A1707 Speakers Buzzing

[Vonevoe]

Hello,

First of all, I'm not a native English speaker, so I'll try my best to be understood. I am a first-year electronics student eager to learn, so I bought a MacBook A1707 and tried to fix it.

This MacBook had water damage.

The left USB-C ports were affected by water, so I replaced them. Both ICs 3100 and 3200 didn't work as expected, so I changed them. The battery was dead, so I bought a new one and replaced it.

Now, the MacBook works perfectly, except for one issue.

When I connect the battery and plug in the USB-C, a few seconds later, both speakers start buzzing for 1 or 2 seconds.

My Research:

I found the board view and the schematic of the board.

I tested the voltage on all the lines connected to J6410 and J6500 (then U6400) with the following procedure:

1. Unplug the battery and the USB charger.
2. Plug in the battery, then the charger.
3. Connect the black lead to the ground and the red lead to the line I want to measure.
4. Wait for the first fans to start spinning.
5. Then, wait for the second fans to start spinning (when the speakers are buzzing).Repeat the process.

The voltage on SPKRCONN_WL_OUT_NEG, SPKRCONN_TL_OUT_NEG, and SPKRCONN_TR_OUT_NEG dropped from 0.120V to 11.80V, then back to 0.120V.

Only SPKRCONN_WR_OUT goes from 0.130V to 0V.

For U6400:

• SPKRAMP_WL_OUT_NEG and SPKRAMP_TR_OUT_NEG obviously 0.120V to 11.80V, then back to 0.120V.
• PP1V8 is normal
• AUD_I2C_1A_SCL: 1.795V
• SPKRAMP_WL_LRCLK: 0V -> 0.380V -> 0V
• SPKRAMP_WL_OUT_POS: 0.110V
• AUD_SPRKRAMP_INT_L: 0.015 -> 0.010
• SPKRAMP_WL_SDIN: 1.78V -> 1.5V -> 1.78V
• SPKRCONN_WL_OUT_POS: 0.130V
• VREF_AMP_WL: 0V -> 2V -> 0V
• AUD_SPKRAMP_RESET_L: 0.030V -> 1.8V -> 0V
• SPRKRAMP_WL_SCLK: 0.050V -> 0.875V

I tried to find the MAX98706 datasheet to trace back to the source of the problem, but I couldn't find it.

I don't have another 820-00281-A logic board for testing.

I don't want to give up, but I'm unsure where to continue to understand the problem.

So here I am, asking for help.

 

[2informaticos]

First of all, welcome to the forum!

I suspect you still have components affected by corrosion.

This board has 4 speakers; two on each side.
Every MAX98706 drives one speaker.

"When I connect the battery and plug in the USB-C, a few seconds later, both speakers start buzzing for 1 or 2 seconds"
Do you have diferent behavior without battery connected (only charger)?
Buzzing sound disappears after 2s?
Once logged into macOS, audio codec is recognized?

 

[Vonevoe]

First of all, welcome to the forum!

Thank you! I'm happy to be here.

This board has 4 speakers; two on each side.
Every MAX98706 drives one speaker.

This MacBook model has four speakers, with two located on each side. Each speaker is powered by a MAX98706 chip. Therefore, in reality, what is referred to as a single speaker in this model actually consists of two separate speakers ?

Do you have diferent behavior without battery connected (only charger)?

I'm unable to turn on the MacBook when the battery is unplugged. However, I did manage to turn on the MacBook and then unplug the battery (refer to the "additional observations" section below).

Once logged into macOS, audio codec is recognized?

Yes indeed.

Buzzing sound disappears after 2s?

I've uploaded a private video showcasing the issue. (Note: YouTube required me to format the video as a short.)

Additional Observations:
It appears that the buzzing noise replaces the usual startup chime ("dong!"). When I attempted to play music or any audio, the speaker responded with a buzzing sound similar to what's in the video, although not as loud.
Additionally, when I unplugged the battery, the speaker buzzed for a fraction of a second.

I initially thought this issue was unrelated and merely a result of 'flexgate', but now I'm starting to question that assumption.


Sometimes the display band flickers slightly, or sometimes it even disappears completely.

Thank you for your help

 

[2informaticos]

Mac detects changes between speaker and headphones?
Do you get correct sound on headphones?

 

[Vonevoe]

I couldn't try with headpone but i use a external speaker (bose)
Mac detect change in both way.

I get correct sound in external speaker.

 

[2informaticos]

Inspect all audio amplifiers for corrosion; pages 60/61 in schematic.

 

[Vonevoe]

There's nothing,
those chips are indeed the MAX98706. I attempted to locate the datasheet. Should I consider replacing all of them?

 

[2informaticos]

I doubt all the speakers to be damaged.
But just in case, test with known good speakers.

When you play video/audio, speakers buzzing continously?

Check diode mode to ground for AUD_SPKRAMP_RESET_L; also the voltage.

 

[Vonevoe]

I doubt all the speakers to be damaged.
But just in case, test with known good speakers.

When you play video/audio, speakers buzzing continously?

Check diode mode to ground for AUD_SPKRAMP_RESET_L; also the voltage.

These speakers are confirmed to be in working condition. I experimented with two sets of brand-new ones.

I don't completely comprehend which parameter affects the diode mode reading's outcome yet, so I will provide a comprehensive description of the measurement procedure.

I tested it with only the logic board, without any additional components connected (no power supply, no speakers, etc.). I placed the red probe on ground (GND) and the black probe on TP_AUD_SPKRAMP_RESET_L, resulting in a reading of 0.333 V.

I tested the voltage in the same environment, with the only difference being the connection of J3300 connector to the USB-C port and a USB-C charger. Additionally, I switched the positions of both probes, resulting in the following readings:

0.033 V -> 1.798 V -> 0.033 V.

Upon retesting, the speakers emitted some faint noise, and it appears that adjusting the volume did not have any noticeable impact. Notably, the speakers became very hot when I left the Mac on for a few minutes, and there was a noticeable odor of burning.

 

[2informaticos]

Do you get 0.333, or 0.033 on reset line (diode mode)?
BTW, do not post "V" after diode mode readings.

Pleae, don't quote an entire post, visible few centimeters above.

 

[Vonevoe]

I obtained a diode mode reading of 0.333 on TP_AUD_SPKRAMP_RESET_L pin.

Tomorrow, some friends of mine will bring me a functioning board for testing diode and voltage.
I'll attempt to measure as many lines as possible.


I appreciate your assistance and advice.

 

[Vonevoe]

I'm not entirely sure how to interpret the diode mode measurements, but I've been studying the schematic. I've tested every diode mode on all of the MAX98706 chips, and the readings appear to be perfect. I'm now contemplating whether the HDA Bridge (U6200) could be the issue here and whether it should be replaced.

Does this plan sound reasonable to you?

 

[2informaticos]

Try changing it...

 

[Vonevoe]

I stumbled upon something unusual by accident. I wanted to check if my two multimeters were providing the same reading, so I tested them on the same location. Then, I measured the same spot on another board (working board) and discovered that the diode mode measurement was different.

	Working Board	Problematic Board
R3550
(1) P5VUSBCX_R	0.175	0.275
(2) PPV5V_S4_X_USBC	0.175	0.275
(3) P5VUSBCX_P	0.175	0.275
(4) P5VUSBCX_N	0.175	0.275

(1) Connected to an inductor, then linked to Q3501 and U3500.

(2) Linked to U3200 and U3100 (currently testing all pins in diode mode).

(3) Connected to R3521 and then linked to U3500 (diode mode measurement between U3500 and R3521 is consistent on both the working board and the problematic board).

(4) Connected to R3572 and then linked to U3500 (diode mode measurement between U3500 and R3572 is consistent on both the working board and the problematic board).

I have tested all the pins of Q3501 and U3500, and there are only two lines that show a difference:

	Working Board	Problematic Board
PP5VUSBCX_LL	0.175	0.275
PP5VUSBCX_DRVH	0.600	0.700

Both of these lines are connected between Q3501 and U3500.
(to be continued)

 

[Vonevoe]

To summarize the story,
I tested all the pins of U3200 and U3100, and I noticed that all the incorrect diode mode measurements were directly or indirectly related to Q301 and U3500. I attempted to replace Q3501, but it didn't change anything.

Then, I decided to replace U3500, and I obtained better results.

Signal	Before replacement	Expected Value	After repalcement
PP5VUSBCX_LL	273	170	200
PP5VUSBCX_DRVH	700	600	633

How much of a difference in diode mode measurements between a working motherboard and a faulty one would indicate that something is wrong?

However, the issue with the speakers was still unresolved. Therefore, I decided to test my initial hypothesis.

I replaced U6400 and U6450, but I believe I soldered U6400 incorrectly. When I turned on the Mac, I no longer had 20V. I found a short circuit on PP3V3_G3H at U8510. I replaced this chip, which restored the 20V, but the Mac still wouldn't boot.

I noticed that I didn't have the (S5 enables) signal. I examined the signal source and the necessary signals, and I found a short circuit on PP3V0_GH3. I removed U7800, but the short circuit was still present, so I identified that U1100 was also shorted.

Unfortunately, I can't proceed further because I lack the expertise to salvage the motherboard, which is now completely unresponsive. It's entirely my fault, but this learning process has been valuable.

Ty for the help