Help Request for Lenovo L470 - No power

[nikal]

Hi

this is my first ever message in this forum that I follow for a very long time now and I have used the power of collective knowledge so many times.

I do repairs, not professionally (for the moment) and recently I stuck on a case with a L470 Lenovo laptop.

This laptop came to me with no power issue. I diagnosed it and I found the following.

1. A shorted and exploded ceramic capacitor (probably on main power rail)
2. Internally shorted the 3.3 ALWP buck converter
3. Shorted (drain to source to gate) the first DC-IN mosfet.

I have replaced the Mosfet and removed the exploded capacitor. Currently The 3.3 buck IC is on order from China. The problem is that despite the removal of all shorted components, the BQ IC doesn't switch ON the first MOSFETs no matter what. BQ has VCC and REGN and ACDET 2.6V however ACOK is low (0V). Is 3.3ALWP a prerequisite for BQ to startup? Given that EC doesn't get power and SMBUS isn't working.


Any advice is more than welcome

Kind Regards

[mcplslg123]

The 3.3V regulator will generate 3.3V LDO which in turn will power the SIO. So it wont work without the 3.3V Regulator soldered.

[nikal]

The 3.3V regulator will generate 3.3V LDO which in turn will power the SIO. So it wont work without the 3.3V Regulator soldered.

Hi,

thanks for your reply. So, the BQ needs to "talk" with the EC in order to switch ON the dc-in FETs as well the charging circuit. Is it so ?

Thanks in advance for your reply and time

[mon2]

The BQ charger IC has a SMBUS (I2C) interface which use the SCL & SDA lines. Respectively, on this interface, there will usually be a single bus master (sometimes more than one) and at least one or more SMBUS slave devices. This BQ charger IC is the slave device and the EC is the SMBUS master that initiates the communication across this 2-wire interface. The SMBUS master creates the SCL (clock line) and 'pings' all expected devices on the network of slaves. Only if the SMBUS address matches, only then the SMBUS slave must reply back to the master. Review the BQ charger IC datasheet to learn about the address supported by this component.

[nikal]

Hi mon2

Thank you for the reply. Yes I know how SMBUS works. Well, the reason I'm asking about BQ is because I did a test in order to see if the only fault is the 3.3V buck converter. I provided power to the laptop DC_IN using the external charger/adapter and then using my bench power supply I supplied 3.3V to the coil. I was expecting the EC to start the laptop and the BQ to open the DCIN FETs but nothing happened. I have checked for other shorts in the board but found none. Ι wonder if the controller IC provides some signal to EC in order to start. If this is the case, then my bench power supply can't simulate the missing IC by just providing the necessary voltage.

Anyway, I guess I will find out as soon as I get the IC and solder it back in place.

Thanks again for your time and help.

[mon2]

Confirm the resistance of the dcin mosfets - with no power. Each should not be shorted and the current loop circuits must be stable with values matching the schematics else the gate voltage will not be boosted. Best to share the full list of voltage to ground on each pin (with function names) of the BQ charger IC. Do this only after the confirmation of the dcin mosfets. The mosfets must be hundreds of k ohms or more if not shorted.

[nikal]

Confirm the resistance of the dcin mosfets - with no power. Each should not be shorted and the current loop circuits must be stable with values matching the schematics else the gate voltage will not be boosted. Best to share the full list of voltage to ground on each pin (with function names) of the BQ charger IC. Do this only after the confirmation of the dcin mosfets. The mosfets must be hundreds of k ohms or more if not shorted.

Hi Mon2

The resistances in the two DCIN FETs are in the range of several hundrends or MOhms either I measure compared to ground or between source-drain-gate.I think that except the pins I mentioned in my first post the rest of the BQ have no voltages or very low voltages (less than 1V) but I will check again and share here the exact values per pin.

Kind regards

[nikal]

What I noticed though measuring again the voltages and looking at the schematic is that the pin 5 (3V_LDO) of the burned IC in the 3VALW circuit becomes 3VL which then powers the EC. So for sure, injecting 3.3V at the coil won't power the EC which in turn is needed for the BQ to startup.
I will wait for the IC and after I replace it I will check everything again and if there is still an issue I will get back here.

Thanks again for your time and help.

[mcplslg123]

If you really want to experiment,then you need to inject 3.3V on pin5(LDO) and not the coil. may work but not sure.
However, wont suggest to experiment if you donr have deep knowledge of laptop circuits.

[nikal]

If you really want to experiment,then you need to inject 3.3V on pin5(LDO) and not the coil. may work but not sure.
However, wont suggest to experiment if you donr have deep knowledge of laptop circuits.

Hi!

thanks for the reply,

I already did it and I saw that BQ restored voltages in many of its pins but there is no gate voltage in the DC-IN FETs. Maybe the EC is burned too but I will not give up yet. I will wait the new buck converter controller and a new BQ which I ordered too.

[Sametbey]

If there is no problem with the Scl and Data pins of the BQ charge control chip, the input mosfets need to be opened. After the 19v is distributed, the LDO receives +3v and then the IO control. Then the charging should start.

[nikal]

If there is no problem with the Scl and Data pins of the BQ charge control chip, the input mosfets need to be opened. After the 19v is distributed, the LDO receives +3v and then the IO control. Then the charging should start.

Thanks for your reply!

If there is no problem with the Scl and Data pins of the BQ

or with the EC, right ?

To be honest, I already replaced the BQ with a spare from another board. The thing is that both BQ (the original that was onboard and the second one that I replaced) produce the REGN voltage of 6V meaning that the internal LDO is working. This is a sign that BQ is OK so something else is going on and I suspect the EC. The thing is that I can't find any other short on the board so the only culprit is the EC but I will replace the 3.3VALW IC in order to be sure.

What do you think ?

Kind Regards

[Sametbey]

In some cases, it may enter protection due to the mosfet on the +19v line in the other coil circuit. You can check. If you bypass the AC input mosfet in the conduction direction -> (1N4001) you can observe the +3V LDO output.

[nikal]

Sametbey

Thanks for the help neighbor 😉

By bypassing you mean to inject 19V after the two DCIN mosfets and the protection diode? To check for +3V LDO/ALW I have to replace first the burned buck IC. Anyway, I will do further test as soon as I receive the IC and replace it in the 3VALW circuit cause my bench power supply doesn't provide multiple outputs in order to simulate DCIN + 3VLDO + 3VALW and observe consumption at the same time.

[mcplslg123]

Once again,I'll request to have patience and wait for the 3V regulator ic before going in for un-necessary adventure at this stage . Why risk damage?

[nikal]

Once again,I'll request to have patience and wait for the 3V regulator ic before going in for un-necessary adventure at this stage . Why risk damage?

Agree mostly because I don't see any progress not matter what I do. I'll replace the 3V regulator and we will take it from this point again.

Thanks for the advice.

[nikal]

Hi again,

After replacing the 3.3VALW and 3.3 LDO rail restored, I found that EC was burned. I had the same EC (IT8586E) from another board (yes I know it needs specific firmware in order to work) and I replaced it just to see if something will be different. Unfortunately the same behavior. The first MOSFET after the DC-IN won't open (no gate voltage). I also replaced the BQ (24780) but this didn't change anything. The VCC power reached the BQ, I also have the REGN voltage and the ACDET (2.6V). If I bypass the first two mosfets and provide voltage to the board after the second mosfet I see that something draws around 200mA for a fraction of a second and then consumption goes down to around 10mA (meaning the EC has power). I couldn't find any short anywhere and I'm suspecting that maybe I have to program the EC with the correct firmware in order for the BQ to open the first MOSFET. I also tried to bypass the SMBUS connection of BQ with the EC by shorting to ground the DATA pin but no change. The board won't get power no matter what.

Any ideas ? Do I have to program the EC with the correct firmware or this is not related to the BQ and the absence of gate voltage to the first MOSFET ?

Thanks in advance for your help

[mon2]

Carefully measure and post each voltage on the BQ charger IC in a table format with the pin # and their functions for a review.

With no power, what is the resistance to ground (in ohms) of the main power rail that follows the DCin mosfets?

[nikal]

Carefully measure and post each voltage on the BQ charger IC in a table format with the pin # and their functions for a review.

With no power, what is the resistance to ground (in ohms) of the main power rail that follows the DCin mosfets?

Hi,

thanks for the reply. Please see the attached image.

The resistance of the main power rail after the two first FETs is starting from several Mohm and stabilizes after a while at around 100Kohm (I guess as soon as the capacitors on the main rail are charged up).

[mon2]

The voltage on CMSRC pin # 3 cannot be at 0v for the charger to function. It must have the adapter voltage as in a normal power path flow onto the main rail. Absent voltage here will make the charger IC believe the adapter voltage is missing.