A winter day in Moscow and a low power camera

[Artlav]

But if you can just disable the converter going to those loads using the converter itself during their off period then the current can be minimal.

Ah.
That's the piece i wasn't thinking about.

Don't look for a converter that can produce 20uA output and consume as much itself, but just connect any good converter to power ONLY the big loads like the camera, RF and the SD card, while using an LDO for the uC.
When it's done, shut down the converter.

In other words, i finally got it why there are "shutdown" pins on the converter chips.

Hm. Best case scenario is twice the run time for the same battery count (or half the battery count, actually).

Practical numbers for now are 3 seconds at 100mA for the camera to stabilize and get a picture, then 13 seconds at 60mA to write the picture down.
4 joules per picture, if we ignore the RF.
That is, 56Wh per year, or 4 D cells with a converter.

I'm leaving it overnight staring out of the window again...

Yeah if you just force the lines on the powered IC to be LOW or INPUT mode when the SD is shut off then that problem goes away, but of course there are also isolating bus transceivers.

Only problem is, i have the RF module share the same SPI bus.
So, full disconnect necessary.

[Artlav]

13 seconds to store a 48KBy image sounds a bit on the slow side if you're really burning an average of 60mA in that time frame.
...
If the MCU SPI write speed is the bottleneck, maybe going with a faster MCU or one with a more capable SPI / SD interface could actually save energy according to the "hurry up and get idle" scheme if you could get the writing done in 10% of the time or whatever.

The bottleneck is the UART camera, running at 115.2kbit/s.
Nothing i can do about that, short of starting from scratch.

However, i haven't optimized the storage system yet, to the point that the SD card might even be running at 100KHz (i get 4 kb/s, while the expected maximum is 12kb/s).

You could skip the filesystem (if you haven't already) for a block sequential storage scheme.  Start with a fully erased media to avoid the need to do any erase cycles on the flash during operation.
...
just write to a PCB mounted flash IC and then you could have much more low level control over the FLASH media / chip management which may lead to power saving opportunities

That's a flat out no.
The neat part of my current setup is that i can remove the card from a running camera (which will just go back to sleep if there is no card present), get it into a PC, watch the pictures, tweak setting in a config file, then plug it back in, and it would continue running without a hitch.

Using a raw write or an onboard FLASH chip would kill all these advantages at negligible gain.

[Artlav]

Made a project page for the camera, http://orbides.1gb.ru/lapse_camera.php , since this deviated in a new and wonderful direction that might be a shame to just lose in the forum archives.

The camera have been running for a day, and all seems to be well.
It made me wonder, whether the RF module is even needed.
All the settings are stored on an microSD card, and it is supposed to sit in some remote location.

Can anyone think up a good reason why a camera like that would need a (slow) RF module?

At first all i wanted is an RF-accessible camera, which grown into a time lapse one.
Time to split the two projects.

The time lapse camera, as it looks now.
Several problems with it - RF module sticks out, the card is in an inaccessible place, camera is not lined up with the mounting holes, etc.
Time for a new board.

If you can live-swap the SD card then the enclosure may be somewhat open environmentally, so being mindful to conformally coat / seal / whatever the thing is relevant for condensation prevention.

Hm, i planned on a sealed enclosure.
However, if it is to ever be opened, it should indeed be coated and prepared for any environment.

[Artlav]

Aaaand it's on Hackaday already.
http://hackaday.com/2014/12/22/a-year-long-time-lapse-camera/

Interesting.