Home of the Squeezebox™ & Transporter® network music players.
Results 1 to 10 of 10
  1. #1
    Senior Member JoeMuc2009's Avatar
    Join Date
    Jan 2009
    Location
    Muenster, Germany
    Posts
    152

    A new riddle Boom

    Hi all,

    another sick Boom just arrived. Besides a mostly burnt display, it also needed a new power supply connector. But that's not all. It can't connect.
    Here is the list of symptoms I could find out so far:


    • Logitech logo does not appear on startup. Instead, the display stays dark longer than usual and the first thing I can see is "obtaining IP address"
    • the Ethernet (wired) MAC in my DHCP server registers as 04:00:04:00:04:00 which clearly isn't right. But it gets an IP anyway. A device name is not shown in the DHCP server at all
    • in the Boom's setup menu (after pushing the "back" button for some seconds):
      • I can see both my Logitech Media Servers, however it is not possible to register with them, and the server logs do not show any contact
      • DNS server and Squeezebox Server are indicated to have IP address 4.0.4.0 which is nonsense of course
      • sometimes, the device's own IP address also reads 4.0.4.0
      • sometimes, the MAC address is shown as 00:00:00:00:00:00
      • sometimes, going to the settings requests a setup lock code which is 1024 (found that out while I could go into the settings without the code). Most of the times however, I can get into settings without the code
      • I have changed the MAC address manually to what is printed on the bottom label of the Boom but this setting is immediately lost again
      • setting the MAC is somehow restricted where it shouldn't be. The second digit of the first MAC segment can only be set to 0 or 1 when using the "up" button on the IR remote. Using "down" reveals the values 9, 8, 1, and 0. All other segments work as they should
      • the factory reset available in the settings did not help
      • in some submenus, random horizontal lines appear on the display's right half

    • the display shows two lines where the pixels are considerably darker than in the other lines. They do show the right content. The display is brand-new, the old one showed the same issue. This must be something the Xilinx processor is doing wrong in controlling the display
    • Xilinx and factory reset both cannot be triggered on startup, neither from the IR remote nor the main user panel. It seems the firmware jumps right past the phase where these keys would be evaluated (remember the Logo does not appear either)
    • running the PCB without the Wi-Fi card, or using a new Wi-Fi card, did not fix it either
    • if the Wi-Fi board is not inserted, the PCB ends up flashing "Wireless ca" in huge letters and kind of shifted (see attachment), probably that's the same text as the "Wireless card not installed!" message that pops up every seconds in the first display line in the menu


    This reminds me closely of the "MAC not set" issue I had (see https://forums.slimdevices.com/showthread.php?104987-MAC-not-set)
    The impression is that whatever causes the behavior is becoming worse over time. Settings are incorrect right from the start, but at least the IP address received from DHCP is valid at first. It gets lost after a bit of time though.
    What happens in the MAC setting screen indicates that some data line is corrupted and at least one bit is lost. The PCB does not have visible damage of any kind. From the "MAC not set" drama I remember that desoldering the flash EEPROM and creating a dump of it is likely to prove that the firmware image is actually okay, the issues appear to be more in the region of the CPU or the RAM, however, changing the RAM chip for a new one did not help last time.
    Also tried gently flexing the board, pushing the CPU, applied freeze spray, all to no avail.

    If only I could trigger the Xilinx reset!

    Looks like another death, dammit.
    Attached Images Attached Images  


    PN me if your Boom / Classic / Transporter display has issues!

    Blog: https://www.blogger.com/blogger.g?ri...50753#allposts

  2. #2
    Senior Member JoeMuc2009's Avatar
    Join Date
    Jan 2009
    Location
    Muenster, Germany
    Posts
    152
    Guys, this is cool. I saved it!!!
    In a microscope-assisted board inspection I found that the soldering does not look great. I resoldered the RAM chip for a start and that brought the device back to life. Just put flux around the contacts and heated each one up quickly. Then powered it up just for kicks. The Logitech logo appeared!
    And it spontaneously reverted to the correct MAC address and now keeps it. It's fully up and running again.
    Really happy now...


    PN me if your Boom / Classic / Transporter display has issues!

    Blog: https://www.blogger.com/blogger.g?ri...50753#allposts

  3. #3
    Senior Member JJZolx's Avatar
    Join Date
    Apr 2005
    Location
    Colorado
    Posts
    11,531
    Boom Doctor.

    Nice work.

  4. #4
    Senior Member
    Join Date
    May 2017
    Posts
    587
    You're the best. Luckily I live in the Netherlands so if one of my SB dies I've got help close by 😉
    SqueezeBoxes: 1x Transporter (Living room) 1x SB2 (shed), 1x Radio (Kitchen), 1x Boom (Dining room), 1x piCorePlayer (jacuzzi), 1x piCorePlayer (Garden) 1x OSMC + Squeezelite (Movie room), 1x Touch (Study 2), few spare unit's
    Server: LMS on Pi3 7.9.1. on PcP 3.21
    Network: AVM Fritzbox, Netgear Smart Switch 24p, 3x Ubiquity

  5. #5
    Senior Member
    Join Date
    May 2008
    Location
    Canada
    Posts
    5,057

    Filament starving

    Hi Joe - I was wondering if there is any chance to idently the part that is responsible for not driving correctly the VFD filament "heating". Have you identified the component on the board but not able to say what it is or still no idea where it is?

    Thanks

    Philippe
    LMS 7.7, 7.8 and 7.9 - 5xRadio, 3xBoom, 4xDuet, 1xTouch, 1 SB2. Sonos PLAY:3, PLAY:5, Marantz NR1603, JBL OnBeat, XBoxOne, XBMC, Foobar2000, ShairPortW, JRiver 21, 2xChromecast Audio, Chromecast v1 and v2, , Pi B3, B2, Pi B+, 2xPi A+, Odroid-C1, Odroid-C2, Cubie2, Yamaha WX-010, AppleTV 4, Airport Express, GGMM E5

  6. #6
    Senior Member JoeMuc2009's Avatar
    Join Date
    Jan 2009
    Location
    Muenster, Germany
    Posts
    152
    Quote Originally Posted by philippe_44 View Post
    Hi Joe - I was wondering if there is any chance to idently the part that is responsible for not driving correctly the VFD filament "heating". Have you identified the component on the board but not able to say what it is or still no idea where it is?

    Thanks

    Philippe
    Hi Philippe,

    to be honest, I never cared much about it. Even if we found the failing part, replacing it would only end in the replacement failing again some day. The circuitry needs an adjustment to eventually stabilize for long term operation. But that's definitely out of my range right now, I'm not good at reverse engineering and schematics will never become available.
    I just noticed that I never took a good photo from the Boom PCB's front side when the display was off. The only picture I have is in the attachment, the suspects being marked by the yellow box. The area is very close to the right-side connection of the filament wire, and usually hidden behind the display. One of the parts is getting really really hot (beyond my IR camera's 105°C limit). I'll have to check the voltages and currents in a defective and a working Boom to compare them - once I get to it. I also never took temperature measurements in a working Boom so it may be by design that this red-hot part operates at such a level. We'll see.
    It's probably an overloaded MosFET or a shorting capacitor that causes the issue.

    Regards,
    Joe
    Attached Images Attached Images  


    PN me if your Boom / Classic / Transporter display has issues!

    Blog: https://www.blogger.com/blogger.g?ri...50753#allposts

  7. #7
    Senior Member
    Join Date
    May 2008
    Location
    Canada
    Posts
    5,057
    Quote Originally Posted by JoeMuc2009 View Post
    Hi Philippe,

    to be honest, I never cared much about it. Even if we found the failing part, replacing it would only end in the replacement failing again some day. The circuitry needs an adjustment to eventually stabilize for long term operation. But that's definitely out of my range right now, I'm not good at reverse engineering and schematics will never become available.
    I just noticed that I never took a good photo from the Boom PCB's front side when the display was off. The only picture I have is in the attachment, the suspects being marked by the yellow box. The area is very close to the right-side connection of the filament wire, and usually hidden behind the display. One of the parts is getting really really hot (beyond my IR camera's 105°C limit). I'll have to check the voltages and currents in a defective and a working Boom to compare them - once I get to it. I also never took temperature measurements in a working Boom so it may be by design that this red-hot part operates at such a level. We'll see.
    It's probably an overloaded MosFET or a shorting capacitor that causes the issue.

    Regards,
    Joe
    ah crap, these are under the LCD, so that makes is a really painful fix. I was thinking about the current drained 100% of the time by the 3 diodes.

    And I would have a hard time to agree that you are not good at reverse engineering, come-on!

    Cheers
    LMS 7.7, 7.8 and 7.9 - 5xRadio, 3xBoom, 4xDuet, 1xTouch, 1 SB2. Sonos PLAY:3, PLAY:5, Marantz NR1603, JBL OnBeat, XBoxOne, XBMC, Foobar2000, ShairPortW, JRiver 21, 2xChromecast Audio, Chromecast v1 and v2, , Pi B3, B2, Pi B+, 2xPi A+, Odroid-C1, Odroid-C2, Cubie2, Yamaha WX-010, AppleTV 4, Airport Express, GGMM E5

  8. #8
    Senior Member
    Join Date
    May 2008
    Location
    Canada
    Posts
    5,057
    Quote Originally Posted by JoeMuc2009 View Post
    Hi Philippe,

    to be honest, I never cared much about it. Even if we found the failing part, replacing it would only end in the replacement failing again some day. The circuitry needs an adjustment to eventually stabilize for long term operation. But that's definitely out of my range right now, I'm not good at reverse engineering and schematics will never become available.
    I just noticed that I never took a good photo from the Boom PCB's front side when the display was off. The only picture I have is in the attachment, the suspects being marked by the yellow box. The area is very close to the right-side connection of the filament wire, and usually hidden behind the display. One of the parts is getting really really hot (beyond my IR camera's 105°C limit). I'll have to check the voltages and currents in a defective and a working Boom to compare them - once I get to it. I also never took temperature measurements in a working Boom so it may be by design that this red-hot part operates at such a level. We'll see.
    It's probably an overloaded MosFET or a shorting capacitor that causes the issue.

    Regards,
    Joe
    BTW, when you unsolder the VFD, do you use hot airflow and heat up all the pins to remove it intact or so you cut them and unsolder the remaining pieces one by one using just a soldering iron. I might change one soon and I was thinking of fixing as well the Vref for filament heating changing the (probably) mosfet. But the existing VFD is not fully dead, so conserving it might be a good idea


    Envoyé de mon iPad en utilisant Tapatalk
    LMS 7.7, 7.8 and 7.9 - 5xRadio, 3xBoom, 4xDuet, 1xTouch, 1 SB2. Sonos PLAY:3, PLAY:5, Marantz NR1603, JBL OnBeat, XBoxOne, XBMC, Foobar2000, ShairPortW, JRiver 21, 2xChromecast Audio, Chromecast v1 and v2, , Pi B3, B2, Pi B+, 2xPi A+, Odroid-C1, Odroid-C2, Cubie2, Yamaha WX-010, AppleTV 4, Airport Express, GGMM E5

  9. #9
    Senior Member JoeMuc2009's Avatar
    Join Date
    Jan 2009
    Location
    Muenster, Germany
    Posts
    152
    Quote Originally Posted by philippe_44 View Post
    BTW, when you unsolder the VFD, do you use hot airflow and heat up all the pins to remove it intact or so you cut them and unsolder the remaining pieces one by one using just a soldering iron. I might change one soon and I was thinking of fixing as well the Vref for filament heating changing the (probably) mosfet. But the existing VFD is not fully dead, so conserving it might be a good idea
    Envoyé de mon iPad en utilisant Tapatalk
    Hi there,

    I'm using a semi-professional desoldering iron from Ersa (https://www.kurtzersa.de/electronics...-2000-a-5.html), The 3rd picture is pretty close to my setup.
    It can be seen in action here on a SB Classic: https://youtu.be/XDod-obw6Iw

    The video illustrates how the entire solder around the pin becomes liquid on both sides of the pad before getting extracted by vacuum. Some pins always remain stuck and require a second pass but overall this makes it really easy to non-destructively remove almost any component with low risk of damaging the board or the extracted part.
    If you are wondering why the vacuum pump starts running *after* the solder was sucked away, that is because vacuum is immediately available and when I trigger it, it does not need to build up first because the entire system is under low pressure already. When the upper threshold of the pressure level is reached (i.e. vacuum gets too low), the pump restores it automatically. It's pretty noisy though so I cannot do any desoldering at night without disturbing the neighbors.


    PN me if your Boom / Classic / Transporter display has issues!

    Blog: https://www.blogger.com/blogger.g?ri...50753#allposts

  10. #10
    Senior Member
    Join Date
    May 2008
    Location
    Canada
    Posts
    5,057
    Quote Originally Posted by JoeMuc2009 View Post
    Hi there,

    I'm using a semi-professional desoldering iron from Ersa (https://www.kurtzersa.de/electronics...-2000-a-5.html), The 3rd picture is pretty close to my setup.
    It can be seen in action here on a SB Classic: https://youtu.be/XDod-obw6Iw

    The video illustrates how the entire solder around the pin becomes liquid on both sides of the pad before getting extracted by vacuum. Some pins always remain stuck and require a second pass but overall this makes it really easy to non-destructively remove almost any component with low risk of damaging the board or the extracted part.
    If you are wondering why the vacuum pump starts running *after* the solder was sucked away, that is because vacuum is immediately available and when I trigger it, it does not need to build up first because the entire system is under low pressure already. When the upper threshold of the pressure level is reached (i.e. vacuum gets too low), the pump restores it automatically. It's pretty noisy though so I cannot do any desoldering at night without disturbing the neighbors.
    A real proper desoldering station is something I’ve not invested into here. I have a hot air gun that I use to reflow my own SMD-base PCBs, desoldering tweezers and manual vacuum pumps but no proper station ... don’t know why I did not get one after all these years
    LMS 7.7, 7.8 and 7.9 - 5xRadio, 3xBoom, 4xDuet, 1xTouch, 1 SB2. Sonos PLAY:3, PLAY:5, Marantz NR1603, JBL OnBeat, XBoxOne, XBMC, Foobar2000, ShairPortW, JRiver 21, 2xChromecast Audio, Chromecast v1 and v2, , Pi B3, B2, Pi B+, 2xPi A+, Odroid-C1, Odroid-C2, Cubie2, Yamaha WX-010, AppleTV 4, Airport Express, GGMM E5

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •