What supply voltage(s) does the output stage run on? Maybe the peaks are clipping
from inadequate voltage swing in the output stage, when the digital volume is at max.
But at least you know how to reduce the problem. As I remember the digital control
should be reduced by 6dB from maximum to eliminate the clipping.
from inadequate voltage swing in the output stage, when the digital volume is at max.
But at least you know how to reduce the problem. As I remember the digital control
should be reduced by 6dB from maximum to eliminate the clipping.
Last edited:
Thanks for the help you two. With your help I've found that the output looks to be asymmetrical clipping. Following the slow trace speed suggestion, the top of the trace while playing heavy modulated music shows limited output of 1.3v peak while the bottom half of the trace goes off the scope screen! My test disk didn't drive the circuits hard enough to display this. The opamp circuit is +- 12 v, so not the limiting factor. But at least, thanks to you guys I have a new issue to pursue. Just got stuck, senior moment.....
Agree with rayma that the output stage power supply would be something to look at. If the voltage looks okay when not playing music, perhaps it sags under some playback conditions?
Do you have a schematic?
Is your scope on DC coupling?
Is there any DC at the audio outputs?
What is the DCV before the output coupling capacitor, if any?
Should be near 0VDC if the op amp works on +/-12V.
Does the supply measure right? Seems it might be the positive DC supply.
Is your scope on DC coupling?
Is there any DC at the audio outputs?
What is the DCV before the output coupling capacitor, if any?
Should be near 0VDC if the op amp works on +/-12V.
Does the supply measure right? Seems it might be the positive DC supply.
Last edited:
Guys thanks for the help. I know of NO available schematics for Oppo bdp products. I admit, this is a " buy it broke, fix it" project. Tomorrow I'll look for the asymmetric issue. Yes, post #5 explains the output opamp offset. Should be zero, well it not dispite equal +- inputs from the dac chip or different opamp installed. Shows offset before caps, zero after caps if installed. DC scope coupling, no, watching thru ac coupling.. This distortion problem exists with or without output coupling capacitors installed. I've had 3 different opamps, the stock PS and an external linear PS running only the analog board with the same outcome. I'll look again at the opamp Power supply. Sag would explain, but again I've monitored it without any obvious issue concerning sag. I agree, normally it appears to be a PS issue. Been there, try again..
Datasheet for CS4382A is available at: https://statics.cirrus.com/pubs/proDatasheet/CS4362A_F2.pdf
An example output stage is shown on page 27 of the pdf, Figure 16. Chances are reasonably good that Oppo implemented the recommended output stage or else used something very similar.
Sometimes its necessary to trace out some circuitry and from that sketch up a schematic for the purposes of troubleshooting. Depending on what Oppo did maybe the datasheet output stage will be enough.
EDIT: As can be seen from Figure 15 of the datasheet, there should be about a +2.5v DC offset at the opamp inputs.
An example output stage is shown on page 27 of the pdf, Figure 16. Chances are reasonably good that Oppo implemented the recommended output stage or else used something very similar.
Sometimes its necessary to trace out some circuitry and from that sketch up a schematic for the purposes of troubleshooting. Depending on what Oppo did maybe the datasheet output stage will be enough.
EDIT: As can be seen from Figure 15 of the datasheet, there should be about a +2.5v DC offset at the opamp inputs.
It a bit hard for these old eyes to trace surface mounted parts and circuit traces. Still working on that.
I did trace the clipping back to the DAC chip output. The dc voltage measures 1.22 on one channel (both -+ outs), the other, 1.17 at the opamp input. Not the 2.5 on the suggested data sheet. Both channels clip on full volume. Stepping down something like 4 steps in the DAC chip volume setting pretty much eliminates the problem/clip issue. Checked the PS voltage to both Vd and Va, spot on. What parameters affect the output offset?
I did trace the clipping back to the DAC chip output. The dc voltage measures 1.22 on one channel (both -+ outs), the other, 1.17 at the opamp input. Not the 2.5 on the suggested data sheet. Both channels clip on full volume. Stepping down something like 4 steps in the DAC chip volume setting pretty much eliminates the problem/clip issue. Checked the PS voltage to both Vd and Va, spot on. What parameters affect the output offset?
Well, if they left out the 22uf electrolytic cap on the circuitry for the + opamp input, then the resistors would form a DC voltage divider. Depending on resistor tolerances the exact voltage could have some variation. Have to trace out the actual circuit they used and sketch up a schematic if you want to analyze it.
I'd replace the audio coupling caps and look at anything in the negative FB loop of the output buffers.
I've have a non modded unit to compare. It also has the 1.2 vdc offset on the input to the stereo(-+ input channel pins) opamp. However there is no clipping at full volume at the dac output or output opamp input pins. I have run the problem unit with other film coupling caps and without output coupling caps with no change. Preliminary inspection of the top side of the circuit board shows a circuitry not the same as the data sheet recommended(4382A) out circuit. No 22uf cap present. Still looking.
The posted picture shows one channel +-input traces in upper right each going into equal value resistors. There are no components (at least on the top side of the CB) between the dac output pins and the 01C resistors. It appears that both +- traces are the same. Tracing one path gives you both. Pin 5 is +, 6 -, 7 output. I don't see any vias along the path.
Here's a full picture. I don't think the output is at
fault as the clipping issue shows on the 4382 outputs.
Hi, in one of posts op mentioned there was a voltage regulators problem with ripple . There's some possibility, that dac chip get too high voltage and gets damaged . Try to measure dc voltages at all dac chip outputs , should be very close to each other. If there's no dc coupling capacitors between dac and opamp , any difference will result in assymetry after opamp and dc offset. Also , if opamps are supplied with +-12v , output should have no dc . Otherwise they would have single supply like +12v with coupling capacitors necessary.
Looking at the pics, kinda looks like the output path for one opamp is marked in red below. Hard to tell for sure from the pic, so you might want to check. Don't know where it goes after the electrolytic coupling cap.
Beyond that each opamp has + power on pin 8 and - power on pin 4. What are the voltages, DC, AC, scope?
Other voltage to check in the same way are shown on pages 19 and 20 of the dac chip data sheet. They are VA, VD, and VLC. Also good to check the voltages at the pins for FILT+ and VQ. In particular bad parts or bad solder joints in any of the above areas might have some effect on distortion.
In addition, looks like there are some SMD transistors marked with Q designators on the silk screen. Don't know what they are for, but might be a worth a look with a scope while music is playing to see if analog audio in present on any of them. Also, sometimes its easier to use a scope if solder a ground pin or test point somewhere on the board to hook the scope probe ground clip. Good to keep the probe ground path short if possible.
Beyond that each opamp has + power on pin 8 and - power on pin 4. What are the voltages, DC, AC, scope?
Other voltage to check in the same way are shown on pages 19 and 20 of the dac chip data sheet. They are VA, VD, and VLC. Also good to check the voltages at the pins for FILT+ and VQ. In particular bad parts or bad solder joints in any of the above areas might have some effect on distortion.
In addition, looks like there are some SMD transistors marked with Q designators on the silk screen. Don't know what they are for, but might be a worth a look with a scope while music is playing to see if analog audio in present on any of them. Also, sometimes its easier to use a scope if solder a ground pin or test point somewhere on the board to hook the scope probe ground clip. Good to keep the probe ground path short if possible.
All voltages have been measured and scope monitored, all in spec. I believe as ximikas indicates, due to the regulator failure (has been repaired) somehow some part of the 4382A was damaged. The clipping is present at the dac chip output, thus the output opamp just reflects what it's feed. As rayma has mentioned it's possible to reduce the dac output volume setting to live with this problem (stops the clipping). How much this will affect the quality of the signal due to bit chopping I don't know. I don't have the skill or equipment to change the 4382A.