[Acer 6930G] Lower +5V problem.

[ben7]

Re: [Acer 6930G] Lower +5V problem.

Sometimes it's better to address the effect than the cause. If the laptop works properly like that, leave it shorted. That MOSFET is just acting as a simple ON/OFF switch there.

Edit: You say the computer resets when trying to check MAIND. That indicates a problem there, it means that whatever is driving the MAIND signal has a very high output impedance... To drive the gate of a FET you would need a relatively low impedance source. Instead of shorting the FET you could do a more elegant solution: Pull the gate up via a 1k resistor to 5V_ALW. That should do the trick, and also have the circuit that's controlling MAIND work properly. There's a good chance that it's still able to drive the signal LOW when the computer is turned off.

It's weird because in the circuit, the mosfet gate is pulled up by a high resistance resistor :S

I guess some parts have changed their values slightly, and the high resistance resistor is no longer enough...

[Francisco.s]

Re: [Acer 6930G] Lower +5V problem.

Sometimes it's better to address the effect than the cause. If the laptop works properly like that, leave it shorted. That MOSFET is just acting as a simple ON/OFF switch there.

Edit: You say the computer resets when trying to check MAIND. That indicates a problem there, it means that whatever is driving the MAIND signal has a very high output impedance... To drive the gate of a FET you would need a relatively low impedance source. Instead of shorting the FET you could do a more elegant solution: Pull the gate up via a 1k resistor to 5V_ALW. That should do the trick, and also have the circuit that's controlling MAIND work properly. There's a good chance that it's still able to drive the signal LOW when the computer is turned off.

I agree with you is an elegant solution, and seems to be that will not be any issues.
edit: I am thinking that mosfet will not conduct with only 5 volts in its gate. There will be another way to do that, I will study the circuit, and post whatever I found.

[Th3_uN1Qu3]

Re: [Acer 6930G] Lower +5V problem.

edit: I am thinking that mosfet will not conduct with only 5 volts in its gate.

It will, don't worry. The MOSFETs designed for use in notebooks and other portable devices achieve minimum Rds(on) with gate voltages as low as 4.5v. Look up the datasheet of any of the 'fets you see on the schematic.

[Francisco.s]

Re: [Acer 6930G] Lower +5V problem.

It will, don't worry. The MOSFETs designed for use in notebooks and other portable devices achieve minimum Rds(on) with gate voltages as low as 4.5v. Look up the datasheet of any of the 'fets you see on the schematic.

Could be you are right, but the fact that they uses 15 volts to feed de gate makes me think that 5 volts is not enough. Perhaps the data-sheet will clarify this aspect. I'll check and post results.

[Francisco.s]

Re: [Acer 6930G] Lower +5V problem.

I have checked the data-sheet, and it confirms what I thought, mosfet FDS8884 needs a VGS = 4.5V in order to get 30mΩ of rDS(on) at a ID = 7.5A.
As we need 5 volts in the source of the mosfet, then we need 9.5 volts in order to do the mosfet fully saturated.
So it is difficult to get this voltage from the motherboard. Only from 15 volts circuit we can get it.


Perhaps will be a good idea to swap mosfets pq14 and pq34, so we can know if problem changes from the 5v to 3,3v.

[Th3_uN1Qu3]

Re: [Acer 6930G] Lower +5V problem.

That kinda flew over my head, i didn't realize that it's on the high side... Well then. There is a 15v rail, isn't it? Connect the 1k resistor from the gate to the 15v rail. Done.

[fddizon]

Re: [Acer 6930G] Lower +5V problem.

check/replace PC66 of MAIND. it might be the culprit.

[Francisco.s]

Re: [Acer 6930G] Lower +5V problem.

check/replace PC66 of MAIND. it might be the culprit.

Could be PC66 has some leakage, but as MAIND signal goes also to many mosfet, and those all works properly, we didn't have it into account.

[crowntoy99]

Re: [Acer 6930G] Lower +5V problem.

check/replace PC66 of MAIND. it might be the culprit.

thnk you