Project

General

Profile

Actions

Issue #1926

open

Maxim Power Management ICs: find how much mA it can provide, and check if it's configurable

Added by Denis 'GNUtoo' Carikli almost 5 years ago. Updated over 2 years ago.

Status:
New
Priority:
Normal
Category:
Wi-Fi
Target version:
Start date:
05/06/2019
Due date:
% Done:

0%

Estimated time:
Spent time:
Resolution:
Device:
Grant:
Type of work:
C programming, Reverse engineering (no code nor documentation available)

Description

It's worth understanding more the characteristics of the MAX8997 and the MAX77693.

This could help in understanding if and how we can make external USB devices more reliable. That is needed as the internal WiFi chips don't work with free software at the time of writing.

The MAX8997 is used at least in the Galaxy SII (I9100) and the MAX77693 is used at least in the Galaxy SIII (i9300), Galaxy SIII 4G (i9305), the Galaxy Note 2 (N7100) and the Galaxy Note (N7100).

Some more background information can be found in the WiFi wiki page.


Files

instr1.png (10.3 KB) instr1.png Henrik Grimler, 10/24/2021 11:41 AM
instr2.png (10.5 KB) instr2.png Henrik Grimler, 10/24/2021 11:41 AM

Related issues

Related to Replicant - Issue #2011: Find a public version of the MAX77693 datasheet / reference manualNew02/24/2020

Actions
Actions #1

Updated by Denis 'GNUtoo' Carikli almost 5 years ago

Moving forward:
  • Since no documentation can be found on the Maxim Power Management ICs (PMICs) it is a good idea to do some measures to find the maximum amount of mA that can be delivered by measuring it.
  • The MAX77860 has a configurable current limit for USB OTG. It is configurable through the OTG_ILIM register which sets the limit when the device is in current sink and current source modes. It might be interesting to do some measurements and see if the chips mentioned above are similar enough and to check if the charging limit has an effect on how much mA can be delivered through the USB port in current source mode.
Actions #2

Updated by Denis 'GNUtoo' Carikli almost 5 years ago

Having access to a current-sink/load might be useful to do that.

Actions #3

Updated by Kurtis Hanna over 3 years ago

  • Related to Issue #2011: Find a public version of the MAX77693 datasheet / reference manual added
Actions #4

Updated by _I3^ RELATIVISM about 3 years ago

  • Type of work C programming, Communication (mails, contacting people, etc) added
Actions #5

Updated by _I3^ RELATIVISM about 3 years ago

  • Type of work Reverse engineering (no code nor documentation available) added
  • Type of work deleted (Communication (mails, contacting people, etc))
Actions #6

Updated by Henrik Grimler over 2 years ago

i9300 running replicant 6 can provide around 820 - 850 mA to an OTG device. I measured it by sweeping the current from 0 to 1 A with two different galvanostats, with some sweep rates between 5 mA/s and 50 mA/s, and aborted when voltage reached below 4 V.

It seems that it might be the voltage rather than current that decides when the OTG power is turned off, once the voltage drops to 4.66 V the connection is terminated. When I tried with two different galvanostats this corresponded to 820 mA and 850 mA, respectively. It is a bit weird that the two instruments gave different values, I will need to run some diagnostic tests to see if there is an issue with one of the galvanostats (I trust the one that gave 850 mA the most).

As far as I can tell these are hard limits, trying to draw 820/850 mA will turn off the power, but at 5 mA below that limit the phone can provide power continuously (at least for 30 s) without issues.

The battery in my tests was charged to around 50 %, being at 100 % or 5 % might give slightly different results.

In the attached graphs you can see how the voltage changes with current, and that it turns off once the voltage reaches 4.66 V. It is a bit hard to see from just the plots, but the voltage drops linearly with increasing current. The raw data shown in the plots can be downloaded here.

Actions

Also available in: Atom PDF