Better characterization of suspend current

[matt_hsu]

於 五，2008-02-08 於 09:25 +0000，Andy Green 提到：
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi folks -
> 
> I was lying awake in bed at 3.30am musing about the suspend current.
> The smart battery says on it that it is "3.7V 1250mAh", so naturally I
> divided 1250 / 5 and I say to myself "wah!  Does it really mean we have
> a 250hr standby time now?  Didn't I see that kind of number on adverts
> for mobile phones?  
	Well, we had tried to measure the suspend current of Touch which from
HTC. Its suspend current is dropped between 3mA~4mA. In addition, its
resuming is pretty fast.

> 10 days?  Cool!"  Then doubts and suspicion crowded
> around me in the dark.
> 
>  - This "5mA" figure never got tied to any specific configuration I saw,
> it just got reported as a single figure.  That needs specifying so the
> tests are repeatable and can be understood in context. 
	Sure. It's necessary to have many repeatable and abnormal tests. So
automated power measuring is the right thing to do. ;-)

>  For example,
> what are we measuring here?  It is Vb at the battery terminal, right?
> 
>  - The GSM modem seems to be "off" according to Allen's last attack on
> the current monster.  Is it off?  What happens to suspend current when
> it is left on so in can wake on incoming call?  
	That's the next question I would like to know. 
	I think we could figure this out these days. But I need GSM people to
help out.   
> Because that will be a
> very common scenario indeed.
> 
>  - People avoided using batteries and use bench PSUs last I saw (hence
> nobody told me about charging completely broken).  What is the situation
> there for this measurement?  It matters because -->
	I sent this patch before. 

http://lists.openmoko.org/pipermail/openmoko-kernel/2008-January/000556.html

	The battery could be charged with 1A at kernel level. 
> 
>  - What we really fight is not the dreaded current monster but the
> shapeshifting sister monster (imagine the current monster in a Japanese
> schoolgirl outfit) the "Power Monster".  When I see people with bench
> PSUs they stuck them on 4.0V with insanely low impendence you expect
> from a bench PSU, so "5mA" in that context is 20mW.  But for a start the
> battery voltage is lower, 3.7V nominal so we would see 5.4mA, right?
> Well what about the efficiency of the various LDOs in the PMU across
> input voltage?  Maybe the overall power changes.  We need to measure Vb
> current over a few voltages that can be seen from a real battery and get
> an idea if overall power changes -- maybe it reduces as the battery
> tends towards 3.3V and LDO dropout goes down.
	Yes. The system would be shutdown immediately when the battery is about
3v3. I think we could add some detecting and handling at u-boot to avoid
booting up when system tends less voltage.
> 
>  - What we really need to be able to compute is a fairly reliable
> estimate of standby time with GSM modem "on for incoming call detect" to
> be arrived at from these measurements, and we need to check that
> lifetime on a real phone actually left in standby until it flakes out.
> If we can only manage 2 days or 3 days, well, it's not great but we have
> a reliable figure we can say to people and we can estimate any
> improvements we manage.
	It sounds great. I will let HW people to know this suggestion and to
see any possibilities to have this observation. 

	Cheers,

	Matt
> 
> - -Andy
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.7 (GNU/Linux)
> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
> 
> iD8DBQFHrB/wOjLpvpq7dMoRAmp4AKCHLzVflrGQ4qYLtlncCwxIaozjNQCfUCUF
> E4gUOO9UUj1Cv9Pqaw3pXbU=
> =1xQp
> -----END PGP SIGNATURE-----