Sansa Clip+ - no battery - powering down after 22 seconds

[jerzybulovski]

Hi all,

I've been given a Clip+ with a duff battery so I took it apart and removed the battery from the unit.

I have put Rockbox on it and I can power the device from a bench power supply via 4V DC applied to the terminals that the battery is normally wired to. I've run the unit from the bench supply for most of the day without any problems whatsoever. In fact, it worked perfectly.

Since I want to use the Clip+ in the car only (from a USB cable + 12V cigar adaptor), I thought that I'd be able to avoid putting another battery in and just be able to power it off the USB connector.

If I try this, however, about 22 seconds after powering the unit up, it goes off. This 22s is consistent for each power up so I think it has to be a timeout....

I did look into the "car adaptor mode" but I think this timeout cannot be 22s (or 20) but 15, 30, 45 etc...so I doubt it is that.

Does Rockbox test for the battery (even on USB "charge") and if it doesn't like what it sees, shut down?

Is there a setting that I can change to avoid this or do I have to amend the source code myself/rebuild?

Before anyone says, I do not wish to put a new battery in it

Thanks,
J

[saratoga]

Probably either the zero battery voltage or the lack of the battery thermal sensor.

Easiest thing to try would be to just put the old battery back in. 

[jerzybulovski]

Thanks for the reply.

I don't have the old battery, unfortunately - it was disposed of.

I guess it could be the thermal sensing because I imagine it's only checked during "charging" ; this wouldn't have been the case when I powered it from the bench supply.

If it's possible, and someone can point me to the right bit of the source, I don't mind modifying it and rebuilding it.

I suppose another option is to apply some signal to the thermal cutout terminal on the PCB to simulate the battery being OK?

Otherwise, I already have some bits to build a 4V regulator but I'd rather not do that.

Thanks,
J

[saratoga]

http://git.rockbox.org/?p=rockbox.git;a=tree;f=firmware/target/arm/as3525;h=060385f0cf7c55659cf6fd9d51c5caffb507e31b;hb=HEAD

Its most likely in the AS3525 target tree.  You might have to trouble shoot a lot to figure out where its happening though.

[jerzybulovski]

Thanks for the help, Saratoga.

I've just been briefly looking at the power management code and it looks like if the battery is read to below the critical voltage level, the player is shut down; however, I think this shouldn't happen if when USB charging...

I guess that brings us back to the thermal protection.

I'll have a look at the link. Thanks, again.

[jerzybulovski]

I've sorted this out.

Built code with "NO_LOW_BATTERY_SHUTDOWN" #define set and hey presto - a working device.

I think some of the other builds have this set for various reasons so I guess it's unlikely to end up with any flash corruption in using it (?)

Really couldn't have been made easier what with the build environment vm etc. The only thing that required a bit of guesswork was how to get the new build onto the device but, in the end, I just copied it over and it worked.

Great piece of software and good to see it being written in C.