MacBook Pro Unibody - No Backlight - Logic Board Question

[tlars]

I think I may have this same issue with a MBP17 (A1212). Any advice or photos of where this fuse may be on the motherboard? I removed my screen assembly and connected it to a same model MBP17 and the backlight worked...so I'm pretty sure I have eliminated the inverter, inverter cable and backlight itself.

same problem but more damage..2010 Macbook Pro A1278, Fluid on Keyboard, let dry 5 days , smoke, give to dad to fix. Screen active but no backlight. lvds cable connector base disolved. (through to multilayer circuitboard) .cleaned with replaced fuse. more smoke. observation:
damage only when fuse connected , only at pins 21+22
3 red leads on LVDS cable.
ohmeter from Diode test point (27V) to copper foil that may have connected to 21+22 is 0.001 ohms.
legs of chip all not shorted.
issue= is LED shorted in Screen , or is damage to connector causing short ( solder from sufacemount has climbed up leads into connector.)
have been lucky that chip did not fry. (smoke says that driver chip still functioning. reluctant to test without more info
next test is to power up with cable disconnected , but expect smoke and more damage ,so last choice.
do we know how to power LED's to test? /send 27V down leads and make screen glow/ do we know what wires ? (not far end of cable mentioned earlier in thread , on this model)

 

[Dadioh]

Yikes

You do know that electronics all work on smoke? The reason we know this is that if you let out the smoke they stop working

Seriously though, that smoke comes from burning material. You think it is the plastic part of the lvds connector? Any chance of getting close up high res photo of the area?

Going from memory the led power is at the far end of the lvds connector. Need to check which end. I would be surprised if LEDs were shorted in the display. Most likely a short at the lvds connector but a picture would help.

 

[sincesystem7]

Infinite respect

Dadioh, I have read this thread from beginning to end several times, and I'm very impressed with your knowledge and apparent skill. I'm hoping that maybe you can help me out here.

I have a Macbook Unibody A1342 (the white one) with this very same backlight problem. If you could show me how to find that paticular backlight fuse, and tell the most basic soldering equipment I need to bridge it, I would I would really appreciate it.

My soldering skills are not above average, but in my current situation this is the only solution I can afford. Any advise at all would be very helpful.

Thanks.

 

[Dadioh]

Dadioh, I have read this thread from beginning to end several times, and I'm very impressed with your knowledge and apparent skill. I'm hoping that maybe you can help me out here.

I have a Macbook Unibody A1342 (the white one) with this very same backlight problem. If you could show me how to find that paticular backlight fuse, and tell the most basic soldering equipment I need to bridge it, I would I would really appreciate it.

My soldering skills are not above average, but in my current situation this is the only solution I can afford. Any advise at all would be very helpful.

Thanks.

Is the A1342 the unibody polycarbonate? If so then it also has an LED screen and I suspect the circuit will be almost identical. Can you post some high resolution pictures of the logic board both front and back near the lvds connector? If the size of the photo is a problem then pm me and we can arrange for you to email pics to me. I may then ask you to ohm out a few spots to aid in determining the circuit.

 

[tlars]

re: Yikes. here is pic. damage is clearer under 10X magnifier (mono / for stamps = $7) I have soldered a physically larger 2amp fuse vertically to one side and a wire return in place of tiny fuse.
have not cleaned board since smoke.
bad as it looks, with the fuse out , the system boots , and the screen seems complete (as far as I can see shining light from behind.)
my objective is either to bypass the I-pex connector from existing LED power .problem:Multilayer Board makes following/ cutting trace difficult
or supply alternate 27V power , then only question is
Which wires on I-PEX are for LED?

re: Yikes. here is pic. damage is clearer under 10X magnifier (mono / for stamps = $7) I have soldered a physically larger 2amp fuse vertically to one side and a wire return in place of tiny fuse.
have not cleaned board since smoke.
bad as it looks, with the fuse out , the system boots , and the screen seems complete (as far as I can see shining light from behind.)
my objective is either to bypass the I-pex connector from existing LED power .problem:Multilayer Board makes following/ cutting trace difficult
or supply alternate 27V power , then only question is
Which wires on I-PEX are for LED?

 

[BrianZ]

Help.

Hi Dadioh,

I m having the same problem with A1297. i found out about this tread by accident as i was searching for a solution , In my case there was a little liquid damage with no burning smell or marks. The unit powers up but no screen.
I believe it is the same issue. Going over the post , I understand that the issue is related to LED driver, but i m not sure on how to go from here.

I have attached a pic , would you please point what may need to be changed.

Thank you for starting this thread. I really appreciate it.

 

[Dadioh]

re: Yikes. here is pic. damage is clearer under 10X magnifier (mono / for stamps = $7) I have soldered a physically larger 2amp fuse vertically to one side and a wire return in place of tiny fuse.
have not cleaned board since smoke.
bad as it looks, with the fuse out , the system boots , and the screen seems complete (as far as I can see shining light from behind.)
my objective is either to bypass the I-pex connector from existing LED power .problem:Multilayer Board makes following/ cutting trace difficult
or supply alternate 27V power , then only question is
Which wires on I-PEX are for LED?

The signal (27V) that drives the LED's is on pins 21 and 22. The returns are on pins 24 through 29. Those burned pins seem to be around the middle of the connector? Maybe 21 and 22?

 

[Dadioh]

Hi Dadioh,

I m having the same problem with A1297. i found out about this tread by accident as i was searching for a solution , In my case there was a little liquid damage with no burning smell or marks. The unit powers up but no screen.
I believe it is the same issue. Going over the post , I understand that the issue is related to LED driver, but i m not sure on how to go from here.

I have attached a pic , would you please point what may need to be changed.

Thank you for starting this thread. I really appreciate it.

I am going to need better resolution to zoom in and see the devices. Also, I don't see the WLED driver chip (the one that drives the LED backlight) in that shot so you will need to take high res pics of the back of the board as well.

Or if you do some google searching maybe you can find a high res picture for that logic board already taken by someone else. If you post the pics I can help locate the fuse.

And to be clear... when you say no screen do you mean there is an image but just no backlight to illuminate it? Easy check is to shine a flashlight in the middle of the screen while booting. If you can see the apple logo then the display signals are working but just no backlight. Please confirm.

 

[BrianZ]

more info : Dadioh

I do get video on external display. and there is a dim image on the built-in display. with a flash light i can see clearly what s on the screen. as per high resolution picture. macrumors have a 1.14 mb limit which results on a low resolution picture. can i send you a high resolution picture by email or other means.

Thanks for your help.

 

[Dadioh]

I do get video on external display. and there is a dim image on the built-in display. with a flash light i can see clearly what s on the screen. as per high resolution picture. macrumors have a 1.14 mb limit which results on a low resolution picture. can i send you a high resolution picture by email or other means.

Thanks for your help.

Drop the images in photobucket.com or any other free online site and then post the links here.

 

[BrianZ]

Higher resolution pic

Drop the images in photobucket.com or any other free online site and then post the links here.

Thank for taking time to look at the pic, hopefully this original pic may help.
if it does not i may have to take a better pictures.

please see link below

http://s1180.photobucket.com/albums/x409/brianz1/?action=view&current=IMGP0073.jpg

 

[Dadioh]

Thank for taking time to look at the pic, hopefully this original pic may help.
if it does not i may have to take a better pictures.

please see link below

http://s1180.photobucket.com/albums/x409/brianz1/?action=view&current=IMGP0073.jpg

Still a bit blurry. Looking for a beige part with a dot on the top. Can you take a photo of the other side of the board.

 

[junn]

See image below. This is a 2A fuse in an 0402 package. I think it is 16V but can't remember. It carries 12V normally. Look for the little white dot on the top. Use a multimeter and check if it is an open circuit (unplug the battery first).

Thanks dude! Another macbook is saved for 0$.
And for all others - you can just short circuit this fuse and its working.
It can be done with a very fine tip soldering iron.
But like DADIOH said, this could lead WLED driver burnout if the logic board failure/damage is unknown/larger.

But - i did it and this mac is working as new, except the keyboard backlit.
Maybe it has one fuse more

And finally - this post can save thousands of $$$ for people all ower the world!
Thanks again!

 

[tlars]

The signal (27V) that drives the LED's is on pins 21 and 22. The returns are on pins 24 through 29. Those burned pins seem to be around the middle of the connector? Maybe 21 and 22?

right again!
with the LVDS cable disconnected ,when I supply 24V+ to pins 21+22 and the negative on pins 25 or 26ish
the back light lights up.GREAT first step. (very low current)
seems to be positive ground.
how are the LED's wired? Serial I thought , but the returns are high resistance from each other on the cable and on the board. each has a resistor then goes to the chip (which may be fried). (multiple 27volt strings in parallel for brightness?)
next test : cable plugged in , wires 21+22 cut back and connected to 24V+ , neg on 25 or 26 .but acts shorted. therefore need to cut return wires but want to understand first. brightness, timing?

 

[Dadioh]

right again!
with the LVDS cable disconnected ,when I supply 24V+ to pins 21+22 and the negative on pins 25 or 26ish
the back light lights up.GREAT first step. (very low current)
seems to be positive ground.
how are the LED's wired? Serial I thought , but the returns are high resistance from each other on the cable and on the board. each has a resistor then goes to the chip (which may be fried). (multiple 27volt strings in parallel for brightness?)
next test : cable plugged in , wires 21+22 cut back and connected to 24V+ , neg on 25 or 26 .but acts shorted. therefore need to cut return wires but want to understand first. brightness, timing?

That is good news.

There are six return lines with 10ohm (going from memory) current limiting resistors. I think there are 6 parallel strings of LEDs. Again from memory I think maximum brightness is 27v and minimum brightness is 18v (the last step before screen shuts off using MacBook brightness controls).

As far as timing goes, the wled driver has an i2c signal from the smc which gets asserted after the smc goes through initialization of power supplies and releases reset of the processor and gpu. Obviously it also responds to sleep modes as well. Not sure where you are going to generate the 24-27v for the backlight. The next highest voltage on the board is about 17.5 volts from the MagSafe. It is supposed to be 18.5v (on paper) but it is less on every one I've looked at.

 

[tlars]

That is good news.

There are six return lines with 10ohm (going from memory) current limiting resistors. I think there are 6 parallel strings of LEDs. Again from memory I think maximum brightness is 27v and minimum brightness is 18v (the last step before screen shuts off using MacBook brightness controls).

As far as timing goes, the wled driver has an i2c signal from the smc which gets asserted after the smc goes through initialization of power supplies and releases reset of the processor and gpu. Obviously it also responds to sleep modes as well. Not sure where you are going to generate the 24-27v for the backlight. The next highest voltage on the board is about 17.5 volts from the MagSafe. It is supposed to be 18.5v (on paper) but it is less on every one I've looked at.

my earlier comment about positive ground took a while to make sense. the "wires" are coaxial. the small center wire is not easy to bare and solder. (was grounding to shield)
fed external 24volt to one Red (21)cut from Ipex and one Blue (26)cut from Ipex.
Screen lights up functionally bright and shows that the rest of the computer is undamaged.(mouse, wireless, Keyboard all working)
proof this machine is worth the remaining salvage effort to assemble a simple voltage boost pump.

 

[Dadioh]

my earlier comment about positive ground took a while to make sense. the "wires" are coaxial. the small center wire is not easy to bare and solder. (was grounding to shield)
fed external 24volt to one Red (21)cut from Ipex and one Blue (26)cut from Ipex.
Screen lights up functionally bright and shows that the rest of the computer is undamaged.(mouse, wireless, Keyboard all working)
proof this machine is worth the remaining salvage effort to assemble a simple voltage boost pump.

Great work!

So they use coax to provide the backlight voltage? That seems odd. Coax is usually used for high frequency signals rather than DC voltages. Good info.

You could build a boost circuit driven by the +5V rail. That way the backlight will only come on after the SMC has done all of it's power checks and asserted the rail. Power draw on the LED's is small so the 5V will not miss the power.

 

[sincesystem7]

Should have had more courage

Hello Dadioh.
After I posted to this thread I chickened out, and sent my book to one of those "fix it or free" places on Ebay. I got it back with two stripped screws and a full refund. This time I'm following your directions to the letter. I managed to get the logic board free, and I have attached photos if you could tell me where the backlight fuse is, and what I need to do to bridge it, I would really appreciate it

 

[macrepair]

MacBook 13" A1278

Dadioh do you know if this part is the WLED driver on MacBook 13" A1278? Looks like a 20 pin QFN package. Would like to replace this but can't seem to find it anywhere or probably just searching for the wrong part number! Top of component reads:

APP_1A
QR1.2F
0904A

 

[Dadioh]

Dadioh do you know if this part is the WLED driver on MacBook 13" A1278? Looks like a 20 pin QFN package. Would like to replace this but can't seem to find it anywhere or probably just searching for the wrong part number! Top of component reads:

APP_1A
QR1.2F
0904A

That is the one. Unfortunately I believe this is a proprietary Apple part so I haven't been able to find a source for it. This part uses an external FET for the boost circuit whereas the National part has an internal FET.

If anyone does discover if this a rebranded part available under a different part number that would be great news but for now I think it comes down to scavenging them off dead logic boards

 

[Dadioh]

Hello Dadioh.
After I posted to this thread I chickened out, and sent my book to one of those "fix it or free" places on Ebay. I got it back with two stripped screws and a full refund. This time I'm following your directions to the letter. I managed to get the logic board free, and I have attached photos if you could tell me where the backlight fuse is, and what I need to do to bridge it, I would really appreciate it

I don't think there is enough resolution for me to pick it out. You want to start by finding the WLED driver chip. It will be a small device with either 7 or 8 pins on each of the 4 sides. On the top you will probably see APP001 or LP8543. You may need some sort of magnification to pick it out (I use a microscope). Then you need to search in that area for a small device, beige in color, with a white dot on the top. This will "probably" be the fuse but if you can get a closeup shot in that area I could maybe help identify it.

 

[macrepair]

That is the one. Unfortunately I believe this is a proprietary Apple part so I haven't been able to find a source for it. This part uses an external FET for the boost circuit whereas the National part has an internal FET.

If anyone does discover if this a rebranded part available under a different part number that would be great news but for now I think it comes down to scavenging them off dead logic boards

Thanks for the info! Guess that's gonna make it a difficult part to source! Do you know where the external FET is located on these boards & do you think the FET is more likely to fail from liquid spills or the WLED driver or both?

 

[Dadioh]

Thanks for the info! Guess that's gonna make it a difficult part to source! Do you know where the external FET is located on these boards & do you think the FET is more likely to fail from liquid spills or the WLED driver or both?

I haven't seen the FET fail yet. Most of the time the fuse blows. I have a had a few where the WLED driver device fails. The WLED driver is generating 27V so it is the most likely culprit in liquid spills. The rest of the logic board is 12V, 5V, 3.3V, 1.0V etc... so they are not as likely to damage during a liquid spill. Long term corrosion is more of an issue with the low voltage stuff.

As far as the location of the FET goes I can't recall off the top of my head. I would have to go look at a board.

 

[macrepair]

There should be a diode near the LVDS connector. On the cathode of that diode (marked with a white stripe) is the 27V that drives the LED's in the display. If you have approximately 27V there but no backlight then the issue is either the LVDS cable or the display. If there is no voltage then it is the WLED driver circuit. You want to be VERY careful probing on the board while it is running. One little slip and you will let the smoke out.

This is what the diode looks like on the 13" MBP unibody.

Having a problem with one of the boards I have here. I have already changed the WLED driver twice thinking the one may be faulty, but only getting between 10-11V where there should be 27V, and still no backlight. Any ideas what else may be wrong in the circuit? Thanks again for your help..

 

[Dadioh]

Having a problem with one of the boards I have here. I have already changed the WLED driver twice thinking the one may be faulty, but only getting between 10-11V where there should be 27V, and still no backlight. Any ideas what else may be wrong in the circuit? Thanks again for your help..

That means the circuit is not boosting. I had one like that. Check the enable pin. I had one with a burnt resistor in the voltage divider that supplies the enable.