• Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage

Zasady działu:Kliknij, aby przejrzeć zasady tego działu

  1. Tytuł wątku powinien zawierać markę i pełny model sprzętu oraz skrótowy opis usterki.
  2. Treść wątku powinna zawierać dokładny opis usterki, co zostało sprawdzone/wymienione/zmierzone, wyniki tych działań, oznaczenie kodowe PCB (w przypadku napraw hardware) oraz sformułowane pytanie. W przypadku problemów z odczytaniem oznaczenia kodowego PCB, należy przeczytać TEN WĄTEK. Jeśli nadal nie można znaleźć oznaczenia, należy zamieścić wyraźne zdjęcia obu stron płyty - należy przy tym usunąć wszelkie folie i moduły, mogące ewentualnie zasłaniać oznaczenia płyty.
  3. Przed napisaniem nowego tematu, każdy użytkownik zobowiązany jest zapoznać się z wątkami w dziale SZKOLENIA i na ich podstawie dokonać próby wstępnej diagnostyki uszkodzenia.
  4. Nie jest dozwolone umieszczanie załączników w postaci kompletnych dokumentacji, a także jakichkolwiek wsadów BIOS / firmware (do tego służy dział PLIKI BIOS DO WERYFIKACJI). Dozwolone jest wyłącznie umieszczenie fragmentu dokumentacji, który ułatwi autorowi diagnostykę/naprawę sprzętu z wątku. W jednym poście można zamieścić nie więcej, niż jedną stronę instrukcji/schematu. Plik nie może posiadać widocznych znaków wodnych, napisów "confidential", adresów email etc.
  5. Nie jest dozwolone zamieszczanie linków do plików, znajdujących się na innych stronach internetowych.
  6. Nie jest dozwolone zakładanie tematu, który opisuje usterkę więcej niż jednego sprzętu. W takich przypadkach należy utworzyć osobne wątki dla każdego sprzętu.
  7. Nie jest dozwolone pisanie tematu lub postu, mającego charakter zapytania o schemat, boardview, zdjęcia płyty, wsad BIOS lub instrukcję serwisową. Do takich zapytań służy dział ZAPYTANIA O DOKUMENTACJE/BIOS.
To jest wątek ARCHIWALNY. Możesz odpowiedzieć w nim tylko wtedy, gdy Twój post zawiera rozwiązanie problemu (Regulamin forum p. 12.1).

#1 Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Cubytus 4 stycznia 2016, 18:50
Hi there,

here's my updated topic with additional information.

Failure description:
I got a water-damaged MBP where an unknown chip (probably related to Thunderbolt and HDMI functions) stays active and hot during sleep. The battery discharges much faster than it would do otherwise, and so deeply that time and date settings are lost and reset to 2008 when the main battery can no longer provide enough juice. Having the wrong date means certificates are not recognized as valid, and prevent wireless connection.

Chip that stays hot is circled in red (this isn't my picture!):
Obrazek

PCB marking codes:
820-2936-B

Already attempted:
According to the only shop that does this kind of repair in this city, replacing the burned chip is just too dangerous for the computer to be worth it, on an otherwise working machine. It's a 40-pin QFN package, very close to other tiny components as shown in the picture. Another repair shop does it across the border, but asks $400 for it, with no guarantee it will work. I don't have the tools or skills myself to attempt this repair.

A hardware test was performed at an official Apple Store, and it didn't reveal any malfunction. That leads me to think the problem is probably low-level.

One possible solution
On computers that don't have a removable main battery and no button cell battery, there's often a supercapacitor holding CMOS and RTC data.

The other option would be to replace this supercapacitor powering the MBP A1278 by a standard 3V battery to power the PRAM and SMC, and add a physical switch to disconnect the main battery while hibernating. Sounds like a less risky alternative as only two solder points on the board would be required.

Questions
The question would be: where is the supercap holding the charge for the PRAM and SMC on a Late 2011 MBP?
Or, what other solutions would be feasible on this computer?

Re: Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Google Adsense [BOT] 4 stycznia 2016, 18:50

#2 Re: Apple MacBook Pro A1278 Paradise chip stays hot during sleep, MB after spillage


przez Vogelek23 4 stycznia 2016, 22:41
Cubytus napisał(a):The other option would be to replace this supercapacitor powering the MBP A1278 by a standard 3V battery to power the PRAM and SMC
This Macbook has no CMOS battery nor supercapacitor. RTC generator is powering from a U2800 SLG3NB148V IC (pin 1). If you disconnect the battery and PSU, CMOS memory is cleared.

Cubytus napisał(a):an unknown chip (probably related to Thunderbolt and HDMI functions) stays active and hot during sleep.
It is a PS8301 Display Port super driver (Parade). This IC can be supplied in S3 state (suspend to RAM), but it shouldn't stay hot even during the normal operation. Otherwise, it needs to be replaced. You can also measure the voltages on all pins of Q7830 and check for possible short (drain-source).

Cubytus napisał(a):According to the only shop that does this kind of repair in this city, replacing the burned chip is just too dangerous for the computer to be worth it
It seems that you have gone to the bad shop. It is not dangerous, but easy for any person who knows about the hot-air soldering. Where do you come from, mate?

#3 Re: Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Cubytus 5 stycznia 2016, 01:31
So I assume that RTC drain on the main battery is so low that Apple didn't bother including a backup power for it?

Would you care to post a picture of where I could find that U2800 chip on the mainboard? I am currently typing this post on the faulty computer and obviously can't look now.

Almost all "repair" shops here don't do any actual repair, just replacement of faulty parts (Montreal, QC). This particular shop does hot air soldering, but only if it were the only way to save an otherwise dead computer.

#4 Re: Apple MacBook Pro A1278 Paradise chip stays hot during sleep, MB after spillage


przez Cubytus 9 stycznia 2016, 21:14
Vogelek23 napisał(a):RTC generator is powering from a U2800 SLG3NB148V IC (pin 1). If you disconnect the battery and PSU, CMOS memory is cleared.

You can also measure the voltages on all pins of Q7830 and check for possible short (drain-source).

Can you point me to a picture or schematic where I could find those two chips on the board?

#5 Re: Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Vogelek23 18 stycznia 2016, 22:14
Cubytus napisał(a):Can you point me to a picture or schematic where I could find those two chips on the board?
Here you have a full documentation: schemat-apple-macbook-pro-a1278-k90i-820-2936-051-8658-t15431.html

U2800:
Obrazek

Q7830:
Obrazek

#6 Re: Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Cubytus 31 stycznia 2016, 21:23
Ok, I was able to test Q7830 as it was on the exposed side of the motherboard. However, it looked a bit different from what the schematic led me to believe. I find about 3V on top and bottom rightmost pins, but no short between them. The leftmost pins are 0V. That is, with battery connected, but computer turned off.

What should I be measuring on pin 1 of U2800?

#7 Re: Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Cubytus 30 marca 2016, 07:14
Update:

Powering U2800 from a 3V coin cell battery
Goal: Not lose time and date if main battery is drained.

I couldn't find a datasheet for SLG3NB148V (U2800), but there was one for SLG3NB ("notebook clock"), with differing numbers at the end. All are 16-pin TQFN packages, and pin 1, on the similar SLG3NB, is VDD_RTC_OUT, or Power Output. This is confirmed by the schematic of board 820-2936-B board I found.
From what I understand, this chip is the clock signal generator that feeds a 32.768KHz signal to the actual RTC.
And VDD_RTC_OUT feeds the RTC with power?

Interestingly enough, datasheets from similar parts say:
SLG3NB120 supports Non-rechargeable CR coin cell batteries

Integrated battery supply switch

Source: datasheet for similar parts.

Then further down the datasheet, I find this:
Obrazek

The following is a capture from an Apple schematic I found:
Obrazek
VBAT is pin 13 on U2800, and indeed, they have at least though about using a coin battery to feed U2800.

Can I safely assume it is possible to power the U2800 clock generator from a 3V CR2032 if all the following circumstances are reunited:
1- I cut the PP3V42_G3H line going to pin 13 of U2800 so as not to give current to a non-rechargeable battery
2- I solder a wire on the newly-cut PP3V42_G3H on U2800 / pin5 side, then connecting it to the CR2032 battery as shown in Silego's datasheet
3- The clock generator U2800 would be powered by the coin battery when computer is off or sleeping, and from the PP3V3_S5 line when computer is active
4- U2800 would make the switch automatically between coin cell battery on pin 13, and PP3V3_S5 on pin 2?
5- pin 1 would be 3.0V on the coin cell battery, and 3.3V on computer power

One problem remains: even if I keep RTC and its clock generator active, would the the other settings be kept?

Testing U2800
To go back to the question you asked about U2800, I saw this:
Obrazek
You asked me to test pin 1 on U2800, but this chip is very difficult to reach as it needs to be tested on a completely working machine and it's on the other side of the motherboard, while I can't use an external display precisely because of PS8301 failing. Please understand testing this chip is a very delicate proposition and would require much time and patience.

From your hypothesis, pin 1 should measure 0 VDC when main battery is exhausted due to PS8301 staying active, and 3.3VDC when main battery is good?

Turning off PS8301
Goal: turning off the faulty chip appears less risky when no hot-air rework stations or skills are available.

Theoretically, would it be possible to turn off PS8301 manually during sleep so as to conserve battery?
Would it crash the computer on wake up?

#8 Re: Apple MacBook Pro A1278 Parade chip stays hot during sleep, MB after spillage


przez Cubytus 29 sierpnia 2016, 19:56
I have finally tested U2800 under battery power:

It was rather difficult to keep the motherboard sufficiently exposed for testing under the binocular miscroscope, and keep it plugged in at the same time. I extended the test leads from the multimeter (set in DCV) with sewing needles.
Datasheet of a chip from the same family (SLG3NB148V has no publicly available datasheet AFAIK): http://www.datasheetlib.com/datasheet/1210772/slg3nb114_silego-technology.html

Obrazek

Pin 1 VDD_RTC_OUT: 3,09V
Pin 2 +V3.3A: 3,09V
Pin 3 X2: 0,9V
Pin 4 X1: 0,2V
Pin 5 VDD: 3,02V
Pin 6 VDDIO_25M_B[: 3,02V
Pin 7 GND: 0V
Pin 8 25MHz_B: 1,7V
Pin 9 25MHz_A: 0,7V
Pin 10 GND: 0V
Pin 11 VDDIO_25M_A: 1,7V
Pin 12 32KHz_A: 0,41V
Pin 13 VBAT: 3,3V
Pin 14 VDDIO_32K_B: 2,2V
Pin 15 32KHz_B: 1,8V
Pin 16 GND: 0V

What about the possibility of powering U2800 through a coin-cell battery? It seems there's an integrated battery supply switch, so if main power is lost, U2800 would stay active using the coin cell.

If yes, what technique should I use to solder anything on the extremely small pin13?

To jest wątek ARCHIWALNY. Możesz odpowiedzieć w nim tylko wtedy, gdy Twój post zawiera rozwiązanie problemu (Regulamin forum p. 12.1).

Kto przegląda forum

Użytkownicy przeglądający ten dział: Brak zidentyfikowanych użytkowników i 3 gości

_______________________________
Wszelkie prawa zastrzeżone. Zabrania się kopiowania jakichkolwiek treści i elementów witryny bez zezwolenia.
Wszelkie opublikowane na tej stronie znaki handlowe, nazwy marek, produktów czy usług należą do ich prawnych właścicieli i zostały użyte wyłącznie w celach informacyjnych.