Lenovo ThinkPad X250 strange power issue

[sarry86]

Hello, I have this laptop ThinkPad X250 that has this strange problem, when connected to the main power source, the fan spin once, keyboard lit up for like 2 secs then turns off, only power button remain with solid green light and the ThinkPad icon at the rear lit solid as well, I have tried everything, wiping, writing bioses that users here requested and were given as working for the exactly the specific model but nothing works, the strangest thing is that, even if I completely wipe the bios and blank check it to make sure it is wiped, the laptop still start as mentioned above, have anyone ever met with this strange phenomenon? 🤔 I wonder how Lenovo bios system works, if anyone know please enlighten me 🙏

[sarry86]

Could it be the problem with MEC1633L? 🤔

[mon2]

Could be assorted issues. Do you still have the original known good bios? It should be used for starters.

Review the voltage to ground on each pin of the charger IC. Post the full list of measurements. Proceed with care so that the charger IC is not damaged from shorting of the meter probes across nearby parts or pins.

[sarry86]

Could be assorted issues. Do you still have the original known good bios? It should be used for starters.

No I don't, just the one I backed up after the problem

Review the voltage to ground on each pin of the charger IC. Post the full list of measurements. Proceed with care so that the charger IC is not damaged from shorting of the meter probes across nearby parts or pins.

GND to :-
EXTPWR = 0V

LPMOD = 3.3V

ACDET = 2.48V

IOUT = 0.02V

SDA = 3.3V

SCL = 3.3V

ILIM = 1.5V

BM = 3.3

CSN = 1.47V

CSP = 1.47V

BTST = 5.90V

REGN = 6.00V

SW = 1.46V

LPREF = 0.5V

PVCC = 1.49V

[mon2]

1) cute kid

2) remove all power. Meter in resistance mode (not diode mode).

3) measure the resistance to ground of pvcc power rail. Post the measurement.

[sarry86]

1) cute kid

2) remove all power. Meter in resistance mode (not diode mode).

3) measure the resistance to ground of pvcc power rail. Post the measurement.

1. Thank you so much ☺
2. Reads 0ohms

[mon2]

The posted meter pic is helpful. This measurement is not 0 ohms since meter is reporting OL (‘over limit'). This means the selected meter scale is too low. Switch to a higher meter resistance scale till you have a proper measurement. We know that the resistance of this rail is higher than 200 ohms. Post the true resistance of this rail.

[sarry86]

The posted meter pic is helpful. This measurement is not 0 ohms since meter is reporting OL (‘over limit'). This means the selected meter scale is too low. Switch to a higher meter resistance scale till you have a proper measurement. We know that the resistance of this rail is higher than 200 ohms. Post the true resistance of this rail.

Here

Both in MegaOhms and KiloOhms range, I don't understand, why is it showing two different readings? And why are the caps charging in resistance mode? 😳

[sarry86]

The posted meter pic is helpful. This measurement is not 0 ohms since meter is reporting OL (‘over limit'). This means the selected meter scale is too low. Switch to a higher meter resistance scale till you have a proper measurement. We know that the resistance of this rail is higher than 200 ohms. Post the true resistance of this rail.

Please help me, is that resistance what is supposed to be?

[mon2]

The new shared resistance values are ok. They confirm there is no short on the tested rail.

Remove all power. Meter in resistance mode (not diode). Measure the resistance to ground of each inductor (usually will be grey in colour with 2 leads).

hunting for low resistance rails. Post the measurements.

[sarry86]

The new shared resistance values are ok. They confirm there is no short on the tested rail.

Remove all power. Meter in resistance mode (not diode). Measure the resistance to ground of each inductor (usually will be grey in colour with 2 leads).

hunting for low resistance rails. Post the measurements.

Sorry for the late reply, here

[mon2]

The measurements look ok. The coils with the lower resistance reading are to be expected as these are coils used for a low voltage rail.

the strangest thing is that, even if I completely wipe the bios and blank check it to make sure it is wiped, the laptop still start as mentioned above, have anyone ever met with this strange phenomenon?

There are 2 independent bioses at work on the logic board. The main bios which you are working with and is read by the host CPU (intel / AMD CPU). Then there is the EC / keyboard controller IC which contains an internal 8051 microcontroller. Respectively, this EC IC has a local bios (sometimes this bios is internal to the package of the IC) which is responsible to provide for simple tasks like keypress decoding. The keyboard is a grid of connections (horizontal / vertical lines) to save on the total # of connections. When you press a key, the 8051 processor will decode where the horizontal and vertical lines shorted to decode the key being pressed. Being a mechanical action, there will be 'electrical bounce' so the 8051 will clean up this 'dirty bouncing signal' to offer a clean logic '1' or logic '0' for the keypress and release action. Otherwise, you may see multiple and random # of keys when pressing only a single time on the keyboard. This same 8051 is also used to monitor your power on/off button so the main bios could be blank and these sideband signals and triggers will still function due to the 8051 controller.

[sarry86]

The measurements look ok. The coils with the lower resistance reading are to be expected as these are coils used for a low voltage rail.



There are 2 independent bioses at work on the logic board. The main bios which you are working with and is read by the host CPU (intel / AMD CPU). Then there is the EC / keyboard controller IC which contains an internal 8051 microcontroller. Respectively, this EC IC has a local bios (sometimes this bios is internal to the package of the IC) which is responsible to provide for simple tasks like keypress decoding. The keyboard is a grid of connections (horizontal / vertical lines) to save on the total # of connections. When you press a key, the 8051 processor will decode where the horizontal and vertical lines shorted to decode the key being pressed. Being a mechanical action, there will be 'electrical bounce' so the 8051 will clean up this 'dirty bouncing signal' to offer a clean logic '1' or logic '0' for the keypress and release action. Otherwise, you may see multiple and random # of keys when pressing only a single time on the keyboard. This same 8051 is also used to monitor your power on/off button so the main bios could be blank and these sideband signals and triggers will still function due to the 8051 controller.

So, this can be seen on the MEC1633L, is there an easy way to reprogram this chip by j-tagging it and use "any" programmer? 😳