Ga door naar hoofdinhoud

Updated version of the 2016 Retina MacBook. Model A1534, EMC 3099. Now features faster Kaby Lake processors up to 1.4 GHz Core i7 with Turbo Boost up to 3.6 GHz and up to 50 percent faster SSD, and supports up to twice the memory.

81 Vragen Bekijk alle

Only 1 second of backlight

I've been trying to diagnose a problem with mid 2017 A1534, board 820-00687.

Doesn't turn on.

When trying to turn on, LCD backlight comes on for about one second, and nothing happens after that, also there is no image while that happens or after (when shining a light on LCD).

Negotiates and takes 20v properly from charger.

Gives a chime when connecting the charger.

I'm a complete novice in board level troubleshooting, so don't be surprised if I don't know something obvious. I did read up everything I could find.

So far I think I've narrowed it down to something around the L8410 inductor.

Schematic says that after the inductor voltage should be around 45v, while instead it's 8.6 exactly like before it.

I assume something should pulse the input into it to boost the voltage for the backlight, but there is no pulsing. Just stable DC.

Probably that "something" is U8401 which controls the Q8406 MOSFET which should be doing the pulsing.

Are there any preconditions to U8401 initiating the pulsing? Or should I assume that it's dead?

Edit: BKLT_EN_R coming from CPU to U8401 does go to 3.3v and stay there.

Update (02/15/2023)

Block Image

Block Image

As for the differences, I'm using a brd file for 820-00244, so far the only difference I found was at the XW8401 slot, which shows a component in the file but not present on the board.

Beantwoord deze vraag Dit probleem heb ik ook

Is dit een goede vraag?

Score 4
7 opmerkingen

@beanman56

Thanks for tagging me in:-)

door

@con_el_queso

Have you tried a different display?

The fact you get 20v(20v is booting) looks promising but the 1 second if backlight concerns me,these models suffers from a issue called flexgate which can cause these sorts of issues

I would try with no display attached(not for long though)just to see what results you get

Also have you tried it in a external display?

Thanks:-)

door

@hellomacos Hi there, thanks for the reply.

Nope, I have no access to a different display because I don't do any of this professionally. Had a dead laptop laying around and decided it couldn't get any dead-er, so might throw my all at it.

Your mention about flexgate got me researching and I couldn't verify that A1534s were in fact affected. Could you point me to resources about that?

I have tried external display just an hour ago and no luck. But I don't know whether I did everything right while attempting.

door

@con_el_queso - This MacBook series didn’t encounter the MacBook Pro 2016/17 systems did with a short cable on the backlight cable.

door

2 commentaren meer tonen

Voeg een opmerking toe

4 Antwoorden

Het nuttigste antwoord

On the 820-00687 board if the backlight comes on for a second I would assume it is the CPU. This model is prone to the CPU failing, so much so I refuse to take in repairs for this model of MacBook.

Was dit antwoord nuttig?

Score 4

3 opmerkingen:

Might still be the case here but I think I've narrowed it down to some problem in the U8401 comms.

SCL and SDA line resistors are similar while J8500 is vacant, but SDA line drops to 0.648K as soon as the ribbon is connected.

door

@con_el_queso Yes keep diagnosing until you have ruled out everything else.

door

@imicrosoldering I think I'm basically done at this point, unless I'm missing something subtle.

Initially I found that backlight power boost converter mosfet wasn't pulsing.

That's supposed to be done by U8401, everything around it seems fine, all rails etc. Except for the I2C comms.

Traces for that end at the tcon connector and that's outside of my technical skills I guess.

door

Voeg een opmerking toe

hello there,

you've already discovered a lot and but... start from the beginning. You need to determine if the motherboard boots. Remove the matrix and connect an external monitor. Correct me if I remember wrong - this model already supports video via usb-c port (displayport 1.2). If so, follow the path (white rabbit :-). Be sure to check the values in the screen port. It would be useful to determine if the screen works. If you have another computer, or have the ability to check with another computer, it would save a lot of time.

Check presence of 3V3/5V_S0SW_LCD and DP_INT_HPD while lcd connected (pp5v_s0sw_lcd & pp3v3_s0sw_lcd)

that's all for beginning.

Was dit antwoord nuttig?

Score 3

26 opmerkingen:

Hi, thanks for the reply.

I will check whether the external monitor shows anything soon enough.

As for the board stuff to check:

DP_INT_HPD - can't check it because it's a direct mid-layer trace from cpu to lcd connector.

3V3/5V_S0SW_LCD - both are present and within range.

door

You can check DP_ONT_HPD in the matrix port, while the other two lines, if they are within range, may mean that the flap sensor is not transmitting instructions, the matrix electronics are not working or the signal tape from the motherboard has simply rubbed off - this also happens sometimes. In this case, you should see the image on an external monitor.

door

@tojaert I checked DP_INT_HPD with LCD ribbon disconnected. Basically nothing in there, just a few mV.

As for the external monitor. I tried to do that as a test but no luck, most likely because I did something wrong. It was messy because the machine has just one type C port. So I could either supply it with external power or try to make it output the signal. I didn't have the dock to manage doing both through a single port.

So I had to power the board from the battery, but when powered from a battery, I don't know what's going on inside at all, no fan spin, no nothing, just one tiny white smd led, which I don't even know the meaning of.

door

I checked DP_INT_HPD from the connector pin, with LCD ribbon disconnected, nothing happening there, just a few mV.

As for the external monitor, I tried to do that, but no luck. probably because I did something wrong.

The whole process was messy because the machine has only one type C port, so I could wither supply power or try make it output the signal but not both.

So I had to try and power it from the battery, but when attempting that, I have no idea what's going on at any give time, no fan spin, no nothing. Only one tiny white smd led, which I don't even know the meaning of.

door

so I guess it's not a backlight problem but a different one. You have to go deeper, to the power supply and battery. Check there. EDP_BKLT_EN comes to mind which is necessary to run the u8401 backlight system.

DP_INT_HPD will only be present at machine startup. Check the power load at startup. Prove that the motherboard starts at all (even no image and backlight).

door

21 commentaren meer tonen

Voeg een opmerking toe

@con_el_queso

Have you tried a different display?

The fact you get 20v(20v is booting) looks promising but the 1 second if backlight concerns me,these models suffers from a issue called flexgate which can cause these sorts of issues

I would try with no display attached(not for long though)just to see what results you get

Also have you tried it in a external display?

Thanks:-)

Was dit antwoord nuttig?

Score 1

21 opmerkingen:

I have tried several times already to answer here but have no idea where my messages are disappearing to after I press "Post comment".

This is my 4th attempt I think.

@HelloMacOS

door

@con_el_queso the @hellomacos is different than the one you used @HelloMacOS :-) iRobot mistook it for Spam and busted your comments. We'll reverse that decision ;-)

door

Huh, it works if I don't try to reply to a specific person:

I have not tried a different display because I have no access to it.

After your mention of it, I tried to research the flexgate but could not verify that A1534s were affected by it. Could you point me to more detailed info about it?

I have tried external display just an hour ago and no luck. But I don't know whether I did everything right while attempting.

@hellomacos

door

@oldturkey03 Roger that. I hope I'm on more chummy terms with iRobot after this :)

door

@con_el_queso Awesome. iRobot is actually pretty cool when it gets to know you :-) I keep my eyes on your rep points for a bit. Just to make sure that it does not get carried away again. A bit overzealous....Welcome to iFixit!

door

16 commentaren meer tonen

Voeg een opmerking toe

@hellomacos - Let’s clarify things FlexGate was only an issue with the MacBook Pro’s 2016 & 2017 as these had a short cable so the backlight flex cable would fail.

Now due to the new method Apple used there is a higher chance of the cables running between getting damaged when the back edge gets a bang (black clutch cover). A second risk if getting an object caught along the inside just at the base of the LCD panel there is thin glass plate (Chin Plate) which covers the ribbons. What can happen is the object presses cracking the glass which then has a sharp edge punch’s into the cable damaging it (or a shard breaks free cutting the cable). So it’s important to inspect both areas for signs of damage.

The fact the chin plate was damage aims me to a damaged display cable!

This needs a retest this time closing the display with an external display connected (you can also use a magnet to fire off the Hall sensor if you know where it’s located). If the image is present on the external that proves the issue is the display assembly needs replacing. If not then the issue is within the logic board. I would still clean up and replace the chin plate to prevent future damage.

Was dit antwoord nuttig?

Score 0
Voeg een opmerking toe

Voeg je antwoord toe

The Man Himself zal eeuwig dankbaar zijn.
Weergavestatistieken:

Afgelopen 24 uren: 0

Afgelopen 7 dagen: 0

Afgelopen 30 dagen: 6

Altijd: 496