SR580 again but the holes should be enlarged. Measure the new 5V line and If everything is fine you should read 5.45V on this line. If the line is below 5.45V change the R509 with 3.4k, 3.6k so that you have 5.45V. Now the 5V line will be post regulated from 5.45V line via LD29150 device. I have use LD29150 in DPAK package.
For the input pin put 470nF and for the output pin 33uF/10v. I have install it in place of JP520. Now 5v analog/digital DAC power supply is ok.
For the input pin put 470nF and for the output pin 33uF/10v. I have install it in place of JP520. Now 5v analog/digital DAC power supply is ok.
Last edited:
In the main pcb place capacitor TC231 and TC236 which are missing. I have use 47uf/25v Elna Silmic II. Remove all ceramic capacitor near opamp (C2138,........,C2143) and place two 47uF/25v Silmic II for +/-9V near TC220. Use the GND hole of TC220 for the Silmic's and the other legs of caps directly to the +9 and -9V trace (scratch the trace).
Here are some modifications I made on OPPO's.
Look, learn and implement.
Mod's can be extended to a higher level - it is up to your budget and what you want to obtain.
vicol audio : oppo 980h audio tuning
vicol audio : oppo 981hd audio tuning
vicol audio : oppo dv-983h audio tuning
Regards,
Tibi
Look, learn and implement.
Mod's can be extended to a higher level - it is up to your budget and what you want to obtain.vicol audio : oppo 980h audio tuning
vicol audio : oppo 981hd audio tuning
vicol audio : oppo dv-983h audio tuning
Regards,
Tibi
Decoupling caps used in OPPO are very cheap X7R. These exhibit very high piezoelectric effect. Beside this X7R have huge variations with temperature and potential applied. In the end, instead to get rid of HF noise, you are inducing a lot of noise in audio band due piezo-efect.
Up to 400Khz an Vishay or Arcotronics MKT/MKP will have same ESR as the ceramic one, without any X7R mentioned problems.
Regards,
Tibi
Nope, for decoupling they are the best, small package size for any given value, so you can get them next to the pins where they can actually DECOUPLE. Adding larger size caps defeats the object.
X7R are used on just about every board made these days including very sensitive analogue design, and other high end designs. They are not going to show the problems you describe to any extent that you would notice. And adding leaded devices is just a waste of time for decoupling, the reason everyone uses X&Rs for decoupling is that they offer the best solution in terms of case size and thus lower inductance...
Its audiophile fashion nothing else to remove them.
X7R are used on just about every board made these days including very sensitive analogue design, and other high end designs. They are not going to show the problems you describe to any extent that you would notice. And adding leaded devices is just a waste of time for decoupling, the reason everyone uses X&Rs for decoupling is that they offer the best solution in terms of case size and thus lower inductance...
Its audiophile fashion nothing else to remove them.
http://www.millertechinc.com/pdf_files/MTI TN095 Capacitor Noise.pdf
"For very low distortion applications (THD < 0.002%), use only NPO or film capacitors
(such as Polycarbonate or
Polypropylene)."
A quick search on internet will reveal a lot of documents related to extremely poor performance of X7R capacitors. X7Rmust be used with extremely care in audio applications.
Regards,
Tibi
"For very low distortion applications (THD < 0.002%), use only NPO or film capacitors
(such as Polycarbonate or
Polypropylene)."
A quick search on internet will reveal a lot of documents related to extremely poor performance of X7R capacitors. X7Rmust be used with extremely care in audio applications.
Regards,
Tibi
http://www.avx.com/tantalum/presentations/tac_ceramic.pdf
The last one. I hope I made my statement clear and no need further clarifications.
Regards,
Tibi
The last one. I hope I made my statement clear and no need further clarifications.
Regards,
Tibi
No they are not, do not quote what is untrue, neither are they for space applications or medical. Sorry but I do Class 3/3a design regularly and they all use X7Rs as decoupling. We are talking decoupling applications here not in the signal path.
So explain why they are used by the electronics industry for decoupling, on class 3 products.
Sorry but for decoupling X7rs are the best and for high speed digital decoupling they are the only caps where you can get the required value in the required (small) case size. We are not talking in the signal path, but decoupling.
I would guess the boards in this lot are covered with X7R decoupling capacitors:
http://www.northropgrumman.com/Capabilities/ANVIC5/Documents/vic5.pdf
All this stuff about the caps is not new, and has been known for years, as a coupling cap it may in some circumstances this effect may cause a problem, but we are not talking about them being used in this way, we are talking about power supply decoupling at a devices pins....
Read up on decoupling.... theres plenty of info out there.
I'm not generalising, I would be interested in any findings, as work I have done on the above, and other analogue designs (phase array radar for one) have used X7Rs for decoupling and there has been no measureable effect even when doing the vibration testing...
Anyway bye, no more arguing.
Marc
Anyway bye, no more arguing.
Marc
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.