Hardware Fix broken Switch? It's not turning on and also not charging

[RathX]

Hey I hope I can find some help here, I'm really frustrated and can't find any solution to my problem.
A few things to know,

• The switch has AutoRCM enabled
• An RCMx86 chip was soldered, but even after desoldering the switch doesn't work (Everything worked fine, before the problem happened)
  
• The battery reads 4.2V, I've charged it externally to be sure that the battery isn't causing the problem

A few days ago I put the switch in stock mode (SysNand) into sleep, the next day, the device wouldn't turn on and on the display it showed a battery icon at 0% and 3000mW. I let it charge for 48h with the original charger, which is also used for the dock. But after that the switch wouldn't turn on at all. My PC doesn't recognize any new connected devices, when I try to connect the switch, no matter of the USB Port (Can't see anything different after connecting in TegraRCM as well).
So I've taken the switch apart, desoldered the chip and measured all the caps around P13USB, M92T36 and BQ24193. But nothing, everything looks OK to me, no beeping and the caps seem to be fine.
I've found out two things marked in the attached picture....
First when I measured the red and green line, there was no connection, so I desoldered the USB choke and put two wires on the red and green connections. So now I can measure a current of 4.2V on each wire, but the switch is still not turning on! (And I now have to order a new choke)
Second the yellow marked cap doesn't seem to have any resistance and doesn't seem to have an anode, when connecting either side with ground.

I'm really lost and I don't know what could be causing the problem, if anyone has any clue, it would be really appreciated!View attachment 265533

 

[jkyoho]

yellow circle one is FUSE not CAP, you should have continuity or 0.x resistance (diode measure around 0.5xx each side)
I suggest remove the fuse, make a blob solder and see if charging voltage go into board first

 

[HollowGrams]

Hey I hope I can find some help here, I'm really frustrated and can't find any solution to my problem.
A few things to know,

• The switch has AutoRCM enabled
• An RCMx86 chip was soldered, but even after desoldering the switch doesn't work (Everything worked fine, before the problem happened)
  
• The battery reads 4.2V, I've charged it externally to be sure that the battery isn't causing the problem

A few days ago I put the switch in stock mode (SysNand) into sleep, the next day, the device wouldn't turn on and on the display it showed a battery icon at 0% and 3000mW. I let it charge for 48h with the original charger, which is also used for the dock. But after that the switch wouldn't turn on at all. My PC doesn't recognize any new connected devices, when I try to connect the switch, no matter of the USB Port (Can't see anything different after connecting in TegraRCM as well).
So I've taken the switch apart, desoldered the chip and measured all the caps around P13USB, M92T36 and BQ24193. But nothing, everything looks OK to me, no beeping and the caps seem to be fine.
I've found out two things marked in the attached picture....
First when I measured the red and green line, there was no connection, so I desoldered the USB choke and put two wires on the red and green connections. So now I can measure a current of 4.2V on each wire, but the switch is still not turning on! (And I now have to order a new choke)
Second the yellow marked cap doesn't seem to have any resistance and doesn't seem to have an anode, when connecting either side with ground.

I'm really lost and I don't know what could be causing the problem, if anyone has any clue, it would be really appreciated!View attachment 265533

Make sure fuse is good like mentioned above and check coil to the right of the fuse, when they go out you will get exact symptons you are having. I have fixed this exact issue before and its not always the same fix:


M92
P13
BQ
USB C Port
Coil

These ics can read fine and still be bad sometimes.

 

[RathX]

yellow circle one is FUSE not CAP, you should have continuity or 0.x resistance (diode measure around 0.5xx each side)
I suggest remove the fuse, make a blob solder and see if charging voltage go into board first

I've went ahead and did the blob solder, there is a ~5V current between the blob and the battery slot when plugging in USB.. I guess a new fuse would be needed now, do you know what the part is called?

I don't know if it should've fixed the issue, but the switch still isn't turning on. :/

Make sure fuse is good like mentioned above and check coil to the right of the fuse, when they go out you will get exact symptons you are having. I have fixed this exact issue before and its not always the same fix:


M92
P13
BQ
USB C Port
Coil

These ics can read fine and still be bad sometimes.

I guess the USB port should be fine, if current reaches the battery slot from USB right? How do I check the coil?

 

[HollowGrams]

I've went ahead and did the blob solder, there is a ~5V current between the blob and the battery slot when plugging in USB.. I guess a new fuse would be needed now, do you know what the part is called?

I don't know if it should've fixed the issue, but the switch still isn't turning on. :/



I guess the USB port should be fine, if current reaches the battery slot from USB right? How do I check the coil?

Multi-meter continuity mode - Both leads, 1 each side = should beep if good.

 

[RathX]

Multi-meter continuity mode - Both leads, 1 each side = should beep if good.

Oh it does beep, so all good with the coil I assume

 

[ppzikos]

look inside usb port with you camera phone or microscope , if there is a bend pin...
when charger is plug ( no battery ) : test M92t36
check if you have 3-6V on pin 5 & 6 & 9
check if you have 0.8 - 1.4V on pin 35 OR 37 ( cc1 OR cc2 ) ( this is for detect orientation charger, only one pin have voltage at same time )
check if you have 0.8 - 1.4V on pin 36

> if all is good except pin 6 of M92, check if is short. if no short change M92 and maybe usb
> if pin 9 have 5V and pin 5 have 0V , change change M92 and maybe usb too.


when only battery is plug test BQ24193
check Voltage pin 1 &2 & 13 & 15
and :
when battery + Charger test BQ24193
check Voltage pin 1 & 2 & 13 & 15

does not measure to the pin of the IC directly, disconnects battery and charger and finds a track or component easier next to it, after having verified that it is well on the desired pin

remove IC for detect short is a bad idea........ but if you want more problem it's good

edit : Switch not need P13 for start , only puts it back when the Switch will work
edit 2 , if you remove your nand the switch boot in rcm mode automatically , no need vol+/power and jig.

 

[RathX]

look inside usb port with you camera phone or microscope , if there is a bend pin...
when charger is plug ( no battery ) : test M92t36
check if you have 3-6V on pin 5 & 6 & 9
check if you have 0.8 - 1.4V on pin 35 OR 37 ( cc1 OR cc2 ) ( this is for detect orientation charger, only one pin have voltage at same time )
check if you have 0.8 - 1.4V on pin 36

> if all is good except pin 6 of M92, check if is short. if no short change M92 and maybe usb
> if pin 9 have 5V and pin 5 have 0V , change change M92 and maybe usb too.

That's a lot of information, thank you!
The USB Port looks fine to me, couldn't see any bent pins.

On M92 I've measured ~4.5V on pins 5&9, but around 0V on pin 6. I've checked the blue marked cap (see picture: blue circle), but there is no short, seems to be okay.
-> Does this mean M92 is faulty?

I've measured ~4V on pins 35 and 36, but I could've also made a mistake, since I need to find better/easier contacts to measure. And ~1V on pin 37.

when only battery is plug test BQ24193
check Voltage pin 1 &2 & 13 & 15
and :
when battery + Charger test BQ24193
check Voltage pin 1 & 2 & 13 & 15

does not measure to the pin of the IC directly, disconnects battery and charger and finds a track or component easier next to it, after having verified that it is well on the desired pin

remove IC for detect short is a bad idea........ but if you want more problem it's good

edit : Switch not need P13 for start , only puts it back when the Switch will work
edit 2 , if you remove your nand the switch boot in rcm mode automatically , no need vol+/power and jig.

I will check BQ24 in a few hours after getting some sleep, and also try to boot the switch without emmc? Should I just remove the module?

 

[ppzikos]

when the module nand is removed, if switch is off you plug usb on laptop and switch boot direct in rcm mode and pc detect switch. for test it's fast.

but if pin 6 is no voltage , the soc have not voltage too (no boot no rcm... etc), pin 6 is needed for start other voltage.(3.3V 1.8V 1.1V etc)

ok i think m92 is bad , but i need mesurement of bq for be sure.
on m92 check , pin 38-39-40 ( there is a 3 caps on top ) it's ldo internal of m92 , 1.5v ,2.8v ,1.5v check if is good.

i go to sleep to

--------------------- MERGED ---------------------------

I've measured ~4V on pins 35 and 36, but I could've also made a mistake, since I need to find better/easier contacts to measure. And ~1V on pin 37.

there is no better contact for this pin, and your mesurement is good.

 

[Modificatorul]

"First when I measured the red and green line, there was no connection, so I desoldered the USB choke and put two wires on the red and green connections. So now I can measure a current of 4.2V on each wire, but the switch is still not turning on! (And I now have to order a new choke) "

You have or you had a short in usb c port,that's why it burned choke on D+ and D - (point 2 and 3 on your picture ) and fuse.

Disconnect usb cable and battery, test continuity between point 2 and 6(vcc) , and point 3 and 6(vcc). if you have continuity beetween Data + and Data - to vcc your usb c port is still shorted and must by removed.

If you don't have a short, set your multimeter now to diode mode and test point 2 and 3 to gnd, post if the multimeter show any values.

Data + and Data - go to esd filter remove it if is present to elimnate problems (not present on all boards) then choke and then directly to tegra x1 soc.
if this logic part in tegra x1 is burned you can't send payloads to start your switch,tegra rcm will not detect rcm mode, and your switch is in auto rcm from your post..

Post the results .

 

[RathX]

Ok, got me a better multimeter today.
Is there a certain spot where is should be measuring for the M24 pins? I've always held one contact on the positive battery pin.
So it's weird with pin 5&6, since the voltages change wether I connect to the cathode or anode pin on the battery slot.

when the module nand is removed, if switch is off you plug usb on laptop and switch boot direct in rcm mode and pc detect switch. for test it's fast.

Result: My PC doesn't recognize the switch at all, even without the eMMC module​

but if pin 6 is no voltage , the soc have not voltage too (no boot no rcm... etc), pin 6 is needed for start other voltage.(3.3V 1.8V 1.1V etc)

ok i think m92 is bad , but i need mesurement of bq for be sure.

I've measured BQ24, the pins had a current like this [only battery]:

• pin 1&2: ~battery voltage (currently 2V; 4.2V unconnected)
• pin 13&15: 0mV AND is shorted when connected to cathode (beeping)

[with battery + charger]:

• pin 1: 1V
• pin 3: 4.2V
  
• pin 13&15: 0mV AND is shorted when connected to cathode(beeping)

on m92 check , pin 38-39-40 ( there is a 3 caps on top ) it's ldo internal of m92 , 1.5v ,2.8v ,1.5v check if is good.

It measures 2.8V on pin 38, 1.4V on pin 39 and 4.2V on pin 40 with battery+cable.​

Disconnect usb cable and battery, test continuity between point 2 and 6(vcc) , and point 3 and 6(vcc). if you have continuity beetween Data + and Data - to vcc your usb c port is still shorted and must by removed.

If you don't have a short, set your multimeter now to diode mode and test point 2 and 3 to gnd, post if the multimeter show any values.

To be honest, I don't know if the d+ and d- points had any issues to begin with, could've been a bad multimeter.. but I didn't measure a short between d+/- and the fuse/point 6.
When connecting D+ and D- with GND the multimeter shows a value of ~0.75V.​

Data + and Data - go to esd filter remove it if is present to elimnate problems (not present on all boards) then choke and then directly to tegra x1 soc.
if this logic part in tegra x1 is burned you can't send payloads to start your switch,tegra rcm will not detect rcm mode, and your switch is in auto rcm from your post..

I guess the part marked on the attached picture is meant, right? I just want to make sure that I don't desolder it when it's still ok. Is there a way to find out if it is still ok?​

 

[Modificatorul]

Ok, got me a better multimeter today.
Is there a certain spot where is should be measuring for the M24 pins? I've always held one contact on the positive battery pin.
So it's weird with pin 5&6, since the voltages change wether I connect to the cathode or anode pin on the battery slot.

Result: My PC doesn't recognize the switch at all, even without the eMMC module​

I've measured BQ24, the pins had a current like this [only battery]:

• pin 1&2: ~battery voltage (currently 2V; 4.2V unconnected)
• pin 13&15: 0mV AND is shorted when connected to cathode (beeping)

[with battery + charger]:

• pin 1: 1V
• pin 3: 4.2V
  
• pin 13&15: 0mV AND is shorted when connected to cathode(beeping)

It measures 2.8V on pin 38, 1.4V on pin 39 and 4.2V on pin 40 with battery+cable.​

To be honest, I don't know if the d+ and d- points had any issues to begin with, could've been a bad multimeter.. but I didn't measure a short between d+/- and the fuse/point 6.
When connecting D+ and D- with GND the multimeter shows a value of ~0.75V.​

I guess the part marked on the attached picture is meant, right? I just want to make sure that I don't desolder it when it's still ok. Is there a way to find out if it is still ok?​

Acording to your picture and the lines you drow the esd ic is shorted to gnd on D+ point 2.

Test again 2 and 3 point to ground in continuity mode if shorted remove the esd ic, and test again and see if the short disaperead, then connect the battery and usb cable and see if tegra show the device now, i asume you have the drivers installed if not, see tegrarcmgui seetings tab.

 

[RathX]

Acording to your picture and the lines you drow the esd ic is shorted to gnd on D+ point 2.

Test again 2 and 3 point to ground in continuity mode if shorted remove the esd ic, and test again and see if the short disaperead, then connect the battery and usb cable and see if tegra show the device now, i asume you have the drivers installed if not, see tegrarcmgui seetings tab.

Ok, I've applied some flux around the ESD, and made sure everything is clean, maybe some point was connected to the shielding right next to the ESD, but now I measure the pins like on the picture.
No shorting to ground on D+ and D-.


Connecting to PC still doesn't show anything in TegraRCM. Could the power button maybe be damaged? Is there a way to jump 2 points/ to make sure the power button is working correctly?

Or should the PC recognize the switch as soon as I connect a USB cable without me needing to press the power button?

 

[Modificatorul]

If i remember corectly when you connect the usb cable should power on.

I whant you to test if D+ and D- is going from mb through port but you need a usb c test board for this.

Probably i'm on wrong path with this data comunication, so probably replacing mn92t36 will solve the isue folow @ppzikos advices and see were it goes.

 

[ppzikos]

pc no detect swith without push power button if m92 is dead.

If you don't have a short, set your multimeter now to diode mode and test point 2 and 3 to gnd, post if the multimeter show any values.

test this please , NO BATERRY AND NO CHARGER in diode mode on your multimeter , red on ground and black for mesure pad 2 and 3 , (after choke if is removed).
it's for test tegra soc , if data line is burn or not in tegra.

-----

m92 pin 38-39-40, your mesurement is very wrong , retry please. only with charger no need push power button.
take mesurement on caps .

BQ24 pin 13&15: 0mV AND is shorted when connected to cathode (beeping) ,
what ? i dont understand it's shorted to ground without battery and charger ? or 13 & 15 short together ?

reminder : do not test a short circuit with the battery or charger.

 

[RathX]

m92 pin 38-39-40, your mesurement is very wrong , retry please. only with charger no need push power button.
take mesurement on caps .

I think I've measured on the wrong points all the time until now, is the marked spot correct on the picture?
Because now I've got the right voltage I think 1.5V, 2.8V, 1.5V on pins 38-39-40. The tip with measuring on the caps made it easier, thanks.

BQ24 pin 13&15: 0mV AND is shorted when connected to cathode (beeping) ,
what ? i dont understand it's shorted to ground without battery and charger ? or 13 & 15 short together ?

If I've measured correct (measured on the marked spot on the battery like in the picture), I read now (BQ24):
Battery only
4,2V on pin 3, 13, 15
and 0V on pin 1

Battery+USB Cable
5V on pin 1
0V on pin 3
3,5V on pin 13 and 15

edit: forgot the picture

 

[ppzikos]

test this please , NO BATERRY AND NO CHARGER in diode mode on your multimeter , red on ground and black for mesure pad 2 and 3 = D+ D - (usb data line), (after choke if is removed).
it's for test tegra soc , if data line is burn or not in tegra.

 

[RathX]

Oh sorry, forgot to mention that part. It's 0.75V on both points 2 and 3

 

[ppzikos]

yes it's good , usb data line of tegra is not dead , i think m92 is dead , & BQ is good , maybe usb is damaged but i'm not sure because all is good. your choke of usb dataline is dead or not ? , if is dead it's strange and you must change usb and choke.

edit :
test m92 pin 6 , with charger + battery + push power , if 0V change M92

 

[RathX]

I'm not sure, should've checked better
I'll order M92 and some chokes and fuses, hopefully the problem will be fixed! I don't get it how it died, but whatever I guess.

Thank you all for your help, I'll update after changing the parts.