Vostro 3560 (LA-8241P, UMA version) runs only on AC

[spleenharvester]

Another LA-8241P here - not the usual shorted cap issue. (EDIT title is incorrect, board runs only on battery)


On AC: Completely dead, 19V into VIN but nothing on B+. Adaptor does not trip and there is no evidence of shorts. There are 40ohms ish across PQ100 (legs 1-3 to 5-8) see attached image for voltage readings.


On battery or if injecting 11V into B+: Board fires up with 2.8V at the power button, shuts down after 2 seconds (I suspect there's a second fault but we'll cross that bridge when we get there).


Annoyingly the other LA-8241P I'm working on are not currently with me so I have no other boards to compare to.


Any ideas?


Cheers

[spleenharvester]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

Update, there is 0.1V on pin 22 of PU100 (ISL88731CHRTZ-T). If I inject 19V into this pin, I get 19V on B+, so I suspect the IC is probably okay.

Which I think narrows it down to PC111 or PR109.

[khaahk]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

I admire that you put it in writing rather than checking both of them right away. I would not have the patience to do that...

[spleenharvester]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

The problem is that nothing on this board is screenprinted so I'm actually having serious trouble finding the components!

The previous post was actually a fairly large measurement error though, pin 22 is actually getting 19V from VIN, so the problem is elsewhere (what I thought was PC111 was actually PC135, so I was injecting at BST_CHGA).

[pyry]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

first test PC101 is not short and if not

remove PQ103 and test if it turns on

[khaahk]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

PQ100 is on my suspect list.

[pyry]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

never mind my fault i mean PQ105B it's not doing his job

[piernov]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

40 ohms across PQ100 seems too low to me. Check PQ100 and PQ101 out of circuit. (resitance between drain-source, drain-gate and source-gate)

[spleenharvester]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

PC101 confirmed not shorted. I will check PQ100 and PQ103 when my new hot air station gets here.

PQ105 is getting 2.5V on legs 2 and 5 (as per leg 1 PQ109) but 0V on legs 1, 3, 4, 6. Is this normal (I'm not understanding very well what this circuit is supposed to be doing?)

I suspect PQ100 might be okay as when 5V is applied across source to drain, it seems to open normally and not heat up too much, just like PQ101 does. But I will make sure to test it out of circuit.

[spleenharvester]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

Is it normal to get the battery voltage coming up at the DC jack when only battery is plugged in and no adapter?

I tried to take a resistance reading between PQ105 and VIN on a pin that is usually 0V and got a massive spark for reasons that are not clear to me. The board behaves as it did before but I never checked whether the batt voltage came up at DC before, I am just wondering whether I killed something.

[BlueMidnight]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

The input FETs (PQ100/PQ101) are turned on by allowing current to flow through PR101 and PR105 to ground. This is a voltage divider that will place ~8V on the FET gates. When the gates are at 19V, the FETs are biased off.

The gates normally rise to 19V if PQ103 is turned on (shorting the gates to 19V directly) and/or PQ105B is turned off (disconnecting the voltage divider from ground and raising all points to 19V as a result). In reality, the board will always toggle both of those transistors at the same time. PQ103 is important as it very firmly holds the FETs turned off by shorting the sources and gates together. PQ105B helps a bit as well by switching off, but also saves a tiny bit of power.

On your board, the voltages on PQ103 show it is biased off. Likewise, the voltages on PQ105B show it is biased on. So, there really shouldn't be 19V on the gates of the input FETs.

Therefore, there must be a failed component causing a short between VIN/P2 and the FET gates (assuming no conductive residue on the board, etc). The most likely candidates are PQ100, PQ101, PQ103, or PC101. And, of course, all these have been mentioned already. Resistors don't really fail shorted unless there are carbon marks around them from them burning up or some such, so those are probably safe to ignore unless you get truly desperate.

To get a feel for the proper operation of that circuit, I've attached two pictures. Each shows a different scenario. One is with a good AC adapter connected. The other is with an AC adapter that is outputting a voltage that is slightly too low. Several other scenarios are possible of course.

That spark you saw could only occur if the circuit was powered (either by being plugged in or by there still being capacitors charged in the circuit). You should never measure resistance on a powered circuit. Also, even when measuring voltage rather than resistance, sometimes it's possible to slip and bridge a couple pins with the multimeter probe. That will also cause a spark and possible damage to the circuit.

And regarding the battery voltage being present at the DC jack... that is not normal...

Motherboards typically have 2 input FETs. All FETs have a body diode that will pass current in one direction even when the FET is turned off. By placing 2 FETs back-to-back so that their body diodes face opposite directions, this blocks current flow both ways.

The FET with the body diode facing in such a way that it block the AC adapter current is called the adapter input FET (ACFET). The other FET blocks battery voltage from feeding back to the DC jack (and into the AC adapter if it is plugged into the jack but not the wall outlet) and is called the reverse blocking FET (RBFET). On your board PQ101 is the ACFET and PQ100 is the RBFET.

PQ100 being shorted would explain all the symptoms. So that would be the first component you should remove and test.

[mcplslg123]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

The input FETs (PQ100/PQ101) are turned on by allowing current to flow through PR101 and PR105 to ground. This is a voltage divider that will place ~8V on the FET gates. When the gates are at 19V, the FETs are biased off.

The gates normally rise to 19V if PQ103 is turned on (shorting the gates to 19V directly) and/or PQ105B is turned off (disconnecting the voltage divider from ground and raising all points to 19V as a result). In reality, the board will always toggle both of those transistors at the same time. PQ103 is important as it very firmly holds the FETs turned off by shorting the sources and gates together. PQ105B helps a bit as well by switching off, but also saves a tiny bit of power.

On your board, the voltages on PQ103 show it is biased off. Likewise, the voltages on PQ105B show it is biased on. So, there really shouldn't be 19V on the gates of the input FETs.

Therefore, there must be a failed component causing a short between VIN/P2 and the FET gates (assuming no conductive residue on the board, etc). The most likely candidates are PQ100, PQ101, PQ103, or PC101. And, of course, all these have been mentioned already. Resistors don't really fail shorted unless there are carbon marks around them from them burning up or some such, so those are probably safe to ignore unless you get truly desperate.

To get a feel for the proper operation of that circuit, I've attached two pictures. Each shows a different scenario. One is with a good AC adapter connected. The other is with an AC adapter that is outputting a voltage that is slightly too low. Several other scenarios are possible of course.

That spark you saw could only occur if the circuit was powered (either by being plugged in or by there still being capacitors charged in the circuit). You should never measure resistance on a powered circuit. Also, even when measuring voltage rather than resistance, sometimes it's possible to slip and bridge a couple pins with the multimeter probe. That will also cause a spark and possible damage to the circuit.

And regarding the battery voltage being present at the DC jack... that is not normal...

Motherboards typically have 2 input FETs. All FETs have a body diode that will pass current in one direction even when the FET is turned off. By placing 2 FETs back-to-back so that their body diodes face opposite directions, this blocks current flow both ways.

The FET with the body diode facing in such a way that it block the AC adapter current is called the adapter input FET (ACFET). The other FET blocks battery voltage from feeding back to the DC jack (and into the AC adapter if it is plugged into the jack but not the wall outlet) and is called the reverse blocking FET (RBFET). On your board PQ101 is the ACFET and PQ100 is the RBFET.

PQ100 being shorted would explain all the symptoms. So that would be the first component you should remove and test.

The best explanation a techie can make. Great

[spleenharvester]

Re: Vostro 3560 (LA-8241P, UMA version) runs only on AC

The input FETs (PQ100/PQ101) are turned on by allowing current to flow through PR101 and PR105 to ground. This is a voltage divider that will place ~8V on the FET gates. When the gates are at 19V, the FETs are biased off.

The gates normally rise to 19V if PQ103 is turned on (shorting the gates to 19V directly) and/or PQ105B is turned off (disconnecting the voltage divider from ground and raising all points to 19V as a result). In reality, the board will always toggle both of those transistors at the same time. PQ103 is important as it very firmly holds the FETs turned off by shorting the sources and gates together. PQ105B helps a bit as well by switching off, but also saves a tiny bit of power.

On your board, the voltages on PQ103 show it is biased off. Likewise, the voltages on PQ105B show it is biased on. So, there really shouldn't be 19V on the gates of the input FETs.

Therefore, there must be a failed component causing a short between VIN/P2 and the FET gates (assuming no conductive residue on the board, etc). The most likely candidates are PQ100, PQ101, PQ103, or PC101. And, of course, all these have been mentioned already. Resistors don't really fail shorted unless there are carbon marks around them from them burning up or some such, so those are probably safe to ignore unless you get truly desperate.

To get a feel for the proper operation of that circuit, I've attached two pictures. Each shows a different scenario. One is with a good AC adapter connected. The other is with an AC adapter that is outputting a voltage that is slightly too low. Several other scenarios are possible of course.

That spark you saw could only occur if the circuit was powered (either by being plugged in or by there still being capacitors charged in the circuit). You should never measure resistance on a powered circuit. Also, even when measuring voltage rather than resistance, sometimes it's possible to slip and bridge a couple pins with the multimeter probe. That will also cause a spark and possible damage to the circuit.

And regarding the battery voltage being present at the DC jack... that is not normal...

Motherboards typically have 2 input FETs. All FETs have a body diode that will pass current in one direction even when the FET is turned off. By placing 2 FETs back-to-back so that their body diodes face opposite directions, this blocks current flow both ways.

The FET with the body diode facing in such a way that it block the AC adapter current is called the adapter input FET (ACFET). The other FET blocks battery voltage from feeding back to the DC jack (and into the AC adapter if it is plugged into the jack but not the wall outlet) and is called the reverse blocking FET (RBFET). On your board PQ101 is the ACFET and PQ100 is the RBFET.

PQ100 being shorted would explain all the symptoms. So that would be the first component you should remove and test.

That is a superb explanation, thank you! I will test PQ100 as soon as my new equipment gets here.

Also I didn't know that about testing resistance while the circuit was powered - that would explain why my multimeter was freaking out a bit beforehand. I might also replace PQ105 while I'm at it just to be on the safe side.