Toshiba A665-14c

[BlueMidnight]

Re: Toshiba A665-14c

FETs can be difficult to measure accurately in circuit. Let's remember those measurements though. But keep in mind that PQ37 and PQ43 are in parallel, so you are measuring both of them at the same time.

We know the problem is someplace after PL11, but there are many possibilities still.

Desoldering things is still the only option you have available, so it's just about choosing what to remove. One possibility is to remove PQ37/PQ43 and put PL11 back. That will test all those capacitors that exist between PL11 and the FETs. It will also allow you to test the FETs outside of the circuit if you need to.

Test procedure for the FETs....

Source = Pins 1-3 (doesn't matter which)
Gate = Pin 4
Drain = Pins 5-8 (doesn't matter which)

Red/Black refers to the multimeter probe of that color.
Multimeter must be set to diode mode.
The expected reading is in parenthesis.

Steps:
1) Red on Drain, Black on Gate (No connection. This also turns the MOSFET off)
2) Red on Drain, Black on Source (No connection)
3) Swap the probe positions. Red on Source, Black on Drain (One diode drop, approximately 0.2-0.5V)

4) Red on Gate, Black on Drain (No connection. This also turns the MOSFET on)
5) Red on Source, Black on Drain (Less than 0.2V)
6) Swap the probe positions. Red on Drain, Black on Source (Less than 0.2V)

7) Repeat steps 1-3 to make sure MOSFET turns off again successfully

Be aware that some meters will not be able to turn on the FET properly. So this test isn't perfect. It should at least tell you if there is a short though.

[BlueMidnight]

Re: Toshiba A665-14c

You could also try removing/sliding each of the capacitors in that area (like the 6 brown/tan ones I see around PQ43 for example) and testing them individually.

Either way I would protect that plastic card reader that is nearby with some kapton tape and/or some layers of aluminum foil so it doesn't start to melt. By the way, I also do the same thing with nearby electrolytic capacitors (wrap them in a thick wad of aluminum foil to act as a heatsink) when I'm using hot air.

[Stevedb]

Re: Toshiba A665-14c

Ok midnight I will remove PQ37 and test it out the circuit

I will use your methods when all the pins Beebs . The mosfet is then bad. Right??

Best regards
Steve

[BlueMidnight]

Re: Toshiba A665-14c

The test doesn't always turn on the FET (depending on the multimeter being used). But you can see if it reads as shorted the whole time.

But remember that PQ43 is in parallel with PQ37. You were actually measuring both at the same time earlier. I would remove both of them if you are heating up one anyway. Then, if you reattach PL11, you can see if B+ still reaches 19V. If it doesn't read 19V anymore, then at least one of those capacitors are bad. It's a quick way of testing them all at once.

[Stevedb]

Re: Toshiba A665-14c

Hi blue midnight

I remove PQ37 and measured this all sides are shorted I measure no diode. I measured pq43 still in the board. But measures good with your methods. I put the coils back I got still 19,51V . I got back lights now white and red. I have only one problem I only find the mosfet on AliExpress or on eBay but from China. It will be long time it will be here. Abouth 34 days. Can I use another type for replacement.

Thanks for all the help.

Best regards.
Steve

[BlueMidnight]

Re: Toshiba A665-14c

There should be many other FETs you can choose from. Search for QFN8 or DFN5x6 N-Channel MOSFETs that you can order easily. Then check the datasheets for each one.


Based on the part number listed on the schematic, the replacement part should have:

Maximum Drain-source voltage "VDSS" of around 30V or higher.
Gate threshold voltage "Vth" or "VGS(th)" of 1.5V-2.5V.
Drain-source ON-resistance "RDS(ON)" of about 7-10 milliohms.
Drain current "ID" or "ID(ON)" of at least 24 amps.


Be aware... if the FET is shorted to the gate it is possible for it to damage PU13. If so, the new FET can be damaged after it is installed. Sometimes you have to replace both the FET and the controller IC at the same time. That's just something to think about. You might want to make sure you have both parts, or at least order extra FETs just in case.

[Stevedb]

Re: Toshiba A665-14c

Hi blue midnight thanks is it easey to replace the IC.
Or do I need special equipment for that.

Is the IC to order easily. Because when I look on eBay on Europe I do not find the IC also not on farnell. Do you know a good site who delivers to Belgium.?

Best regards
Steve

[BlueMidnight]

Re: Toshiba A665-14c

I highly recommend having some good flux to replace an IC like that. The gel kind. Not the watery stuff that comes in pens, and not the solid waxy stuff. It makes it much easier. Flux allows the solder to really grip onto the pins and pads. That force pulling it to the pins and pads also makes it not want to accidentally bridge neighboring pins as easily.

But there is a lot of cheap bad flux out there on eBay and such places. So, I'd recommend checking for some at a decent electronics company if possible. Either a local one or one of the global brands like DigiKey or Mouser

Clean the pads with solder wick. Add a little flux. Apply a very tiny amount of solder to each. Add flux before putting on the IC with hot air, and it will just center itself automatically on the pads when you get the solder hot enough. You can push it down against the PCB at that time (I prefer a stainless steel dental pick for moving ICs around) so it makes the best contact possible. But if you push it down, a little blob of solder might squirt out if there was too much on the pads. That's why you only want a tiny amount of solder on there.

Once it is attached, you can clean up the pins around the edges with a soldering iron (if necessary). A blade-shaped tip or a smaller, curved tip is best for getting good contact with the edges.

When searching for these parts, sometimes you have to leave off the letters at the end. I see some in the EU on eBay if I search for "ISL62883". I also see a "ISL62883C" but that isn't what you want because it has a lower max current. But if you are not in a rush, you could try just the FET first (order several at least... they work on many different boards anyway). There is a chance that the IC is okay. It's up to you.

[Stevedb]

Re: Toshiba A665-14c

Hi blue midnight

Thanks for the tutorial explaining.
I found TPCA 8050 h but looks a little different then TPCA8030H . I found also IRFH8334TRPBF and TPCA8057H expensive but looks the same are those good for replacing. Both on eBay. I will try first the MOSFETs better.

Best regards

Steve

[BlueMidnight]

Re: Toshiba A665-14c

Any of those will work. The ratings are all okay. And some have pins, while others have a larger body with pads instead of pins. But they all fit in the same place.

[Stevedb]

Re: Toshiba A665-14c

Hi midnight

I have today replaced the mosfet and the laptop is fixed.

Thanks for all the help.

Best regards

Steve

[Stevedb]

Re: Toshiba A665-14c

Hi bluemidniight

I couldn't without your help

Many thanks man

Best regards

Steve

[BlueMidnight]

Re: Toshiba A665-14c

Hey good job. That's awesome. Glad I could help

[Stevedb]

Re: Toshiba A665-14c

Yes you was a great help. Just the mosfet was bad. Ic was still good. I am very happy. Did you also read the other topic of my? I think the lapto not repairable anymore because somebody put the laptop in the oven what do you think abouth that topic.

Best regards

[Stevedb]

Re: Toshiba A665-14c

Hi midnight


Those leds i mean

Best regards

[BlueMidnight]

Re: Toshiba A665-14c

I can't think of any reason why the lights on the left would be dimmer than the ones on the right unless the LED board is misaligned in its slot. You can lift out the board to confirm whether it's the alignment or the lights themselves.

[Stevedb]

Re: Toshiba A665-14c

Hi midnight i did that are the leds self. When i remove the cover they are less bright. Strange i found that.

Best regards

[BlueMidnight]

Re: Toshiba A665-14c

The 5V rail is connected to the LEDs through the cable. The PCB has the LEDs and resistors. And then the return path is through the cable to pins on the EC (which are connected to ground to turn on each LED).

The LED brightness is determined by the type/health of the LED and the amount of current going through it. Nothing is impossible. But I can't think of a good reason why the LEDs would suddenly become worn out, or the resistors would increase in resistance, or the transistors in the EC would have a high resistance. The most likely scenarios in my mind are that the cable isn't making good contact, or maybe those LEDs were always like that, long before you ever worked on it.