Announcement

**mon2** · 11-10-2021, 05:49 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi. The 69 ohms is not considered a short.

Low voltage rails are often low in resistance. Based on your meter readings, you do not have a short so voltage injection is not (yet) recommended.

Each and every coil should have the resistance checked against ground. I think you have done this.

In your area, you have a resident guru on macbook repair:

Paul Daniels
https://www.rossmanngroup.com/macboo...-in-australia/

If you wish, you can explore to troubleshoot via this forum or simply contact Paul. He has many videos on YouTube and is also on this forum.

**mon2** · 11-10-2021, 05:58 AM

Re: Mid 2015 15" Retina Pro green light but won't start

From @Piernov's logiwiki website about your logic board:

These boards absolutely require a good charged battery to fully boot into macOS. As usual it's not uncommon to get 2 or 3 bad batteries in a row so make sure you are using a known good one, one that for sure works with others 15" 2015 DG boards.

With a bad battery or no battery, the machine will reboot during or at the end of macOS boot. In this case disconnecting the touchpad often allows it to boot fully since it causes throttling and lowers the overall power consumption of the machine.

**DadBuk** · 11-10-2021, 07:55 PM

Re: Mid 2015 15" Retina Pro green light but won't start

Thankyou for taking the time to reply mon2.

Since starting to look at this board I have been watching a lot of MacBook repair YouTube channels including Paul Daniels - both entertaining and informative.

Due to the cost, for this board I am not planning to send it out again for professional work just in case the first technician was correct and it is not fixable - but am finding the troubleshooting process interesting so will see how far I can get with some more research and some advice from this forum.

The battery is due for replacement as is puffy so if I were to have any success with the pulsing rails I will be replacing.

If I am following correctly, the pulsing rails indicate an issue somewhere on the board, is that limited to a short somewhere or could there be many causes? If I cannot locate a short/partial short, what else should I look for? I have tried checking some of the lines that enable the pulsing rails but they are present and pulsing so maybe not the correct thing to focus on.

I had checked all of the rails for shorts, except PPBUS_S4_TPAD and PPVIN_S4_TPAD as I cannot find them but am also now checking each coil which may not be necessary given previous checks but is useful in terms of learning how the boards is laid out.

Thanks

**DadBuk** · 11-12-2021, 03:45 AM

Re: Mid 2015 15" Retina Pro green light but won't start

I checked all the coils and the readings were consistent with the rails measurements except for L6604/L6605, L6611/L6614 being 0Ω and L6606/L6607 being OL - but that is all audio stuff so I am guessing it is not related to the cycling power rails and the coils are more to do with audio. Any suggestions on where to look next would be appreciated.

Thanks

**mon2** · 11-12-2021, 08:16 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi. For the pulsing rails, check that the current sense resistors and the feedback for the INPUT / OUTPUT for the current sense resistor are making it back to the local power supply controller.

As per another post a few minutes ago, the local power supply controller must be able to detect the INPUT voltage and then is looking for a minute voltage drop across the current sense resistor as it leaves the resistor. If that leg is OPEN, as an example, the local power supply will shut down the power supply. You may have corrosion on that area so use alcohol to clean it up. If the leg of the current sense circuit is OPEN, the power rail is thought of as being under too much of a load and the power supply will enter 'chirp mode' = rail will pulse ON / OFF for safety reasons.

**DadBuk** · 11-21-2021, 04:15 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Thanks Mon2.

At the moment I am having problems with my USB microscope - it is one of those converted Microsoft LifeCams - so gives a reasonable view of the components but the driver has started to crash after about 15 secs making things difficult. For the moment I am using a Supereyes usb microscope and have a trinocular microscope on the way.

In the meantime - I have been looking through the schematic and there are many sense circuits, most not on the main power rails I think. Were there specific sense circuits that you were referring to above.

I was reading an article about the PM_SLP_S4_L 11+1 prerequisites (https://www.it-tech.com.au/mac-q-a/pm-slp-s4-l-timing/). This board has all of these but PM_SLP_S4_L is not constant. From the oscilloscope I can see that it pulses to 3.3 volts for milliseconds which per Case Study 4 of the above it-tech article could mean that the PCH/CPU is ok. That case study goes on to talk about the SPI chip and corroded components. Another article I read talks about checking the SPI termination resistors. I checked these areas and cannot see anything visually.

This board had a small amount of water on it and the battery was disconnected after about 1 hour. When I opened it I could see some 'bubbling' around Q7630 but after that it went to the repair shop and they put it through the ultrasonic. So at the moment, it looks very clean so I cannot locate anything visually that looks suspect. I think that the repair shop did some work on Q7630 and J9510 per the attached pics.

So I am rambling a bit at the moment - I have been reading a lot of articles and looking at a lot of youtube repair videos about pulsing rails but just can't put it all together. What are the likely causes of the pulsing rails. Does it make sense to see which pgood signals are either present or at least pulsing to try and work out which rails are good and which rail or area it is that is cause the pulsing.

Hope I am making sense.

Thanks

Paul

Attached Files

**mon2** · 11-21-2021, 07:32 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi. Unable to fetch the schematics to review at this time - I think the server is down for the sub we have.

In the meantime, you can power down. Meter in resistance mode and check if the mosfet that you suspect is defective or not.

Review this great webpage and the table on the same:

http://electronicsbeliever.com/how-t...0good%20MOSFET.

You are checking for the resistance between:

source & drain
source & gate
gate & drain

Often, on such 8 pin packages - the source = pins 1-2-3. Gate = pin 4. Drain = 5-6-7-8.

The pulsing is most likely due to some power rail turning ON and then OFF as it believes the rail is under too much load.

For this reason, check if the coils (each one) is having a too low resistance to ground. I think you have done this already. The low voltage CPU / GPU rails are expected to be low but that is normal for low voltage rails.

Was going to suggest to find the main PPBUS_G3H rail which often features a fuse. Remove that fuse and leave the fuse pads open. Then power on.

Does PPBUS_G3H remain enabled and is supplying the proper voltage without pulsing?

**DadBuk** · 11-23-2021, 06:12 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi, just an update.

Q7630 does seem to be defective. It is a dual N channel power mosfet. First channel has a voltage drop of .454v D-S and open S-D and high resistance . Second channel has voltage drop .053v D-S and .039v S-D and 60-70 ohm resistance both ways.

So I guess next steps would be to source a replacement for it and U7600 as I read somewhere a damaged mosfet can damage the IC. Q7630 is a NTMFD4951NF but that part number does not come up to buy. I read that any 2 channel mosfet with the same form factor would be ok (cannot find that post now atm) but ISL95870 seems just as hard to come by.

Other option I am following up on is to get a suitably priced donor board that has both components so I am keeping an eye on FB Market Place and ebay. Once I source something and hopefully manage to replace them I will post an update.

Paul

**mon2** · 11-23-2021, 06:39 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi. My subscription schematic database was down. Their server was recently fixed.

1) Meter in DC volts. When power is on with adapter, what is the voltage reading at the following points ?

a) L7630, pin # 2 = PP1V05_S0_REG_R (this should be 1.05V)

b) same rail but measure after the current sense resistor (R7640) = PP1V05_S0 (should be still about 1.05V)

c) PPVIN_S5_HS_COMPUTING_ISNS (feed into dual mosfet Q7630)

d) P1V05S0_DRVH = high side mosfet driver voltage

e) P1V05S0_DRVL = low side mosfet driver voltage

For now, do not consider this as a defective mosfet till more measurements are known. For low voltage rails, it is normal to have low resistance on such rails. Just not super low like < 10 ohms but best to take and post the above measurements.

If you find that the 1.05V rail is not working, power down and remove adapter & battery. Meter in DC ohms scale (lowest setting). Then red probe on PP1V05_S0 and black meter probe to ground.

What is the resistance ? This test should be able to confirm if the CPU is possibly dead due to a very low resistance to ground.

**DadBuk** · 11-23-2021, 06:12 PM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi,
those measures are:

a) L7630 pin 2 PP1V05_S0_REG_R = 1.04v pulsing to 0v

b) after the current sense resistor (R7640) = 1.04v pulsing to 0v

c/d/e)
Pin 1 (G1) P1V05S0_DRVH = 1.05v pulsing to 0v
Pins 2, 3, 4, 9 (D1) PPVIN_S5_HS_COMPUTING_ISNS = 12.59v constant
Pins 5, 6, 7 (S2) GND = 0v
Pin 8 (G2) P1V05S0_DRVL = .22v pulsing to 0
Pin 10 (S1/D2) PP1V05S0_LL = 1.04 pulsing to 0

PP1V05_S0 to ground = 64ohm

Thanks

Paul

**DadBuk** · 11-28-2021, 06:22 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi, further to my update above, I was going over this startup sequence https://logi.wiki/index.php/G3_to_S3..._and_Broadwell to see where this board gets to. Everything checks out down to PM_SLP_S5_L which is pulsing 3.3v - 0.v as is PM_SLP_S4_L. As mentioned above all the power rails from S4 down are also pulsing.

Any ideas on what else I could check on this board.

Thanks

**mon2** · 11-28-2021, 07:21 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Check and post the voltages on the SPI Rom during the power cycling.

**DadBuk** · 11-28-2021, 07:06 PM

Re: Mid 2015 15" Retina Pro green light but won't start

Hi,

Here are the measurements for U6100 and U6101. Thanks

U6100
1 SPI_MLBROM_CS_L = 3.32
2 SPI_MLB_IO1_MISO = 3.31
3 SPI_MLB_IO2_WP_L = 3.32
4 GND
5 SPI_MLB_IO0_MOSI = 3.31 / 0
6 SPI_MLB_CLK = 3.31 / 0
7 SPI_MLB_IO3_HOLD_L = 3.32 / 0
8 PP3V3_SUS = 3.31
9 GND

U6101
1 GND
2 PP3V3_SUS = 3.32
3 SPI_MLB_CS_L = 3.28
4 GND
5 SPIROM_USE_MLB = 3.28
6 GND
7 SPI_MLBROM_CS_L = 3.32
8 PP3V3_SUS = 3.32

**mon2** · 11-28-2021, 09:13 PM

Re: Mid 2015 15" Retina Pro green light but won't start

Do you not see voltage oscillations on pins 1&2 of the SPI flash rom with your scope?

I would expect that MISO pin, at least, should show this pattern.

MISO = master in (CPU), slave out (SPI flash rom).

What type of scope do you have? Can it perform logic analyzer functions like SPI decoding?

On first use, the host will send out SPI commands to the flash device to extract ID or perform reads from addresses, etc.

If corrosion has halted the communication lines between the parts, this can cause the pulsing of the logic board.

Consider to perform resistance checks for each pin of the SPI flash to the mating pins for the component to other parts.

**DadBuk** · 12-31-2021, 06:13 AM

Re: Mid 2015 15" Retina Pro green light but won't start

Back to this board again.

Not sure whether there are voltage oscillations on pins 1 & 2 of U6100 - there is a spike each time the board cycles - I have attached 3 screen prints for pins 1 - 3 which all look similar. My scope has SPI but is 2 channel so can only decode one at a time and cannot use CS. I set it up on pin 2 SPI_MLB_IO1_MISO on channel 1 and pin 6 SPI_MLB_CLK on channel 2 (first time doing this so hoping that is correct) with a timeout of 2.2us but there is no decoding - just a flat line on MISO[HEX] and some spiking on the 2 channels, I guess due to the clk signal cycling. Not sure if that is due to their being no comms on MISO or more likely my setup so I will do some more research on how to do it.

Is it odd that on U6100 that SPI_MLB_IO1_MISO is not power cycling but SPI_MLB_IO0_MOSI is.

I had a close look at the traces around U6100 and there is no corrosion and there is continuity to the components around it.

I have been checking the PG signals and most are either cycling or constant - but CPU_PWRGD is 0v - with nothing at all even on the scope from power on. I am not able to find any info that says exactly what this means - ie, is it indicating that vcore is not stable which we already know.

Any feedback is appreciated. Thanks

Attached Files

Announcement

Mid 2015 15" Retina Pro green light but won't start

Mid 2015 15" Retina Pro green light but won't start

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment