[Bug 806] Standby voltage set to 2.1V, not 1.8V as datasheet says.

[bugzilla-daemon at bugzilla.openmoko.org]

http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=806





------- Additional Comments From speedevil at mauve.plus.com  2008-01-10 14:33 -------
Sorry about the delays - the bugzilla mail was getting sent to the wrong mailbox.

It looks like I was partially mistaken - I hadn't found a spec for vddalive, so
ignored it and simply looked at the vddrtc rating - which explicitly says 2.1v
isn't allowed.

Where I got the information from was
http://www.samsung.com/global/business/semiconductor/productRightMenuDown.do?doc_file=26537
- the datasheet linked from http://wiki.openmoko.org/wiki/Neo1973_hardware#Processor

On page 487, table 24-2 Recommended Operating Conditions

I think the initial problem is that the wiki page says
 STBY_1V8

    * Standby Voltage for S3C2410
    * Generated by PMU D3REG
    * Used by
          o S3C2410 (vddalive/rtc) 

On that table, Vddrtc is clearly stated as 1.8V+-0.15 at both speeds.
The datasheet never actually specifies as I can see it a voltage for VDDalive,
it merely implies it as notes on clock timing constraints.

2.1V is actually a bit above the recommended maximum of 1.95 of VDDrtc, so
operation is not guaranteed.
As the absolute maximum is 2.7V, damage will almost certainly not occur.

(the PMU supply in question has a granularity of 300mv as I understand it, as
indicated by your other calcs - typo I assume)

The 'right' thing to do would seem to be to on shutdown go to 200Mhz mode first,
then off.
This is at best a small power saving, 40uA at most.
Or to put another way will decrease standby in RAM time (with GSM and everything
else off by some 10%)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.