Announcement
Collapse
No announcement yet.
User Profile
Collapse
-
It's not even about RAM. A modern web browser will run like absolute garbage on this CPU, and I'm not even talking about loading a web page, just the web browser user interface.
About RAM, 4 GiB is acceptable to load a web page even on Windows 10, if you're not running other stuff in the background. But the northbridge will most likely not handle more than 2 GiB of RAM anyway.
I'm all for trying to keep old systems running for as long as possible, but trying to browse the web on this machine even with a lightweight-enough Linux distribution is going to be a nightmare. On...
Leave a comment:
-
If battery is plugged in, you'd have battery voltage on +19VB anyway.
Something is probably wrong in the charging circuit.
Leave a comment:
-
With this diode it is normal to observe a voltage on +19VB all the time.
Let's assume Schottky diode drops are 0.2V.
Let's say if +19VB was 0V due to a short to ground, then VB_DETECT will be around 0.2V, so Vgs of PQ601A is 2.9V which should be enough to turn it on. In that case VB_SHORT# is pulled to 0.2V, which turns off PQ601B, so that VB_SHORT is pulled up to 3.1V. Therefore PQB2_GATE will be 2.9V, so that ACIN_CHG is pulled to 0V.
If there is no current across PD604 at all, then +19VB is above 2.9V, VB_DETECT is 3.1V, Vgs of PQ601A is 0V so it is turned...
Leave a comment:
-
Version above is 46.
Here is XPS 9380 0x08AF NVM40 since I bothered extracting it to compare with a dump....
Leave a comment:
-
Latching comparator circuit.
If IN+ > IN- → OUT = 3.3V and stays there. Resets when IN- > 3.3V, i.e., +19VB > 13.5V. Not sure how that's supposed to happen if charging circuit is turned off.
At power-up, HW_ACAV_IN = 0 and +3.3V_ADP_DCIN = 3.3V so IN+ < IN-, so OUT = 0, until charger initializes then both HW_ACAV_IN and +19VB come up but stay in IN+ < IN- configuration. IN+ needs to come up after IN-, which is achieved by RC circuit PR606/PC603.
I hope I'm reading that right. To be honest I don't know what they are trying to achieve with that circuit.
Leave a comment:
-
ArchLinux, i2c-tools package, `decode-dimms` works fine. Using at24 driver for DDR3 (24-series I²C EEPROM) and ee1004 driver for DDR4 (34-series I²C EEPROM with JEDEC EE1004 spec) apparently.
24-series EEPROMs on DIMM sticks are often write-protected by pulling the WP pin high permanently on the PCB, so the IC may have to be desoldered anyway.
Leave a comment:
-
-
I know it's unusual around here, but some people may like to learn, these are good machines to start with (well, not dead GPUs but still).
Some people also do what they can to save more stuff from going to a landfill....
Leave a comment:
-
Boards for A1181 are usually pretty robust, shouldn't need to touch them much.
I'd suspect improper drying first, but maybe something got damaged.
ALL_SYS_PWRGD is a good checkpoint as always.
Board should turn on without battery if using correct Magsafe. Not 100% sure if that specific model turns on automatically when Magsafe is attached if there is no main battery pack and no RTC battery connected.
Regardless, it has power-on pads at R5010 to turn it on, as usual shorting SMC_ONOFF_L to ground.
Leave a comment:
-
Phases are switching at several hundred kHz. Look at the spec sheet of your multimeter. Need an oscilloscope.
Leave a comment:
-
As far as I know this model (and some others from Lenovo) have the solder joints weaken under the CPU due to the low-temperature solder (138C°) and the underfill material as well as chassis flexing. CPU has to be reballed with regular leaded solder (183°C) or even lead-free (220°C).
Leave a comment:
No activity results to display
Show More
Leave a comment: