As far a high pitch sounds go, there's a number of DC converters in the Nook.
The screen uses +/- 15V to drive it.

Well, I don't have to go back to stock really.
I know how the power consumption is supposed to work.

I do know that USB host mode is a power hog, and it's not through supplying current to external devices. Even with a powered hub it uses a lot.

I think that I might have fixed it.
Before, I was having it connect on ADB over WiFi when it was supposedly asleep.

Originally Posted by Renate NST

I do know that USB host mode is a power hog, and it's not through supplying current to external devices. Even with a powered hub it uses a lot.

I was under impression, with USB host mode enabled it drains battery a lot without any USB slave device connected…
Am I wrong about it?

Originally Posted by Renate NST

I think that I might have fixed it.
Before, I was having it connect on ADB over WiFi when it was supposedly asleep.

What have you found?

---------- Post added at 05:29 AM ---------- Previous post was at 05:25 AM ----------

Renate,
Since you have milliammeter connected, could you tell what the consumption is with USB host mode enabled and disabled (without any USB slaves)?

Here's some numbers that I've collected.
Measuring current at the micro USB can be difficult until the battery is fully charged.
Even then, the charging circuit will try to "top up" with pulses for a while.

Power consumption:
Full power off: 9 mA (This is not what would be consumed if the charger were not connected.)
On and idle: 65 mA
Booting up: 160 mA (100% CPU)
Host mode: 160 mA (nothing attached)

Power drain:
USB keyboard: 5 mA (no LEDs on)
USB thumb drive: 65 mA (but still not working connected directly)

So the touch screen not shutting down is the worst scenario and you lose about 30% overnight.
Last night the touch screen did shut down, but I lost about 10% anyway.

Could you measure couple more, please:
Sleeping mode and WiFi (idle, no transfers)

Since you measuring between charger and nook, not between battery and nook, can we use something like P3
Just need to subtract AC 110v to DC 5v conversion loses, right?

Originally Posted by Renate NST

So the touch screen not shutting down is the worst scenario and you lose about 30% overnight.
Last night the touch screen did shut down, but I lost about 10% anyway.

You wrote early – you’ve found the cause and fit it, right?
Or was it something related to “Host mode: 160 mA (nothing attached)”?

Still where those 160 mA goes?
Is it simply maxing CPU out by polling something?

Originally Posted by ApokrifX

Still where those 160 mA goes?

After a lot of screwing around with this I've determined that measuring input current on the USB connector is too unreliable.
The problem is that the charging circuit switches in and out even when it says that it is disabled.
It is going through the charging circuit because the current settings modify it correctly and even turn it off completely.

This is similar to the problems with USB host mode. We can't get any straight info out of the driver.
It shows "await_vrise" quite often, even when it is in host mode.
Also, await_vrise should have a timeout in fractions of a second and should never be a steady state.

So, it looks like host mode is not using significant CPU time in any case.

Back to the original problem, I have seen how the touch screen does not shut off sometimes.
Rebooting it in full gets it back on track.
I've not seen what causes it to get in the non-shutdown mode.

Originally Posted by Renate NST

I've not seen what causes it to get in the non-shutdown mode.

If you can get logcat logs when nook does and does not shutdown properly and compare, it might give you a clue.
At least the IR detector diode you have let you see right away if screen didn’t shutdown properly.
Is there a way to send nook to sleep from command line?

So USB host mode without anything plugged in is causing some drain. Has anyone an idea of how much roughly?

I may have stumbled on something that MIGHT help here. I have always put my Nook to sleep by pressing the power button and after a year for heavy use, it is starting to sound and feel like it might be ready to break. So, I started looking for another way to make my Nook sleep. I found that the bottom button (looks like a power button icon) in button savoir will put my Nook to sleep if button savoir has Superuser permissions. So this is the method I have begun to use (for the last week or so). I have noticed that occasionally, even before I rooted my Nook, that pressing the power button would bring up the screen saver, but not actually put the device to sleep. Since I have been using button savoir to put my Nook to sleep, this has not happened once. Perhaps this procedure might help someone with this problem.

Originally Posted by David0226

I found that the bottom button (looks like a power button icon) in button savoir will put my Nook to sleep if button savoir has Superuser permissions. So this is the method I have begun to use (for the last week or so).

David: When I try that, my Nook (ST) responds with "Feature not supported on your device. Requires 2.2 and above." How is this working for you?

Originally Posted by wilo108

David: When I try that, my Nook (ST) responds with "Feature not supported on your device. Requires 2.2 and above." How is this working for you?

Hmmmm.

Good question. Unfortunately, I don't have a good answer. I will say that the first time I try it after a reboot, it writes a message to the screen that button savior has been granted superuser privileges. Then the next time and every time until I power off the device it works. I initially rooted using tinynoot, then tried MininalTouch without unrooting. So perhaps I have gotten my device into some weird state where this works. I wish I knew a lot more about android ....

Sorry.