Question Can this laptop be repaired ?

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Apr 19, 2020
32
2
35
Hi everyone,

Some month ago, one of my friend probably killed my HP Envy 13 (ad-003nf) by causing an electrical overload to one of its USB port. When it happened, the screen turned off immediatly, and the computer never started again. I bought a new computer since, but as it is a 1300e laptop, I can't just throw it away without trying to repair it.

When I plug the power on, the LED on the start button turns on, but nothing happen, and the Charging Led remains turned off.

I tried to plug an external HDD to an USB C port which can charge an external device even when the computer is not running, and the LED on the HDD do turns on.

I guess the motherboard is dead... Can somebody confirm that ?

Thank you for your response !

Have a great day
 
Usb ports are fine and not shorted we tested that already, also hes getting 3.3v to bios chip and power button. So might think bios got corrupted, i had units getting bios corrupt after mosfets failing and replacing them, had to replace bios chip or reflash.

edit: im not sure when the programmer arrives he ordered it, i just dont know how to flash with soic clip so cant help with that.
 
Usb ports are fine and not shorted we tested that already,
Good but not great, if he fed higher voltage (>5v), this could cause many troubles to everything else that sits on the 5v bus... Let's figure out what happened first :)

that can also hes getting 3.3v to bios chip and power button. So might think bios got corrupted, i had units getting bios corrupt after mosfets failing and replacing them, had to replace bios chip or reflash.
That means the 3.0v power phase is operational and starts... Not much but better than nothing.
For flashing, before even trying to reflash it, we need to make sure it isn't hooked up to anything that will mess with the flasher. Need to measure resistance between Pin #4 and Pins #1,2,3,8,7,5... (Datasheet, Section 3.1, Figure 1a). All should read a high impedance and reasonable resistance on Vcc and /CS.

edit: im not sure when the programmer arrives he ordered it, i just dont know how to flash with soic clip so cant help with that.

CH341a (at least the one I have) uses 5v output and needs a level shifter to meet the IC spec (Datasheet, Section 8.1, 8.2). Alternatively, there is a mod to it, to operate at 3.3v (which is within the spec). I did not test the mod myself, but it looks safe.

As to the troubleshooting itself, I would start with checking the other power phases for shorts (you have 3.0v so far), and then (if none) for appropriate voltages, see the image below. There may be more below the cover.
From there on to check if the POST card gives any clue... if silent - checking the SMC and so on.

 
Last edited:
Apr 19, 2020
32
2
35
It was 1 year ago so I do not remember every details. From what I remember, there was an arduino module connected on one side to a DC motor and on the other side to a USB 2.0 port of my laptop. He made all the connection, and I had loaned him my computer. When he powered the Motor with 220V, my computer turned down and had never turned on since. There must have been a short somewhere, but the strangest thing is that either the arduino nor the motor were killed, so I don't think that 220V reached my computer.

For the test of resistance, between pin 4 and the others, here are the results:

1: infinite
2: infinite
3: infinite
5: infinite
6: infinite
7: infinite
8: 1200 ohms

I made the test on the 2000 ohms mode, so infinite just means >2000 ohms. I suspect the 200k Ohms mode doesn't work on my multimeter, values are weird: on pins 3, 7 and 8, value on this mode rise from 10 to 30k Ohms... ( even on the pin 8 where the resistance is 1200 ohms)

I hope I am clear, tell me if something is not

Here are the tension I got on these components : https://ibb.co/c8WjFLc

Have a great day
 
For the test of resistance, between pin 4 and the others, here are the results:

1: infinite
2: infinite
3: infinite
5: infinite
6: infinite
7: infinite
8: 1200 ohms

I made the test on the 2000 ohms mode, so infinite just means >2000 ohms. I suspect the 200k Ohms mode doesn't work on my multimeter, values are weird: on pins 3, 7 and 8, value on this mode rise from 10 to 30k Ohms... ( even on the pin 8 where the resistance is 1200 ohms)

The above means the BIOS chip can be flashed without de-soldering. It is safe to try anyway.

Here are the tension I got on these components : https://ibb.co/c8WjFLc

Ok. I believe you skipped checking resistances in those points first. There might be short between one of those and the ground and it is a good idea to check it before applying power (in general).
So far, something does not start on the ones with 0v. Those chokes are usually a part of the voltage regulation circuitry, but the problem might be on the "load" side or in the start-up sequence logic too.
Are there any other chokes like those on the board and underneath the protective film?

  1. Are you friends with a soldering iron? SMD/BGP reworking?
  2. Do you have access to a thermal imaging camera? (alcohol is a good workaround when something is obviously burning, but it is hard to spot 5C temp difference with it while rising the voltage to raise the heat is going to fry more components, while using thermal imaging makes it a breeze.)
  3. Do you have access to an oscilloscope?
  4. Lab PSU?
Component-level reverse engineering on multilayer circuitry without schematics is not an easy task. If one of the answers to 1/4 is a "No", your best bet is taking it to a tech who deals with electronic repairs, so he can have a better look into it.

Have a great day
You too :)
 
The above means the BIOS chip can be flashed without de-soldering. It is safe to try anyway.



Ok. I believe you skipped checking resistances in those points first. There might be short between one of those and the ground and it is a good idea to check it before applying power (in general).
So far, something does not start on the ones with 0v. Those chokes are usually a part of the voltage regulation circuitry, but the problem might be on the "load" side or in the start-up sequence logic too.
Are there any other chokes like those on the board and underneath the protective film?

  1. Are you friends with a soldering iron? SMD/BGP reworking?
  2. Do you have access to a thermal imaging camera? (alcohol is a good workaround when something is obviously burning, but it is hard to spot 5C temp difference with it while rising the voltage to raise the heat is going to fry more components, while using thermal imaging makes it a breeze.)
  3. Do you have access to an oscilloscope?
  4. Lab PSU?
Component-level reverse engineering on multilayer circuitry without schematics is not an easy task. If one of the answers to 1/4 is a "No", your best bet is taking it to a tech who deals with electronic repairs, so he can have a better look into it.


You too :)


Its so much easier to find shorts when you have power supply and can inject voltage any point you want on the board. Then just use hand to feel which "area" gets hot and freeze spray / isoprophyl alcohol to identify it more.

About the ch341a programmer he ordered has a 3.3/5v output jumper : https://www.ebay.com/itm/SOIC8-SOP8-Test-Clip-CH341A-Flash-BIOS-USB-Programmer-fur-EEPROM-25CXX-24CXX/114031747365?

I also have old one which has only 5volts, actually it reads 5.5-6v on the multimeter
 
Its so much easier to find shorts when you have power supply and can inject voltage any point you want on the board. Then just use hand to feel which "area" gets hot and freeze spray / isoprophyl alcohol to identify it more.
Everyone has his own technique :)

About the ch341a programmer he ordered has a 3.3/5v output jumper : https://www.ebay.com/itm/SOIC8-SOP8-Test-Clip-CH341A-Flash-BIOS-USB-Programmer-fur-EEPROM-25CXX-24CXX/114031747365?

I also have old one which has only 5volts, actually it reads 5.5-6v on the multimeter
I have two of those (bought a couple to help that other guy in the motherboard thread... and there is a jumper too, it only controls the Vcc probably). I made the mod (pin28 off the pad and to 3.3v regulator) on one of them to test what it does. It indeed helps.
I tested pins 1-4 (/CS - GND) on the one w/o the mod (bottom ch341a and left DMM) and on the modded one (also added a USB tail to it for convenience).
5Hi5MwJ.jpg
 
Everyone has his own technique :)


I have two of those (bought a couple to help that other guy in the motherboard thread... and there is a jumper too, it only controls the Vcc probably). I made the mod (pin28 off the pad and to 3.3v regulator) on one of them to test what it does. It indeed helps.
I tested pins 1-4 (/CS - GND) on the one w/o the mod (bottom ch341a and left DMM) and on the modded one (also added a USB tail to it for convenience).
5Hi5MwJ.jpg


You went few steps further than me, nice mod. +1 for using Fluke multimeter i have one too :)
 
  • Like
Reactions: vov4ik_il
You went few steps further than me, nice mod. +1 for using Fluke multimeter i have one too
:) the mod credit goes to its owner, I did not really dig into it, just had one spare piece to try it. He is a Fluke owner too :tearsofjoy:
The level shifters do the same thing too, with no soldering required.

As for the OP, if you want to use ch341a to fash this BIOS, make sure to do the mod or get a level shifter, so that it does not fry anything... But it looks like you need to figure out what happened to the voltage regulating circuitry on this board anyway. 0v is not normal in most cases.
 
Apr 19, 2020
32
2
35
Unfortunatly I only have a multimeter right now, and none of what you said.

So I believe I am going to try to reflash the BIOS, and if it does not work, I might just by a whole motherboard. I found this site that offers the same model of motherboard : https://sps24.eu/product-eng-3677-M...041TX-2GD44PA-DSC-MX150-2GB-i5-7200U-8GB.html

I never bought spare components before and this one is quite expensive, is it a good idea ?

Thanks
 
Apr 19, 2020
32
2
35
Hi,

I checked for 3.3v and 5v on several components, they remain steady when I press the button. I could not find any components where there was actually 19.5v, apart from the pins where the Power cable is plugged. When I press the button, this remain at 19.5V too.

I can't follow the 19.5V line, as there is too much components near it...

On this image, with power cable plugged : https://ibb.co/sysQYDV

  • Green components are alimented with either 3.3 or 5V
  • Red ones do not show any electric tension on either pins. Is it normal ?

I received the programmer ! But to be honest I do not know how to use it... I am going to search a method on internet, but do you have a link where everything is well explained ?

Thanks !
 
it might be dead startup chip (ENE or ITE chip) can you find this on the board its a big one
You might be right. Looks like the controller is not running the board startup. But you might be wrong - it might be a burned input fuse or something :)

@MaximeGeorges , please make a good close image of the IC to the left of the choke that marked “R10”. Lets see what kind of animal this is. Do not worry about flashing for now. We have a flashing guide but it is irrelevant for now.
 
Apr 19, 2020
32
2
35
Before pressing :
  • VIN_SEN (47) : 0V
  • VDD33 (44) : 2V
  • VDD18 (26) : 1V
  • EN (37) : 0v
  • PE (36) : 2V
  • SLP_S0# (35) : 2V
After pressing, the values remain the same. The pins are very clothes to each other, so especially for the 36 and 35, i might have touch both of them at once, but I was very carefull not to, so I dont think so.

Plus, my multimeter does only have a 300V caliber for continuous signals, so my accuracy is only of 1V, which is not very precised... I could have another one, but in a few days. So 2v might actually be 3... or 2.9
 
multimeter does only have a 300V caliber for continuous signals, so my accuracy is only of 1V, which is not very precised... I could have another one, but in a few days. So 2v might actually be 3... or 2.9
Thats not enough. Will wait.

Also be careful not to short any legs when you do that, use a sharp probe - you can easily do more damage.

So far it looks like the board is sleeping (0v @ EN). We need to find the KBC on the board. Take more photographs of similar sized ICs.
 
Ok, I will get a new one at the beginning of next week normally, and redo the tests.

I found this chip, which should be what we are looking for. But there are no pins reachable from what I can see. But I can see the lines between the chip and the resistances around, so maybe I can check voltage on the resistances ?

https://drive.google.com/open?id=1GsorOtbncF0BaFnqa8BfIOzEPiSUw9Kz

Thanks
I have trouble finding a datasheet for this one (it8988vg). This is the bug that is responsible for launching startup sequence upon power button pressing (among other things it does). Without pinout there is not much I can do. Maybe @lga1156_ftw has it?
The only thing that you can do without pinout is find out (using ohm-meter) which of the gpio legs is connected to the power button. Usually one of the power button legs is grounded and the other one is connected to gpio via pull up resistor bridge or the other way around. Everything has to be powerless for that matter.
 
Apr 19, 2020
32
2
35
Hi, I do have a more precise multimeter now. I rechecked impedances of the BIOS chip, between pin 4 and the others. Here are the results on the display:

1 : 5 Mega ohms
2 : 5 Mega ohms
3 : 68 kOhms
5 : 5 Mega ohms
6: 5 Mega ohms
7 : 16 kOhms
8 : 14 kOhms


(Datasheet) : https://www.winbond.com/resource-files/w25q64fv revq 06142016.pdf


For the other chip :
  • VIN_SEN (47) : 1.2V
  • VDD33 (44) : 3.2V
  • VDD18 (26) : 1.8V
  • EN (37) : 0v
  • PE (36) : 3.2V
  • SLP_S0# (35) : 3.2V
After having pressed the buttons, values remain the same.

I did not find any continuity between the "not groud pin" of the power button and any pins of the KBC. I will recheck later to be sure...
 
Hi, I do have a more precise multimeter now. I rechecked impedances of the BIOS chip, between pin 4 and the others. Here are the results on the display:

1 : 5 Mega ohms
2 : 5 Mega ohms
3 : 68 kOhms
5 : 5 Mega ohms
6: 5 Mega ohms
7 : 16 kOhms
8 : 14 kOhms


(Datasheet) : https://www.winbond.com/resource-files/w25q64fv revq 06142016.pdf

You can flash the BIOS IC with a clip, just need to make the mod to that device you bought... But so far this does not seem to be the problem.

For the other chip :
  • VIN_SEN (47) : 1.2V
  • VDD33 (44) : 3.2V
  • VDD18 (26) : 1.8V
  • EN (37) : 0v
  • PE (36) : 3.2V
  • SLP_S0# (35) : 3.2V
After having pressed the buttons, values remain the same.
This IC has to be "enabled" to start the other voltage regulating circuits behind it. The signal might have to come from the ITE IC. You can ring EN leg to all the resistors around KBC and you will probably find one. It does not have to be 0Ω, could be 1kΩ, or even 10kΩ as there is usually a current limiting resistor and pull-up or pull-down resistor on those GPIOs to avoid fluctuation. If you have a laboratory power supply, you can inject 3.3v to this leg and check the other voltages on those chokes that were reading 0, they should come up. But that will not make the board boot.

I did not find any continuity between the "not groud pin" of the power button and any pins of the KBC. I will recheck later to be sure...
Same as above, there should be a resistor. Ring on the outside of the resistors. The KBC seems to have 128 pins, some will probably be between the layers or on the bottom of the board. I looked up other ITE 128 pin KBCs (1, 2) and the power button may possibly be pin 18 or 21, but it does not necessarily have to be the case here. You can try waking it up by checking what does it have on sleep (0 or 3.3v and applying the opposite via 1kΩ resistor.

I do not see a better way to continue troubleshooting beyond this point without schematics for it (KBC Datasheet). These controllers are usually either pre-programmed or self-programmable on first boot, but once programmed have to be cleared with jtag to program again. If it is faulty and needs to be changed - getting a new one might not be enough to get things working again. You will also need some tools to replace this BGA.