Hi Edo, thanks for your update. I actually think this is progress. The fact that replacing the TDA1305T did not change the behaviour strongly suggests that the DAC itself was not the root cause. What I find more interesting is that:

• cooling around C401/C403 restores the sound,
• replacing C401/C403 did not solve the problem,
• and TR405 starts oscillating when the audio disappears.

That makes me wonder whether C401/C403 are simply close to the actual faulty component, rather than being the fault themselves. At this point I would focus on TR405 and the components directly around it. It may be useful to measure the base, collector and emitter voltages of TR405 before and after the fault occurs. Once the freeze spray arrives, I would also try cooling TR405 and the surrounding components individually rather than cooling a larger area of the PCB. To me, TR405 and its local circuitry now look like stronger suspects than the DAC.

Hello, I begin to have some doubts regarding the possibility to repair the PCB32.

Right today, I have readen that the PCB 32 is a multi-layer board (4 layers); do you think that the issue could be in an internal conductive track?

In this case, we can not check nor repair internal tracks.

I would not worry too much about the 4-layer PCB at this stage. Yes, an internal track or via failure is possible on a multilayer board, but based on the symptoms I do not think it is the most likely explanation yet. The reason is that you already have several strong clues pointing towards a localized thermal issue:

• cooling around C401/C403 restores the sound,
• replacing C401/C403 did not solve the problem,
• TR405 starts oscillating when the fault occurs,
• the behaviour is repeatable and temperature-dependent.

To me, that still sounds more like a component, solder joint, via, regulator, transistor or local control circuit problem than a hidden internal PCB track. I would wait for the proper freeze spray and continue narrowing down the exact area by cooling individual components one at a time. In particular, I would focus on:

• TR405,
• components directly connected to TR405,
• nearby capacitors, resistors and diodes,
• and any ICs or regulators in the same section of PCB32.

Only if the freeze spray testing fails to identify a specific component or area would I start considering an internal PCB defect as a more likely possibility. At this point I still think there is a reasonable chance that the actual fault can be isolated and repaired.

Hello,

finally we had the possibility to test PCB32 components with the cooling spray.

the result was that the sound remains for more time cooling the LF33cdt.

So we will buy some and we will replace it in 2 weeks, 'cause the technician is going to holidays.

I will let you know.....

Thanks for sharing again Edo. This is actually the most promising clue so far. What caught my attention is that cooling the LF33CDT does not just bring the sound back temporarily, but appears to extend the operating time before the fault occurs. That makes the LF33CDT a much stronger suspect than the components investigated earlier. Since the LF33CDT is a 3.3V regulator, a thermal problem there could potentially explain many of the symptoms observed so far, including why the DAC initially appeared to be the culprit. If I were troubleshooting this, I would not only replace the LF33CDT, but also:

• Inspect and reflow its solder joints,
• Measure its input and output voltages before and after the fault,
• Monitor the 3.3V rail while the unit warms up,
• Check whether the output voltage changes when the regulator is cooled.

At the moment, this seems like the strongest lead you have had so far. Please keep us updated after the replacement.

Hello,

GOOD NEWS, AT LEAST!

We finally got to replace the LF33CDT and now the BS9000 sounds great!

Te tested it for more than 1 hour, without any problem.

The funny thing is that I have bought 4 of this component at the price of 7 Euros.

Once again, thank you for all your support.

 

That is excellent news! Congratulations on finding the real fault. This has turned out to be a very interesting troubleshooting case. In hindsight, the LF33CDT explains many of the symptoms. A regulator can become thermally unstable without completely failing, so measuring a nominal voltage with a multimeter does not necessarily mean the supply is healthy. Small voltage dips, oscillation or excessive noise can easily upset the DAC and surrounding circuitry. I think your systematic approach deserves credit:

• Verifying the digital signals,
• Replacing the suspected DAC,
• Using thermal testing,
• And finally narrowing the fault down with freeze spray.

Thank you for taking the time to report the final solution. Threads like this become very valuable for other Beosound 9000 owners who may encounter the same problem in the future. Great info for the community! Enjoy your rescued & fully working BS9000!

Hello, unfortunately I sang victory too soon.

Once replaced the LF33CDT, as you already know, we tested the BS9000 for about one hour, changing often CD and song, without any problem.

I brought it home and reconnected it.

In the same day I listened to various CDs without any problem.

The next day I turned on CD, and The CD begun to spin, the counter advanced properlìy, but no sound.

I tried more times, always with the same result.

The day after, I turned on RADIO from another room (trough Beosound 2000), radio worked fine, Then I tried with CD and it worked well.

Yesterday and the day before, always no sound. Not in yhe loudspeakers, nor in the headphones.

We have to start searching again where the trouble is...

There is something more: sometimes the commands do not reach BS 9000 (that is in option 0): happens sometimes that I give the command "CD" with the Beo4 to Beovision 7 (with option 2) and the BS 9000 does not turn on.

It even happened that, giving the command "n. music" with Beo4, the display on Beovision 7 is not "n. music", but "n. music 1" and Beosound 5 (in "master audio off") does not turn on.

So I think that:

BS 9000 is still not working properly;

I even have a problem with masterlink connections (or BS9000 working not properly, crashes the masterlink connections)

It's always more complicated.

Sorry to hear that Edo. The new symptoms actually make the diagnosis even more interesting. The fact that:

• the CD spins correctly,
• the counter advances normally,
• radio always works,
• and sometimes switching to Radio first and then back to CD restores the sound,

suggests to me that this may no longer be a simple thermal failure. I now wonder if there is an initialization, muting or power sequencing issue on PCB32. The replacement of the LF33CDT may have improved the original thermal problem, but there could still be another fault in the same area. I think it would be useful to compare two situations:

1. Cold start → select CD → no sound.
2. Cold start → select Radio → then switch to CD → sound OK.

In both cases I would check:

• Is the analog audio present at the output of the TDA1305T?
• Is the 3.3V output of the LF33CDT stable (preferably with an oscilloscope, not only a multimeter)?
• Does the muting circuit behave differently between the two cases?

I would also carefully inspect and, if necessary, reflow the solder joints around the LF33CDT, its input/output capacitors, nearby vias and connectors. A marginal solder joint or via could easily explain why the behaviour changes after source switching. The good news is that the investigation now seems to be focused on a much smaller area than before.

A maybe important thing I forgot to tell you is that in the last few days, I did 4 times the test TM71,  with CDs in all positions except #1, as suggested in service manual.

Any time the test did not get to its end, because first open and close the glass, and it's OK; then the sledge, attempting to move from position 1 to 6, always crashes into the CD in position 2, and the BS9000 turns OFF.