NP300e5c Killing input FET after some time.

[CidiRome]

Hi there.

I'm wrapping my head around this Samsung that keeps killing Q10 (the first FET in the 19V power line) after a few minutes at least 15 (this last time I was not close to it when it went down).

I've seen some youtube videos where they just short this FET, but I don't like this kind of approach: if the FET is there it is for some reason that I may or may not even understand.

As I didn't find the same exact model FET to replace, I've scavenged other boards for FETs with similar values, but they keep failing.

There must be something down the line that is killing the FET after a while.

Note that the FET doesn't get burned, it just stops working and sometimes it even work for a few seconds and stops after.

Does any of you that are experienced in this things can point me to the reason of the FET being killed?

Cheers.

[ktmmotocross]

Re: NP300e5c Killing input FET after some time.

check what is current draw from psu and replace that mosfet with some with higher amperage

[piernov]

Re: NP300e5c Killing input FET after some time.

What do you mean by "stops working"? Is it shorted or not? Usually the failure mode of the DC in MOSFET is short between source and drain. If it's not shorted the issue is probably something else, such as the charger IC not triggering the gate.

[mcplslg123]

Re: NP300e5c Killing input FET after some time.

Neither you mentioned the original fet used on the mobo nor you mentioned the replacement fets used. So how do anyone know, whether they were compatible or not??

[CidiRome]

Re: NP300e5c Killing input FET after some time.

Hi.

I will try to answer all questions, so hopefully you can help me.

- I've replaced it again and while working it was drawing up to about 1.5A with the battery out and up to about 2.7A with the battery in (charging).
- Stops working: they do no short, after what I consider failing they sometimes allow the notebook to turn on, but mostly the greed led turn on and off again. It is worse with the battery in (at least discharged like it is now).
- I've tried a few FETs that I didn't give much attention to the Amps because I knew that this notebook is low end/power.

The original FET was a 7200, the first replacement was B20N03, the one after I don't remember, and the last one was a different package type that I manage to install for tests is a 7170.

With this last one I hooked the notebook to my bench power supply and it was working for hours (and charged the battery) and I left over night and today it don't work anymore.

The strange thing is that the schematic seems to indicate different references for the FETs: the one that I replaced was originally a 7200, but the schematic says it is a 7403 that is even a different type of FET...

Please see the attached pictures.

Any Ideas?

Cheers.

[anshad]

Re: NP300e5c Killing input FET after some time.

''- Stops working: they do no short, after what I consider failing they sometimes allow the notebook to turn on, but mostly the greed led turn on and off again. It is worse with the battery in (at least discharged like it is now).''


Based on your explanation, i don't think that the MOSFET is failing/faulty. It looks like the MOSFET driving circuitry sometime is failing to drive the MOSFET. In other words, the MOSFET is not receiving proper gate drive voltage sometimes. What you should do is measure the gate voltage when the MOSFET is passing the voltage and also when it is not passing the voltage. Then you will be easily able to determine the exact root cause.

Btw, what is the board part no ?. Usually Samsung boards are identified by ''BA41-XXXXA'' format.

[CidiRome]

Re: NP300e5c Killing input FET after some time.

Hi.

The Board P/N is BA41-01978A (second Picture of my last post).

I will try to measure the gate voltage and will get back to you.

Cheers.

[CidiRome]

Re: NP300e5c Killing input FET after some time.

Hi.

I had a few seconds to measure the voltage before it went off (with the same FET) and it is about 25V relative to ground, that would be about 6V relative to Drain.

I remember measuring the voltage Gate to Drain and it was about 5V.

When it goes off the voltage is absolute zero (to Ground) or -17V to Drain.

I thought that it shutting down would be because something ordered it to, but the gate circuitry seems to be very passive... The thing is that I don't know where it is on the board and the also the schematic has P-Channel FETs while there are N-Channel FETs on the board.

Any ideas?

[piernov]

Re: NP300e5c Killing input FET after some time.

I can't find the schematics for this board so maybe you got a different one.
25V to ground is indeed perfectly fine for N-Channel MOSFET.
0V means the charger IC stopped generating the gate signal. The charger IC is what controls the gate of the DC-in MOSFETs. So if voltage is straight up 0, it can be an issue around the charging circuit in general.
First you have to identify what kind of charger IC you have on the board.

[CidiRome]

Re: NP300e5c Killing input FET after some time.

Hi.

The schematic I have is the one that I'm attaching and I downloaded it from this forum. Based on the P/N BA41-#####A written on it, it matches BA41-01978A written on the board itself.

Anyway, if this FET is controlled by the charger IC, it doesn't match with the schematic that has a simple passive circuit for this FET's gate (considering it could be a bit different for a N-Channel FET present in this board).

And is there any reason for this notebook to work flawlessly for a few hours when I replace the FET and stop working after that? This is why I would think that something is wrong with the FET itself and not with other circuitry, at least to the level of something making the FET stop working instead of something stop driving the FET...

I'm also attaching pictures of the power/charging circuitry.

Cheers.

[piernov]

Re: NP300e5c Killing input FET after some time.

Schematics is for an AMD platform called Petronas-15, I think your board is an Intel platform called Scala3-15.

Your charging circuit is different. You have a bq24735 charger IC. The gate of the MOSFET is driven by the ACDRV pin of this charging IC. It has an internal charge pump to get the gate voltage 6V higher than source voltage, so about 25V.

If the MOSFETs are not shorted drain to source it's unlikely they are the culprit. Not that they don't have other failure mode, but it would be very unusual in this circuit, especially multiple times. Maybe heat is doing something.

[CidiRome]

Re: NP300e5c Killing input FET after some time.

Hi.

This is the second time I come across a board with this BQ735 and the previous case I couldn't find a dooner board in my stock where I could get this chip from (I found BQ725, but not one 735).

This time I was not pointing to this chip, but it would be nice to know it is the culprit before ordering a new one.

I'm not inclined to do the universal repair of shorting the FET SOURCE-DRAIN pads with a wire like I've seen doing in a few youtube videos for this notebook model...

If there are some more ideas, please feel free to tell what you (plural, any of you) think.

Cheers.

[SMDFlea]

Re: NP300e5c Killing input FET after some time.

scala3-14 schematic ,same BQ735 .

BA41-02093A(HST) BA41-02092A(GCE) BA41-02094A(PTC) BA41-02098A(GBM)

[anshad]

Re: NP300e5c Killing input FET after some time.

''And is there any reason for this notebook to work flawlessly for a few hours when I replace the FET and stop working after that? ''

When you replace a SMD component, you have to apply fair amount of heat using SMD rework station. This heat can have an effect on nearby components and it is very common for some intermittently failing component to either break total or work fine for sometimes when heat is applied. For decades, electronic technicians are using heat and cooling spray to find intermittently failing components.

Refer page 40 of the above schematic for a matching charging circuit using BQ24735. Measure voltages of all pins.

[anshad]

Re: NP300e5c Killing input FET after some time.

''I'm not inclined to do the universal repair of shorting the FET SOURCE-DRAIN pads with a wire like I've seen doing in a few youtube videos for this notebook model...''


Go for this route only as a last resort. If you bypass the VIN protection MOSFETs, some boards may refuse to charge the battery pack. In fact most of the boards i have checked so far behaves like that. Yours might be different but only way to know is actually trying.

[petehall347]

Re: NP300e5c Killing input FET after some time.

get the fet to switch on like its supposed to . why cheat and chance destroying something else ? ok if low value unit i suppose and looking for a new one anyway .

[CidiRome]

Re: NP300e5c Killing input FET after some time.

Hi.

With the correct schematic I was able to diagnose a bit more.

The two resistors (marked in the picture I'm attaching) compose a voltage divider to sense then input DC in. What I noticed was the the bigger (in ohms) resistor resistance was floating, normally going down in resistance along with the temperature going down.

If this resistor resistance goes down causes the voltage to go up and what I believe happens is that the IC detects that the input voltage is climbing and shuts down the FET.

I don't know if I'm 100% correct, but I replaced the resistor with one of similar value and the notebook has been working flawlessly since then.

Thank you for your help.

Cheers.

[ktmmotocross]

Re: NP300e5c Killing input FET after some time.

ic needs to have voltage from resistor divider to ACDET pin in some level/around 2.7V/ if it is more or lover than this level, it shuts down fets./see datasheet/

on battery work normal

[anshad]

Re: NP300e5c Killing input FET after some time.

Hi.

With the correct schematic I was able to diagnose a bit more.

The two resistors (marked in the picture I'm attaching) compose a voltage divider to sense then input DC in. What I noticed was the the bigger (in ohms) resistor resistance was floating, normally going down in resistance along with the temperature going down.

If this resistor resistance goes down causes the voltage to go up and what I believe happens is that the IC detects that the input voltage is climbing and shuts down the FET.

I don't know if I'm 100% correct, but I replaced the resistor with one of similar value and the notebook has been working flawlessly since then.

Thank you for your help.

Cheers.

Good job and a proper repair work. Imagine what will happen if you used the jumper bypass short cut instead. The adapter volt will pass but as far as the charging IC thinks that the AC input voltage is insufficient, it will refuse to charge the battery. The laptop may essentially works like a portable desktop PC without battery backup. Don't blindly swallow everything see on YouTube.

[REpairAT]

Re: NP300e5c Killing input FET after some time.

I just had this particular notebook here for repair, finally fixed (For everyone else having the same- or similar fault).

In my case it was a BIOS issue (or more likely EC FW, that comes with the update).
Symptoms were, that the notebook only ran on batty, gate voltage was too low on input MOSFETs.
so what I did:

• Bridge gate to enable input MOSFETs and charge the battery
• Update BIOS to latest version (attached, original from Samsung Update)
• Remove bridge

suddenly, gate voltage is correct and notebook runs on AC and Battery again.

Bios - https://www.badcaps.net/forum/showthread.php?t=37276

Hope I could help. Cheers!