Sansa Fuze+

[pamaury]

Jean-Louis Biasini sent me his first Fuze+ which suffers from the same problem (internal memory is not writable anymore). I'm looking at the problem but I have no solution for now. I will add support for booting from the sdcard if the internal storage fails. That will not fix the broken ones but hopefully prevent total failures for the others. Has anyone of you tried to send it back to sandisk ?

[luetzel]

No, I haven't sent it to sandisk. Meanwhile, I managed to see all partitions.
Looks actually exactly as in other posts that I've seen:

Disk /dev/sde: 15.9 GB, 15925772288 bytes
1 heads, 1 sectors/track, 31105024 cylinders, total 31105024 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000

Device Boot      Start         End      Blocks   Id  System
/dev/sde1   *      109568    31105023    15497728    1  FAT12
/dev/sde2            1024      101375       50176   53  OnTrack DM6 Aux3
/dev/sde3          101376      105471        2048    1  FAT12
/dev/sde4          105472      109567        2048    1  FAT12


 fdisk can remove/create them and write to the table, however, when trying to format it with "mkfs.vfat, /dev/sde1" it fails with an error.
Post Merge: April 22, 2012, 05:27:34 AMAh, I forgot to add that the OF starts, but shuts down because of a database error, requesting a reformat to fix it. The rockbox build installed is from 120417.

[megal0maniac]

I'm sad to say that my Fuze+ is dead.
Plugged it into my computer (while off), had the bootloader "USB mode" screen.
When I unplugged it, the screen was very bright (more than the usual CRT effect)

Could only see half of the icons, if I went into a menu and back to the main menu, all the icons were gone until I went into another menu.
I went into settings, got a "Divide by zero" error and turned it off.
Now it won't turn on, regardless of what I try.

Have even opened it and de-soldered and reconnected the battery, but it won't do anything.
Battery was low at the time, but still had about 2 hours of play time.

RIP Fuze+...

[pamaury]

Have you tried the recovery mode ?

[megal0maniac]

Absolutely nothing is working.
Not recovery mode, no charging, nothing.
It's totally dead.
Post Merge: May 07, 2012, 09:36:07 AM!!!!
Got it into recovery mode... But now what?
Does this mean there's hope?
Can I dump a firmware image like this?
Post Merge: May 08, 2012, 04:25:50 AMChatted to pamaury on IRC.
It seems as though the player can no longer locate the internal memory whatsoever.
A slightly more severe failure than most, but internal memory related once again.

I've been told that the only foreseeable method of recovery would be to modify the i2c eeprom to boot of a micro SD card, which wouldn't exactly be recovery as the internal memory would still be inaccessible, but would make the player usable once again by circumventing it completely.

The eeprom is write-protected, but this should theoretically be possible to change by pulling the write protect pin to GND.
Then it's a matter of knowing what to write and where to write it, I'd imagine.

Any credit for useful information goes to pamaury.
Just thought I'd post this for informative purposes.

Bottom line: My Fuze+ is essentially dead, without chance of recovery at this stage.
(It isn't possible to determine whether this is Rockbox related or not, but as it says on the wiki, you install Rockbox at your own risk.)

[Enclavejedi13]

I have the same problem as megal0maniac, except my internal memory is still intact. But I am unable to charge my Sansa Fuze+ at all. When I plug it in, it still shows the bootloader usb mode, but I can't even turn it on. What should I do?

[pamaury]

You should boot into the OF using volume down.
I manage to write the epprom on the Fuze+ Jean-Louis sent me but currently it requires to short pins. I'm still trying to find a software only way.

By the way I think I found the reason of the tuner bug. I committed a fix but did not test it yet so if someone can confirm/invalidate, it would be nice.

[megal0maniac]

I have submitted an RMA request to Sandisk.
Problem is I can't find proof of purchase and there's a scratch inside the cover from when I opened it, on the transparent plastic over the screen. 
So we'll see whether or not they replace it.

If they don't, I'm all for shorting pins out. It isn't like I could really make it worse...
In a silly way, I'm almost hoping that Sandisk don't replace it so I can try and fix it myself.

[halonachos117]

I have not noticed any bug that have been *fixed*  Maybe next time it will actually work.
Now, after the screen goes black, I can return to the radio screen, but the signal strength turns to 0dBuV, even though the signal sound great. If one tries to change the station, it freezes just like before. However, now you can return to the main menu, and power down. You can also power down from directly inside the radio screen.

And just like before, once I set my radio to some preset, I can't set the radio screen to seek to non-preset stations. As soon as I enter/return to the radio screen, it switches from "scan" mode to "preset" mode. I can set it back to "scan" mode, but that would be useless, since it will go back to "preset" mode as I describe above.

PS: I'm using version d26a35d-120509, if that make a difference.

PS2: Once again, thanks for being awesome, Pamaury!
Post Merge: May 09, 2012, 06:26:34 PMUpdate: Okay, so I've decided to try my luck, and updated just a few minuets ago. Now it seems to work much better, although I've only tested it for a few minuets. Perhaps your updateed radio code hadn't made it into the automated builds yet?

Anyway, so far so good. I will continue to test and see if I can find any bugs.

[pamaury]

Yes in fact I changed something after that along with some other things (pin use tracking and I fixed some could which would disable the gpio output on the fm i2c pins!). I have tested it today and the radio works reliably now. I also implemented RDS support but my fuze+ doesn't seem to have a good enough reception to meet our RDS parser requirement. I have implement a better RDS parser (see patch http://gerrit.rockbox.org/r/#/c/237/) but it needs discussion. I am interested in any feedback regarding RDS, am I the only one to have poor RDS reception ?

[alienkid10]

Can't test RDS since seemingly no stations around me use it, but radio works better now.

[buckminster]

I'm not getting anything for RDS (it just says "No RDS") when it works fine in OF.

I also noticed that the volume hardware buttons aren't functioning while on the Radio WPS. The presses are registered (wakes up the screen), but they don't adjust the volume. Touchpad volume control works fine. If I exit the Radio WPS, volume buttons work as expected.

EDIT:
On the plus side, I think your modifications fixed some problems I was having with playback from my SD card. Occasionally, I would get some garbling of the music and haven't been able to reproduce it since downloading the latest build. I don't remember the version of the previous build I had, but it was from the last few weeks (prior to your latest changes).

[pamaury]

sorry, I forgot that I did not commit the RDS stuff, it was only a patch. I finally committed the RDS support so now you can test it If it doesn't work and you can compile, I encourage you to apply http://gerrit.rockbox.org/r/#/c/237/ to see if it makes a difference.

[buckminster]

Running 1d21e54-120512 and no RDS being displayed (just "No RDS"). I just rebuilt my system and I don't have the time to download and compile right now. Is there any debug info I can provide you to help?

[pamaury]

Hi,
I would like to get some feedback on the audio volume of the radio. As you probably noticed, even at low volume the radio is still too loud. This is because the SoC cannot attenuate the tuner audio too much. Hopefully there is a way to attenuate the signal directly in the tuner but due to some limitation, the attenuation will have to be constant. So the question is: how much should I attenuate the radio output ? Of course, doing so will allow lower volumes but disallow higher ones. Alternatively, would a setting for this be useful ? (note that having a good default value is still important).