Hello Forum,
I am hoping to get some advice on where to start looking for problems that I can fix on this old sony str-de545.
I have the schematics from the sony service manual. Symptoms are right channel sound is running at about 20-30 percent of the left channel in terms of volume. Other than that function is normal. I have also found R699 and R691 are running hot. Around 105C. They are found in section 3-8 Main 3/3 Grid B5. The other thing that is weird with this are the voltages. Im seeing the B+ and B- running 11 volts higher than they are called out in the schematic. So The 46.7 B+ from the schematic is running at around 58V and the 30 volt is running around 41vdc. B- is the same magnitude but negative. I am not a analog guru and not the best at determining the voltages though these transistor circuits in the schematics. The nice thing is that sony has voltages on the schematic which reduces the needed knowledge for some of that. Examination on the right channel amplifier around Q751, 752 753, 754 the voltages are about what the schematic shows. Plus minus a tenth of a volt. The main observation Im seeing so far the differs from the schematic are the B+ and B- voltages. Where they are running at listed +11v (-11) compared to the schematic. What I am finding weird is that on the main schematic 3/3 page if I look at the B+ and B- voltages coming off the bridge it looks like on that node 9.7v is the voltage as it goes into a few 7805 regulators. on the nedigive side it looks like -9.7 on the schematic. But in the circuit down stream I was measuring those higher voltages. Weird.
How do folks usually diagnose an issue like this? Maybe put a sign wave in on both left and right channels and see how it differs as you work though the circuits. Making sure the magnitude is equal as you go though the stages?
On the hot resistors I'm not exactly sure where I should be looking to determine what has failed. Those are found in section mainboard 3/3 at the top of the page. I have already touched up all of the soldering on the back.
I have experience with electronics, but not in analog amplifiers. Suggestions on where overall to start looking for issues would be helpful, or advice, like toss it and buy a new one. Ha.
Thank you
Matt
I am hoping to get some advice on where to start looking for problems that I can fix on this old sony str-de545.
I have the schematics from the sony service manual. Symptoms are right channel sound is running at about 20-30 percent of the left channel in terms of volume. Other than that function is normal. I have also found R699 and R691 are running hot. Around 105C. They are found in section 3-8 Main 3/3 Grid B5. The other thing that is weird with this are the voltages. Im seeing the B+ and B- running 11 volts higher than they are called out in the schematic. So The 46.7 B+ from the schematic is running at around 58V and the 30 volt is running around 41vdc. B- is the same magnitude but negative. I am not a analog guru and not the best at determining the voltages though these transistor circuits in the schematics. The nice thing is that sony has voltages on the schematic which reduces the needed knowledge for some of that. Examination on the right channel amplifier around Q751, 752 753, 754 the voltages are about what the schematic shows. Plus minus a tenth of a volt. The main observation Im seeing so far the differs from the schematic are the B+ and B- voltages. Where they are running at listed +11v (-11) compared to the schematic. What I am finding weird is that on the main schematic 3/3 page if I look at the B+ and B- voltages coming off the bridge it looks like on that node 9.7v is the voltage as it goes into a few 7805 regulators. on the nedigive side it looks like -9.7 on the schematic. But in the circuit down stream I was measuring those higher voltages. Weird.
How do folks usually diagnose an issue like this? Maybe put a sign wave in on both left and right channels and see how it differs as you work though the circuits. Making sure the magnitude is equal as you go though the stages?
On the hot resistors I'm not exactly sure where I should be looking to determine what has failed. Those are found in section mainboard 3/3 at the top of the page. I have already touched up all of the soldering on the back.
I have experience with electronics, but not in analog amplifiers. Suggestions on where overall to start looking for issues would be helpful, or advice, like toss it and buy a new one. Ha.
Thank you
Matt
The 9.7 to 11v difference isn't significantly too high... I wonder if the others are higher because the manual covers different models and a key difference seems to be output power. The higher output variants would run on higher rails.
R691 and R699 are listed as 2 watt metal oxide types which suggests they may well run very hot.
I think you have to work with the fault you can easily observe and that is one channel being lower than the other. I would begin with a scope check of the input signals and then follow that through the audio stages.
You could 'split' the amp and look directly at the input to the power amp stages and see if the audio level is lower at that point.
R691 and R699 are listed as 2 watt metal oxide types which suggests they may well run very hot.
I think you have to work with the fault you can easily observe and that is one channel being lower than the other. I would begin with a scope check of the input signals and then follow that through the audio stages.
You could 'split' the amp and look directly at the input to the power amp stages and see if the audio level is lower at that point.
Hello Mooly,
Thank you for the response. Ok, Ill go with the theory that the rail voltages are normal. Yeah starting from a audio source where I can find the same input signal values and then moving though the amplifier stages is the best I can do at this point I think. Thank god the schematics and service manuals for these things are available, otherwise it would be a lost cause.
So as I work though this thing, presuming I located a bad IC. where does one go about locating replacement parts for these amps. In the little looking I have done, I have noticed the main boost stage amplifier transistors are not available from folks like Digikey or Mouser by the part numbers listed in the service manual. Im guessing some of these maybe some Sony custom made ic or transistor produced at a Sony fab which will limit their availability unless Sony has a repair store where they sell parts. Have you hear of such a thing or some other kind of repair part place? The amplifier ICs maybe that way. Transistors on the other hand I would hope that there maybe a cross. Are there any good resources for crossing some of these parts. For that matter just finding their datasheets.
Thanks again for your help!
Matt
Thank you for the response. Ok, Ill go with the theory that the rail voltages are normal. Yeah starting from a audio source where I can find the same input signal values and then moving though the amplifier stages is the best I can do at this point I think. Thank god the schematics and service manuals for these things are available, otherwise it would be a lost cause.
So as I work though this thing, presuming I located a bad IC. where does one go about locating replacement parts for these amps. In the little looking I have done, I have noticed the main boost stage amplifier transistors are not available from folks like Digikey or Mouser by the part numbers listed in the service manual. Im guessing some of these maybe some Sony custom made ic or transistor produced at a Sony fab which will limit their availability unless Sony has a repair store where they sell parts. Have you hear of such a thing or some other kind of repair part place? The amplifier ICs maybe that way. Transistors on the other hand I would hope that there maybe a cross. Are there any good resources for crossing some of these parts. For that matter just finding their datasheets.
Thanks again for your help!
Matt
This is quite a good resource for data sheets but at this point I wouldn't start thinking 'what ifs'. IC's are usually super reliable and way down the list of suspects.
http://www.datasheetcatalog.com/
Have you tried a factory reset? That might be the first thing to do actually. Page 6 of the service manual.
http://www.datasheetcatalog.com/
Have you tried a factory reset? That might be the first thing to do actually. Page 6 of the service manual.
I though I did the factory reset with it, but Im more then willing to try it again as that is easy and quick. Ill let you know how that goes. Thank you for the weblink.
Matt
Matt
So I had not done a factory resent, but a parameter clear. I did the factory reset and no luck same issue. To bad I was hoping for an easy fix. Ill have to break out the scope now.
Well... My 545 suffers from a similar problem.
When over 80 or 90% Volume, sound shifts to the left speaker. Both speakers end headphones.
It only happens with sound field selections, so I presume it is DSP related since Volume control is done by an encoder.
When over 80 or 90% Volume, sound shifts to the left speaker. Both speakers end headphones.
It only happens with sound field selections, so I presume it is DSP related since Volume control is done by an encoder.
Sony does not make any components in house, to my knowledge.
They tend to use bought out parts, and they had an associate company, name starting with S, that made small electrolytic caps.
But as in all DSP and complex stuff, there are continuous minor changes in the firmware, and you really need the service manual and a scope, apart from meters to check what is going on.
As a start, check for leaky caps and resistors, that and a wrong voltage might give a clue.
Clean the board thoroughly, use a magnifying glass to check for issues, then be thankful you have a good channel to compare, just see what happens on both sides, a reduced level will isolate the fault.
Then, for the most part, it is an easy fix.
OTOH, Trinitron TV had over clever designs, and were unreliable in Indian conditions, and Sony's practice of charging a premium for Chinese made junk has made them less than reputed.
They had a joint venture with Samsung for LCD panels in Japan, which they exited for financial reasons, and now are known to use very poor Chinese panels in their TVs, which occasionally de-laminate.
Only their top models use better quality panels, and their SMPS are flimsy.
So, at least in India, it does not have a good reputation, more eye candy.
But many people trade in the sets for larger screens while they are working, and if it breaks, the new buyer has paid a low price anyway...32" in working condition get about $60 here, new are triple at times.
They tend to use bought out parts, and they had an associate company, name starting with S, that made small electrolytic caps.
But as in all DSP and complex stuff, there are continuous minor changes in the firmware, and you really need the service manual and a scope, apart from meters to check what is going on.
As a start, check for leaky caps and resistors, that and a wrong voltage might give a clue.
Clean the board thoroughly, use a magnifying glass to check for issues, then be thankful you have a good channel to compare, just see what happens on both sides, a reduced level will isolate the fault.
Then, for the most part, it is an easy fix.
OTOH, Trinitron TV had over clever designs, and were unreliable in Indian conditions, and Sony's practice of charging a premium for Chinese made junk has made them less than reputed.
They had a joint venture with Samsung for LCD panels in Japan, which they exited for financial reasons, and now are known to use very poor Chinese panels in their TVs, which occasionally de-laminate.
Only their top models use better quality panels, and their SMPS are flimsy.
So, at least in India, it does not have a good reputation, more eye candy.
But many people trade in the sets for larger screens while they are working, and if it breaks, the new buyer has paid a low price anyway...32" in working condition get about $60 here, new are triple at times.
Those Sony STR's are notorious for bad solder joints.
I've had them on the service bench many times.
Overheating results in such issues.
Go over the whole main board - connectors, heatsinked components, resistors - check for poor/overheated solder joints.
I've had them on the service bench many times.
Overheating results in such issues.
Go over the whole main board - connectors, heatsinked components, resistors - check for poor/overheated solder joints.
Since you have a gain disparity, maybe start at the gain-setting parts in the power amp feedback loop. A likely candidate would the 100uF to 470uF capacitor that has one lead Grounded, and the other to the resistive divider that sets overall gain. The capacitor may be open (dry) or have poor solder.
This only applies if the sound quality of the soft channel is decent; if there's distortion, a DC offset, or other sound quality issues, you'll have to look for other causes.
This only applies if the sound quality of the soft channel is decent; if there's distortion, a DC offset, or other sound quality issues, you'll have to look for other causes.
Yeah I found several failing solder joints and have repaired those with some good 63% lead no clean solder. I have to say I am amazed that Sony after all these years and with the advent of dirt cheap Chinese pcb's with plated thru-hole boards they still use the cheap old one layer non plated thru hole boards. They could still do one layer, but just plate the holes. Would make their lead free solder work better in my opinion. But selling maybe million units the few cents adds up vs the negative product endorsements.
Rick, yes the soft channel sounds pretty clear with minimal distortion as well as I can hear. So Ill look for those caps and start replacing them. Im sure I have some good 105C 5k hour caps around this place.
NareshBrd, I have the service manual and a scope so I can and will be looking further into each channel and working forward to see if I can find the issue.
MAAC0 Unfortunately this unit has the issues with a softer volume all the way down to the bottom of the volume range. I have to crank it up significantly to hear it. When I switch to the other channel its blasting at that volume.
Thank you all for your responses!!
Matt
Rick, yes the soft channel sounds pretty clear with minimal distortion as well as I can hear. So Ill look for those caps and start replacing them. Im sure I have some good 105C 5k hour caps around this place.
NareshBrd, I have the service manual and a scope so I can and will be looking further into each channel and working forward to see if I can find the issue.
MAAC0 Unfortunately this unit has the issues with a softer volume all the way down to the bottom of the volume range. I have to crank it up significantly to hear it. When I switch to the other channel its blasting at that volume.
Thank you all for your responses!!
Matt
Ok, I have been tearing this thing apart and still struggling to get it fixed. I followed the signals back to the dsp processor. I was not able to really see the signal on the input side of the input side of the integrated UPC2581 amps. Their output showed the smaller signal on the right channel. Between the amp and the dsp the signal had so much noise on it on my scope I was not able to see anything other than noise down in the lowest voltage setting on my tek 3034c scope. There is not much there in that area between the dsp and the upc amps. A few in series caps which I replaced for good measure and the muting circuit. I disabled the muting circuit by pulling that transistor that muted the signal and no change. So If there is something in that area of the amp that has failed then I cant see it. The next area I focused on was the primary boost amp. This one is difficult for me cause I don't understand how the amp works. My amp expertise is limited to what I did in school 30 years ago and that was minimal at best. the mail signal line right front (RF) exhibits the problem and is visible with the scope with out problem. Plenty of signal. It looks like its around 10% of the signal of the other channel. The only thing I can see that is out of normal from voltages marked on the schematic is around the current detect part of the amp. around Q755 I'm seeing C=-55 B=+0.0015(Should be -0.5), E=-0.05(shouldbe-0.5) On Q756 C= -0.05(should be-0.6), B=-0.02(should be 0) E=+0.003 (should be zero). I was thinking that maybe I have a bad transistor there so I swapped them from another channel. No change. At this point my last thing I have to try which I think is pretty limited at best would be to swap the main boost transistors q754 q753, but from what have mentioned here does anyone think that might work at this point? I'm just about to put this thing in the recycling as I don't think I want to spend to much more time on it unless someone has some ideas. Cause I'm just about out of them for this thing.
Thank you
Matt
Thank you
Matt
Work with that.
If you can see the right channel output is low then you must confirm the input is also low (or not).
Your scope should be able to resolve this but you must make sure you are using the correct audio ground point to minimise noise. It is simply the ground used in that area of the circuit that goes back to pins 35 and 36 on the processor chip. Anywhere on that line should be fine.
If you touch the probe tip to that ground (with the ground lead connected of course) then you should really see nothing, no noise or hash. If you do then you might have a ground loop issue with the measurement set up.
You should see the same signal level on C701 and C751
A quick and dirty test is just to lift the PLUS end of either of those caps and link it to the other channels cap. That way both power amps get identical signals, and you should see the same output on both.
There look to be two outputs per speaker feed L- and L+ which means it is used in a bridged mode. Never ground either output but you should see the same signal on each although one will be 180 degrees phase inverted. The scope won't show the phase inversion using a single channel on the scope, but if both signals are present you have to assume they are OK.
Hello Mooly,
Thank you for the ideas. Ill look into that. I really did not want to pull those heatsinked transistors and hopefully this will save that action.
Ill let you know how that goes. The voltages that were inconsistent with the schematic around the current detect circuit are not far enough out to be an issue potentially?
Thank you
Matt
Thank you for the ideas. Ill look into that. I really did not want to pull those heatsinked transistors and hopefully this will save that action.
Ill let you know how that goes. The voltages that were inconsistent with the schematic around the current detect circuit are not far enough out to be an issue potentially?
Thank you
Matt
The voltages you measure look correct, those in the manual look suspect (which often happens).
If you look at what those voltages are you will see they reflect the voltage across the 0.22 ohm emitter resistors of the output stage. 0 volts on the emitter of Q754 and -0.5 volt on the base of Q755 implies 0.5 volts across the 0.22 ohm. That would give a current flow of 2.2 amps in the output stage.
Also the junction of the two 0.22 ohm resistors should be at approx 0 volts and that would be the DC offset value. The manual shows different voltages for the two stages, Q755/6 and the identical stage above it Q605/6. The stages are identical, you have to mentally juggle that one is drawn 'upside down' compared to the other.
If you look at what those voltages are you will see they reflect the voltage across the 0.22 ohm emitter resistors of the output stage. 0 volts on the emitter of Q754 and -0.5 volt on the base of Q755 implies 0.5 volts across the 0.22 ohm. That would give a current flow of 2.2 amps in the output stage.
Also the junction of the two 0.22 ohm resistors should be at approx 0 volts and that would be the DC offset value. The manual shows different voltages for the two stages, Q755/6 and the identical stage above it Q605/6. The stages are identical, you have to mentally juggle that one is drawn 'upside down' compared to the other.
Hello Mooly,
Well I think I found it. Looks like the culprit was C751 a 4.7nf cap located between the mute and the IC702 coming off pin 16 of the processor. I was still not able to get a clean look at the sinewave I was putting into the amp via the optical port. I was getting noise down in that section of the circuit. I even tried as you suggested and mounted a ground right at the analog ground next to the audio processor. It did improve the noise but only by about half. What I noticed was the Amp looks to be differential between the processor and the upc2581 parts. From there I went and looked at those parts on that other input. I found the second cap was bad. I did swap it out with a 10uf instead of the 4.7. I don't have any 4.7s at the moment. What I did notice is that the sound is just marginally lower on the right channel now which I can adjust with the fader. With the larger caps in series I would have though that would have been better for energy transfer and made it louder. But maybe there is still another minor issue. Sony choose to use electrolytic though out the design and even in places I would have placed ceramics, like the 10uf and 4.7uf parts I have been replacing. Would replacing them with ceramics some how negatively effect the amplifier performance? Maybe its simply economics with respect to thru hole parts.
Mooly thank you for your help and encouragement. Without it that Amp would be at the recyclers as you read this. Its currently running. 🙂
Matt
Well I think I found it. Looks like the culprit was C751 a 4.7nf cap located between the mute and the IC702 coming off pin 16 of the processor. I was still not able to get a clean look at the sinewave I was putting into the amp via the optical port. I was getting noise down in that section of the circuit. I even tried as you suggested and mounted a ground right at the analog ground next to the audio processor. It did improve the noise but only by about half. What I noticed was the Amp looks to be differential between the processor and the upc2581 parts. From there I went and looked at those parts on that other input. I found the second cap was bad. I did swap it out with a 10uf instead of the 4.7. I don't have any 4.7s at the moment. What I did notice is that the sound is just marginally lower on the right channel now which I can adjust with the fader. With the larger caps in series I would have though that would have been better for energy transfer and made it louder. But maybe there is still another minor issue. Sony choose to use electrolytic though out the design and even in places I would have placed ceramics, like the 10uf and 4.7uf parts I have been replacing. Would replacing them with ceramics some how negatively effect the amplifier performance? Maybe its simply economics with respect to thru hole parts.
Mooly thank you for your help and encouragement. Without it that Amp would be at the recyclers as you read this. Its currently running. 🙂
Matt
C751 is a coupling cap and is in series with C465 from pin 16. Two caps are used to allow the transistor mute circuit to operate and switch silently (Q365)
Electrolytics are probably the best part for the job tbh and used correctly are transparent to audio. I wouldn't start looking at replacing with non polarised tbh.
I'm going to say you've done really well to find a faulty cap like that but I'm also a bit 'doubtful' because replacement of a part like that if faulty should 100% fix the fault. The only time you get multiple issues is when something untoward has happened in the units past such as spillage for example.
A larger value cap simple alters the high pass characteristic of the cap and R752 it works into. Those two parts form a high pass filter (we can ignore the 1k). The 10uF will fractionally lower the low frequency cut off point but you won't tell any difference.
Is there any chance heat from unsoldering and replacing that cap has actually heated other parts nearby and masked the real problem?
It might be worth looking at Q365, perhaps even just isolating the emitter to see if the level comes up fully. Also, the 10uF C465.
Electrolytics are probably the best part for the job tbh and used correctly are transparent to audio. I wouldn't start looking at replacing with non polarised tbh.
I'm going to say you've done really well to find a faulty cap like that but I'm also a bit 'doubtful' because replacement of a part like that if faulty should 100% fix the fault. The only time you get multiple issues is when something untoward has happened in the units past such as spillage for example.
A larger value cap simple alters the high pass characteristic of the cap and R752 it works into. Those two parts form a high pass filter (we can ignore the 1k). The 10uF will fractionally lower the low frequency cut off point but you won't tell any difference.
Is there any chance heat from unsoldering and replacing that cap has actually heated other parts nearby and masked the real problem?
It might be worth looking at Q365, perhaps even just isolating the emitter to see if the level comes up fully. Also, the 10uF C465.
Hello Mooly,
Thank you for the compliment on potentially finding that cap. It was not as much fun as I would have hoped. I almost gave up. But was a good learning experience. The main thing that I struggle with is having a pile of schematics dropped in your lap and trying to understand what is going on. Once you have a clear idea of generally how and why all the circuits are there and doing it makes it a lot easier to diagnose. My failure initially was not seeing that the signal from the sound processor was a differential signal going to the integrated amp. I though it was single ended at first. So I failed to follow the other signal path and find that cap initially. I usually design more microcontroller embedded systems electronics and rarely fix others designs. I will say I do like how well the sony folks have done on documenting their circuits. Something I could try to achieve in my own designs, but we usually go so fast that we dont have the time for that level of detailed documentation. As small as we are Im usually the only one who looks at the schematic anyway. That is until I retire.
I will look at q365 and see if lifting the emitter brings the level up. This unit has had no other issues as far as I know like spillage. This unit was purchased new by my parents maybe 15 to 20 years ago and been in their tv stereo rack until it started giving them trouble. They then pulled it and placed it into a box where it sat for a few years before I started looking at it here in the past 30 days. So no spillage like events. Maybe general poor air flow around it. I have touched up almost every solder joint on the circuit board with a 63% leaded solder. I was mainly thinking theses cheap Sony caps have maybe dried out over the past 2 decades of heating. They are cheap enough to reorder from digikey. So when you say heated other parts nearby and possibly masked the problem, are you suggesting that maybe there was a cold solder joint that I possibly fixed? Or are you suggesting something else?
Talk with you soon Mooly!!
Thanks again for your help!!
Matt
Thank you for the compliment on potentially finding that cap. It was not as much fun as I would have hoped. I almost gave up. But was a good learning experience. The main thing that I struggle with is having a pile of schematics dropped in your lap and trying to understand what is going on. Once you have a clear idea of generally how and why all the circuits are there and doing it makes it a lot easier to diagnose. My failure initially was not seeing that the signal from the sound processor was a differential signal going to the integrated amp. I though it was single ended at first. So I failed to follow the other signal path and find that cap initially. I usually design more microcontroller embedded systems electronics and rarely fix others designs. I will say I do like how well the sony folks have done on documenting their circuits. Something I could try to achieve in my own designs, but we usually go so fast that we dont have the time for that level of detailed documentation. As small as we are Im usually the only one who looks at the schematic anyway. That is until I retire.
I will look at q365 and see if lifting the emitter brings the level up. This unit has had no other issues as far as I know like spillage. This unit was purchased new by my parents maybe 15 to 20 years ago and been in their tv stereo rack until it started giving them trouble. They then pulled it and placed it into a box where it sat for a few years before I started looking at it here in the past 30 days. So no spillage like events. Maybe general poor air flow around it. I have touched up almost every solder joint on the circuit board with a 63% leaded solder. I was mainly thinking theses cheap Sony caps have maybe dried out over the past 2 decades of heating. They are cheap enough to reorder from digikey. So when you say heated other parts nearby and possibly masked the problem, are you suggesting that maybe there was a cold solder joint that I possibly fixed? Or are you suggesting something else?
Talk with you soon Mooly!!
Thanks again for your help!!
Matt
I was thinking more 'components' than 'joints' tbh and how additional external heat can sometimes bring a faulty or intermittent part back to working condition for a while. The real key to finding these issues is to use the scope. It might be worth spending a bit of time seeing why you can't get a clean viewable signal. Grounding is super important with test equipment on circuits with a lot of digital processors running.
Yeah Im not sure on what the deal with the noise is. Do you have any thoughs on the peak to peak values for the signals that are coming from the sound processor might be? I wonder if I need to try looking at those signal ina differential mode on the scope to improve the noise floor. Two probes and take the difference of the two signal on the display. One of those math functions in the scope. I was able to hook the ground right to the analog ground at the processor. That helped but did not totally eliminate the noise. I also ran my scope in the 20mhz bandwidth setting which also helped from the full bandwidth setting. I half wonder if it might be my probes? Ill have to dig around and see if I have any new ones. Maybe Ill check a different scope too. Unfortunately I only have digital scopes. I think I have another 3054 around here somewhere. I might also try switching these things over to battery mode to maybe reduce the noise signature from them? I cant make it today, but maybe later in the month I can go over to the Tektronix surplus store and see if they have any working old analog scopes. Every once in a while they can be found for less then $100 or so.
Interesting though on the fixing those parts with heat temporarily. I have not heard of that before, but good to learn. Its possible. I did go though and hit most of the components on that board with heat and new solder so I could have masked over the issue for sure.
Thank you Mooly!
Matt
Interesting though on the fixing those parts with heat temporarily. I have not heard of that before, but good to learn. Its possible. I did go though and hit most of the components on that board with heat and new solder so I could have masked over the issue for sure.
Thank you Mooly!
Matt