Tried a lot but it's all a big FAIL. I really love the Boom concept but
it's annoying that each one may just fail from one minute for another
and never return.
I have the second unit of this kind now. Got the first one dirt cheap
off eBay and, as I neither had any service info nor another Boom for
comparison, I made quite some mess out of it. Even ended up trying to
replace the processor because I suspected it of having fractured solder
balls, but with little experience I screwed it up completely. The board
is now a donor, and all other parts of it likewise.
Now the second unit that I recently bought behaves just the same:
- powering up just illuminates the keyboard backlight
- the unit does not react to any buttons pushed (so any form of
factory/Xilinx/other reset is useless)
- no ethernet connectivity (same for WiFi of course)
- no display (neither the filament wires glow faintly as they used to,
nor is there anything shown on the display)
- no sound
- no reaction to IR remote control (except at the moment of powering
up where it prevents the keyboard illumination)
- unit stays rather calm in comparison, and consumes way less power
- this situation appears out of nowhere
- the power supply is not to blame as it flawlessly powers other Booms
I now have multiple other Booms available for comparison but I found
nothing so far. All the crucial voltages are at least close enough,
nothing there to explain why one Boom is not turning on while the other
is. Measured and compared voltages in ~50 spots but everything looks
fine. Reseated the WiFi board, also tried to run the mainboard without
the WiFi board in place, tried all sorts of resets. Took a Flir camera
to find any excessive heat spots but none can be found except the usual
suspects. I noticed though that a working Boom gets considerably warmer
in most areas.
The only difference in the broken one, compared to a working one, is
that some voltages are not turning on, probably because they activate
only after the Boom processor has finished booting. Also looked for
shorted capacitors and missing parts (I had cases where parts actually
fell off the board!) - no avail.
Looks like the processor gets stuck well before being ready to accept
any commands. Like it does not get a clock frequency or a voltage rail
is missing. This is rather hard to measure because most pins are covered
due to the nature of BGA soldering, and the other side of the board is
usually covered by the display, making it hard to probe anything there.
Today I tried to replace the CPU, using way better equipment than last
time, and presumably with more success. I am confident the soldering job
worked out well, the new processor is spot on in the right position and
has the same (low) clearance from the board in all places. So I am
basically positive all processor contacts reach the board.
Just for kicks, while the CPU was off, I powered the board and found
that the behavior is exactly the same as before: keyboard backlight on,
nothing else. So the unit is basically braindead, no matter whether a
CPU is present or not. Needless to say, putting the new CPU did not
help, even though the replacement process went smoothly.
I am just a hobbyist, not an electronics engineer - unfortunately! No
idea how to go on from here. And I think I went rather far, but no
compensation. I need to confess this is way beyond my skills. The fact
that lots of SMD components on the board cannot be identified by their
marking (even Google does not know them) adds well to the frustration.
It is practically safe to assume the Boom mainboard has a JTAG interface
somewhere. It might help one of the experienced to learn at which point
the processor is failing. But you would have to know and understand the
firmware, the processor instructions, and a lot of the Boom architecture
to make any sense out of it. No chance to get there without any
documentation (at least for me).
Some observations I made, maybe others can verify:
- the display filament power (outer three pins on each side) goes up
to ~3V when the power supply gets connected, but slowly drops to 0V.
It should have stable 5V on the left and ~2.5V on the right. Seems to
me like a short is detected during power-up, and large parts of the
circuitry are immediately shut down
- the +55V display grid supply pin (1st pin on the left of the 12-pin
block) is at 0V all the time. Probably the HV generator is not turning
on because the CPU never tells it to
- likewise, a 1.5V current detected in a working boom at C185
(amplifier section, I guess) never turns on in the defective one
There are multiple threads in this forum covering the same issue. Sorry
if anybody gets annoyed I just started yet another one but I imagine it
might help to discuss this matter in a dedicated thread. And "Boom
suddenly died" is pretty close to what people affected will look for.
So this is for the people who could not recover their Boom with a
built-in reset function or reseating the WiFi board, or replacing the
power supply. Logitech won't help us any longer. Let's try to help each
other. I still feel the Boom is worth it.
5x Squeezebox Classic SB3
2.5x Squeezebox Boom
1x Transporter
1.5x Controller
------------------------------------------------------------------------
JoeMuc2009's Profile: http://forums.slimdevices.com/member.php?userid=23131
View this thread: http://forums.slimdevices.com/showthread.php?t=99020
_______________________________________________
Boom mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/boom
