A2337 2020 MacBook Air

Water will get to the audio jack board and fry the audio IC if still no sound after replacing audio board check SPKRAMP_1V8_RESET_L if zero then UR900 is suspect. Run wire from CR901 pin 2 that has 1.8 volts to pin 1 of CE904 this will force SPKRAMP_1V8_RESET_L high and allow speakers to power up. As of writing I could not find UR900 (SN74AVC4T234) for purchase it is recommended you replace UR900 but this workaround works fine so far assuming UR900 is being told to turn on and SPKRAMP_RESET_L is present.

• The most common cause for an 820-02016 to be stuck on 5v and 0.00 to 0.02 is PPBUS_AON voltage low or missing.

• Always check that both USB-C ports are reading the same. One port reading 5v and 0.10 to 0.25 while the other reads 5v and 0.00 to 0.02 is more indicative of a CD3217 issue than a issue with PPBUS_AON or any of the CD3217s support power rails.

Differential Diagnosis for 5v and very low amp draw (0.00 to 0.02 amps) on the 820-02016 logic board.

• PPBUS_AON short, low or missing.

• Device stuck in DFU mode due to corrupt firmware.

• Missing or shorted CD3217 support voltage such as PP3v3_S2_UPC, PP5v_S2 or PP1v25_S2.

• Corroded CD3217, ISL9240, or other non specific areas of corrosion that impedes communication between PMICs and SOC and SOC and other components.

➤Diagnostic Steps

1. Check if the device is stuck in DFU or recovery mode.

- Plug the MacBook into another Mac or MacBook via its master port (Left side, top port, closest to the display) and open Apple Configurator 2 to verify that it is or isn't in DFU mode. If the device is in DFU mode, Proceed to "Device stuck in DFU mode due to corrupt firmware" in repair steps below.

2. Check voltage on PPBUS_AON.

To check voltage on PPBUS_AON, place your multimeter in DC mode, and place the black probe on ground, and the red probe on PPBUS_AON. A good spot to measure from on the 820-02016 logic board is C5960.

~ 12.60 - 13.1v = Normal

~ 12.28-12.35 = Possibly normal, however may indicate a communication issue between the SOC and one of the PMICs or between the SOC and other components as well as corrupt firmware.

~ 0V = Short to ground or absent voltage due to a creation problem. Proceed to the repair steps below for shorted or absent PPBUS_AON voltage.

3. Check PP3v8_AON_VDDMAIN voltage.

To check voltage on PP3v8_AON_VDDMAIN, place your multimeter in DC mode, and place the black probe on ground, and the red probe on PP3v8_AON_VDDMAIN . A good spot to measure from on the 820-02016 logic board is C5887 or C5889.

Normal voltage = ~3.8v.

If voltage is missing or low, proceed to the PP3v8_AON_VDDMAIN missing or low repair steps below.

4. Check CD3217 support voltages.

- PP5v_S2 - Normal voltage =~5v. Recommended spot to measure from is CC334 or CC335.

- PP3v3_S2_UPC - Normal voltage =~3.3v. Recommended spot to measure from is CF400 or CF500.

- PP1v25_S2 - Normal voltage =~1.25v. Recommended spot to measure from is SWV011 (Pin 1)

If any of the S2 rails above are seen missing, proceed to CD3217 support S2 rails low or missing repair steps below.

If all above findings are negative for any issue, one or both of the CD3217s may have failed or may have a surrounding circuit issue such as a broken trace or blown resistor. A firmware issue is also possible, but unlikely if the device is not in DFU or recovery mode.

CD3217 Diagnostic Process

- There is no definitive diagnostic test to confirm that a CD3217 is the cause of an issue a board is experiencing other than replacing it, however some findings can suggest a fault with one or both of the CD3217s.

You can determine which CD3217 is in use by measuring voltage on the input fuse that correlates with the chip.

FF201 correlates with UF500 FF200 correlates with UF400.

Test 1. - Measure PPDCIN_AON_CHGR_R on both ports to determine if the internal MOSFET in one CD3217 is not allowing charger voltage through to the rest of the system. If PPDCIN_AON_CHGR_R is present on one port, but not the other, the CD3217 that is not allowing voltage though is the likely cause of the issue. "

Test 2. Measure the LDO outputs of both CD3217s to determine if one of the ICs isn't outputting proper voltage to its LDO lines. The LDO lines from the CD3217s power internal portions of the chip as well as power the TBT ROM, which is needed for proper function of both CD3217s.

PP3v3_UPC0_LDO - Supplied from UF400 and powers the TBT ROM (UF260). Normal voltage = ~3.3v. Normal diode mode to ground reading = ~.500 .

PP1v5_UPC0_LDO_CORE - Supplied from UF400 and powers internal CD3217 functions. Normal voltage = ~1.5v. Normal diode mode to ground reading = ~.510

PP3v3_UPC1_LDO - Supplied from UF500 and powers internal CD3217 functions. Normal voltage = ~3.3v. Normal diode mode to ground reading =~.500.

PP1v5_UPC1_LDO_CORE - Supplied from UF500 and powers internal CD3217 functions. Normal voltage = ~1.5v. Normal diode mode to ground reading = ~.510.

If you find that any of the LDO voltages are low, check for a short to ground on the correlating line. If no short to ground is found, replace the CD3217 that the line correlates with.

➤Repair Steps

Device stuck in DFU mode due to corrupt firmware: - Revive firmware via Apple Configurator 2.

NOTE: You must be running the latest version of MacOS for this to work consistently. Check for MacOS updates prior to reviving/restoring firmware.

How to revive firmware: Plug the device you are working on to another Mac or MacBook via its master port. The master port on the A2237 is the top left side USB-C port (closest to the display.) If you are confused, please see the diagram listed on Apple's how to page - [1]

Once plugged in, open Apple Configurator 2. You should see a big square icon pop up that says "DFU" or rarely, "RECOVERY". Click the icon, Navigate to the top menu bar click "Actions" then "Advanced". Select Revive device. You will see a progress bar appear. This process can take anywhere from 2 minutes to over 30 minutes in some cases.

SELECTING RESTORE WILL WIPE ALL USER DATA!

PPBUS_AON shorted, low or absent:

PPBUS_AON shorted to ground:

Inject around 1 V with a 5 amp limit into PPBUS_AON with a DC power supply. A larger tantalum capacitor is preferred to inject voltage to, as you will often need a lot of current to be pushed into the line to cause the shorted component to heat up. 1 V is usually sufficient. C5960 is a good place to inject voltage to on the 820-02016 logic board.

With voltage being injected, perform thermal imaging of the board. If thermal imaging is not available, feel around the board to see where it is getting warm. Once the area is localized, add a small amount of isopropyl alcohol to the area to localize the shorted component.

Once the shorted component is localized, replace the shorted component.

For more info on finding short circuits, please see - [2]

PPBUS_AON absent or low with no short:

PPBUS_AON is created by the Intersil/Renesas ISL9240 (U5200) which is a buck or boost converter depending on its input voltage. If USB-C voltage is 5 V, the ISL9240 will boost the 5 V input to ≈12 V. If USB-C voltage is 20 V, the ISL9240 will buck (lower) the 20 V input to ≈12 V.

When PPBUS_AON is absent with no short to ground, we need to first make sure that PPDCIN_AON_GHGR_R is making it to the ISL9240 (U5200). C5201 or C5202 is a good place to measure from. Measure PPDCIN_AON_CHGR_R with your multimeter in voltage mode. You should get the same voltage that you are getting on the USB-C amp meter (5 V in this case, because that is what we are troubleshooting.) If you get 0v on PPDCIN_AON_CHGR_R, check R5220, however it must be noted that if you have 0v here, one of the CD3217s is probably bad. See "CD3217 Diagnostic Steps" above.

If you are getting 5 V on PPDCIN_AON_CHGR, the ISL9240 is likely the cause of the problem, however we still need to check a few things before replacing the chip. The ISL9240 relies on a series of current sensing resistors to measure the amount of power being used by its output rail, PPBUS_AON. If one of these resistors is blown, the chip will think that something on the output is pulling too much power, and it will disable itself as a protection measure. To check the current sensing resistors, put your multimeter on resistance mode. Check each resistor listed below by placing your probes on each side of the resistors. The values should be within 5–10% of the reference listed below. Values slightly outside of the reference should be considered insignificant and ignored. If you find a resistor out of spec, replace the resistor AND the ISL9240, as the resistor likely blew as a result of the ISL9240 internally shorting.

R5221 — 1.00 Ω

R5222 — 1.00 Ω

R5261 — 1.00 Ω

R5262 — 1.00 Ω

If the current sensing values are normal, replace the ISL9240.

PP3v8_AON_VDDMAIN low or missing

- Step 1. Check for a short to ground on PP3v8_AON_VDDMAIN. C5887 is a good place to measure from. Normal diode mode to ground reading = ~.350.

If a short to ground is identified, find and replace the shorted component. For more information on how to find short circuits, please see this page - [3]

-Step 2. Check the enable signal to U5700, P3V8AON_PWR_EN (~5v)

P3V8_PWR_EN is sourced from CHGR_EN_MVR from the ISL9240 (U5200) and passes through a delay switch (U5340) where it is outputted as P3V8_PWR_EN.

If P3V8_PWR_EN is missing, check voltage on R5340 (CHGR_EN_MVR). If CHGR_EN_MVR is missing, replace U5200.

If CHGR_EN_MVR is present, but P3V8_PWR_EN is missing, potential faults could include U5340, or surrounding traces/resistors.

- Step 3. Ensure that the VIN (PPBUS_AON) is making to U5700. Occasionally a via or trace can break resulting in a voltage not making it to some parts of the board. PPBUS_AON can be measured adjacent to U5700, on C5704 or C5804.

If all above findings are negative for issues, replace U5700.

CD3217 support S2 rails low or missing

In most causes, low or missing voltage on S2 rails will be due to a short to ground, or an enable issue.

PP5v_S2 - Produced from UC300.

Step 1. Check for a short to ground. The vast majority of cases where PP5v_S2 is missing will be due to a short to ground.

If a short to ground is found, find and replace the shorted component. For more information on finding short circuits, please see - [4]

If no short to ground is found, proceed to the next step.

Step 2. Check the enable signal for PP5v_S2, P5VS2_EN.

P5VS2_EN is produced by the PMU, U8100.

A missing enable signal can mean U8100 is bad, or there is a communication issue with U8100 and the SOC/other components. A firmware issue can also * *likely* cause missing enable signals from PMICs on the M1 machines. As of 1/14/2022, this remains rare.

Step 3. Check the VIN voltage to UC300, PPBUS_5VS2_VIN.

The VIN voltage listed above is sourced from PPBUS_AON and voltage measured on it should remain consistent with PPBUS_AON voltage. If voltage is low, possible causes could include a broken trace or RD550 blown or otherwise out of spec.

If the above findings are negative, and there are no shorts to ground or signs of visual damage around UC300, replacement of UC300 should be performed.

anecdotal findings - I had a board booting but with no backlight. Traced it to PP5V_S2 only being ~2v rather that the expected ~5v. Both P5VS2_PWR_EN and PPBUS_5VS2_VIN were present at UC300. Upon inspection of the UC300 area, I replaced some out of spec resistors, repaired a corroded pad on one and ultimately replaced UC300. Still only ~2v on PP5V_S2. Then I looked at the resistor that I had somehow missed RC332 P5VS2_BST/_R. It was well out of spec and when it was replaced P5VS2_BST/_R came back to ~5v and the backlight was restored. Also note that you will have no BKLT if PP5V_SW_LCD is missing. Check at pin 1 CP712.

PP3v3_S2_UPC - LDO produced from U8100.

Step 1. Check for a short to ground. Normal diode mode to ground reading is ~ 0.420.

The incidence is unknown for PP3v3_S2_UPC missing without a short or voltage leakage to ground, however possible differential diagnoses can include corrupt firmware, where a DFU revive may resolve the issue or a failed PMIC (U8100), or an issue with the SOC.

This section will be updated when more M1 powered boards present with similar issues. As of 1/18/2022, issues with PMIC voltages remain rare.

PP1v25_S2 - Produced from U7700, BUCK 13 LXO.

Step 1. Check for a short to ground. Normal diode mode to ground reading is ~0.490.

The incidence is unknown for PP1v25_S2 missing without a short or voltage leakage to ground, however possible differential diagnoses can include corrupt firmware, where a DFU revive may resolve the issue or a failed PMIC (U7700), or an issue with the SOC.

This section will be updated when more M1 powered boards present with similar issues. As of 9/21/2022, issues with PMIC voltages remain rare.

Differential Diagnosis for 5v and amp draw cycling between 0.00 and 0.45a as measured on a USB-C amp meter

Nearly always caused by a short to ground, particularly on a rail powered by a PMIC. PP1v8_S2 being shorted to ground has caused this issue, however other rails produced by U7700 or U8100 can likely cause the same symptoms and presentation. The cycling amp draw occurs because the PMIC attempts to start it's power sequence but senses an overload and shuts down. The process is repeated continuously.

• Short to ground on any of the PMIC buck output lines on U8100 or U7701.

➤Diagnostic Steps

High Resolution thermal imaging is highly specific for the detection of short circuits on M1 boards and should be the first line diagnostic method when short circuits are suspected.

1. Check for shorts to ground on outputs of U8100 and U7701.

L8100 / L8101 / L8102 (PPVDD_PCPU_AWAKE) - Normal diode mode to ground reading ~0.007 / ~45Ω

L82B0 / L82B1 (PPVDD_ECPU_AWAKE) - Normal diode mode to ground reading ~0.114 / 140Ω

L8270 / L8271 (PPVDD_CPU_SRAM_AWAKE) - Normal diode mode to ground reading ~ 0.116 / 450Ω

L8120 / L8121 (PPVDD_SOC_S1) - Normal diode mode to ground reading ~ 0.012 / 110Ω

L8110 / L8111 / L8112 (PPVDD_GPU_AWAKE) - Normal diode mode to ground reading ~ 0.010 / 35Ω

L8280 / L8281 (PPVDD_DISP_S1) - Normal diode mode to ground reading ~ 0.055 / 233Ω

L8290 (PPVDD_DCS_S1) - Normal diode mode to ground reading ~ 0.176 / 310Ω

L84E0 - (PP1v4_LDO_PREREG) - Normal diode mode to ground reading ~ 0.285 / >30kΩ

➤ Repair Steps

- Replace the shorted component as indicated by thermal imaging.

- Case Report from user: TCRS Circuit An 820-02020 presented with no power. Upon inspection, no diffuse signs of liquid damage were seen on the system board or enclosure components. Thermal imaging was performed which revealed abnormal pulsation and enhancement of U8100 and L8230 (PP1v8_S2). A short to ground was measured. 1v and 5 amps was injected to the line which revealed abnormal enhancement of UC820. (PP1v8_S2 to PP1v8_S2SW_VDD1 switch.) UC820 was replaced which restored 20v on the USB C amp meter, however amperage was still abnormal at 0.08 amps. Thermal imaging revealed some heating of the CPU when power was first connected. Further in depth visual inspection revealed C1105 to have signs of cracking and excessive heat. A short to ground was measured on PP1v8_S2SW_VDD1. C1105 was removed, with resolution of the short, however function was not fully restored, and PP1v8_S2SW_VDD1 was still seen to be missing despite normal VIN and EN lines to UC820. UC820 was replaced again which restored full function of the board. We can conclude that in this particular case, UC820 was likely damaged from overload due to a short on its output.

- Case Report from user: @TIPTOPGSM_RO If all the values are as per specs, then you need to flash the usb-c eeprom UF230, i managed to fix this issue after finding low voltage on PP5V_S2, i had 0.75v instead of 5v, issue was caused by fried UP700, which then fried UC300, after replacing all above, all power rails were in specified specs, device still stuck at 5v 0,45amps drawing, i just replaced the UF230 from donor board since i had no socket available to reprogram the eeprom so it booted up straight to the login screen.

Differential Diagnosis for a clicking sound from system board with low voltage as measured by a USB-C amp meter.

• Corroded or failed PP3v8_AON creation MOSFET. (Q5800, Q5820, Q5840.)

• Corroded or failed PP3v8_AON buck converter IC. (U5700)

➤ Diagnostic Steps

1. Measure voltage on PP3v8_AON. (Normal voltage = 3.8v. ) You will likely find that the voltage is low and pulsing slightly or completely absent. This confirms the diagnosis.

2. Perform a visual inspection of the system board. This failure is very uncommon without liquid damage. Often times, you will find that one of the 3v8_AON MOSFETs will be corroded, or that U5700 itself will be corroded. If the MOSFETs (Q5800, Q5820 or Q5840) are corroded, proceed to the Repair Steps below for "PP3v8_AON creation MOSFETs corroded" below. If the PP3v8_AON buck converter IC (U5700) is corroded, proceed to the "PP3v8_AON buck converter IC corroded" repair steps below.

➤ Repair Steps

PP3v8_AON creation MOSFETs corroded

- Replace Q5800, Q5820 and Q5840 together. It's important to replace all 3 MOSFETs at the same time as if one MOSFET failed, it may have damaged the others. Replacement of U5700 is not necessary as long as it isn't corroded and as long as the clicking sound does not persist after replacement of the MOSFETs.

PP3v8_AON buck converter IC corroded

- Replace U5700. Replacement of the PP3v8_AON creation MOSFETs is not required unless they are corroded or if the clicking sound persists after replacement of U5700.

Verify correct voltage on PP3v8_AON after the relevant repair is performed.

Differential Diagnosis for 20v 0.01 amps as measured with the board out of the enclosure.

-Not indicative of a fault.

Normal operational characteristic of first generation M1 boards. (820-02016 and 820-02020.) Place the board in the enclosure to properly test the functionality of a repair.

Differential Diagnosis for 5v / ~0.20-0.25a with power cycling of the USB-C amp meter on the 820-02016 logic board.

• Short to ground on PP3v3_S2.

• Short to ground on PP1v25_S2.

Note: Other possible diagnoses are possible, and will be added when/if they are discovered.

➤Diagnostic Steps

High resolution thermal imaging is the preferred first line diagnostic method on M1 based boards when a short to ground on a S2 rail is suspected. Voltage injection is usually not required. If high resolution thermal imaging is unavailable, proceed to the steps below to identify a short on PP3v3_S2 or PP1v25_S2.

1. Measure diode mode readings to ground on PP3v3_S2. - LC710 is the best place to measure for a short on PP3v3_S2 on the 820-02016 logic board.

Place your multimeter in diode mode, and place the red probe on a grounding point such as a screw hole and the black probe on LC710.

Normal diode mode reading to ground on PP3v3_S2 is .440

2. Measure diode mode readings to ground on PP1v25_S2. - Pin 1 of SWV011 is a good place to measure from. This is a large pad from a non-stuffed component.

Normal diode mode reading to ground on PP1v25_S2 is ~0.490.

Shorts to ground are generally considered when the diode mode to ground reading is significantly less than the above normal values. Readings can vary due to temperature or normal wear.

If a short to ground is measured, inject voltage to the associated line. Generally, 1v at 3-5 amps is sufficient to find most short circuits. NEVER INJECT MORE VOLTAGE THAN THE LINE IS RATED FOR! Carefully feel around the board for a warm area, then localize the shorted component using isopropyl alcohol. The alcohol will rapidly evaporate off of the shorted component.

For more information on finding short circuits, see this article. - [5]

➤ Repair Steps

Short to ground localized on PP1v25_S2 or PP3v3_S2

- Replace the shorted component.

Differential Diagnosis for no camera detection after liquid damage on the 820-02016 logic board

• Corroded TCON connector with corresponding short on the display. (LP670 will also be blown.

• Blown 5v supply filter. (LP670.) Always inspect the camera connector/cable on the display if this filter is blown.

• Bad webcam in display due to liquid ingress.

• TCON board fault.

➤Diagnostic Steps

1. Check continuity on LP670.

- Place your multimeter in continuity mode and measure across the component. LP670 is a filter, so it is essentially a wire. Your reading should be close to 0.01. You will often find this component visually exploded/burned with adjacent trace damage. Usually, there will also be corrosion on the display camera connector which will require replacement of the connector and sometimes the entire display assembly. (Left side of the primary display connector.) If LP670 measures normally, proceed to the next steps. If LP670 is blown or burned, proceed to the "LP670 blown or visibly burned repair steps below."

2. Check camera function with a KNOWN GOOD display.

- You don't have a board issue until you know you don't have a parts issue. If LP670 measures normally, test with a known good display and display cable, or place the board in a known good housing. If the camera works normally with a different screen/enclosure, the display is the issue and will need to be replaced. BE SURE TO CHECK LP670 FIRST!

➤ Repair Steps

LP670 blown or visually burned

Understand that this fault occurred because of a catastrophic short within the TCON board of the display. If you replace LP670 without addressing the display issue, LP670 will fail again.

• Generally, when LP670 blows, there will be corresponding pad and trace damage and jumper wires will usually be required.

- 36 or 37 AWG enamel coated copper wire is recommended for jumper wires in this application due to the power requirements of the webcam and ambient light sensor, which are both powered off this filter.

- The closest recommended jumper point for the input side of the filter (PP5v_S2) is CP717 if the traces and vias for PP5v_S2 are not damaged in that area. If damage is seen on and around CP717, use R5900 as a jumper point. Run wire to LP670 and pin 1 of UP700 as necessary. Ensure you have continuity between PP5v_S2 and CP717 if you are using that capacitor as a jumper point.

- Using a conformal coating on the jumper wire is strongly recommended.

- Use a 120 Ohm ferrite filter rated for 1.5 amps in a 402 package size only, preferably from a donor board. Do not substitute a resistor, wire or fuse.

- After LP670 is replaced, and jumper wires are run as needed, replace the display assembly or replace the camera connectors on both the cable side and TCON board side on the display. If any internal layer damage is noted to the TCON board or camera cable, replace the display assembly.

Only a display for a A2337 MacBook Air should be used. A older display for an Intel based MacBook Air will not work and may cause further damage.

The diagnostic and repair process remains mostly the same on the first generation M1 boards compared to other USB-C and Retina machines as the backlight circuits are essentially the same.

Ensure you actually have no backlight! A2337 displays can fail in a way where they will be dimly lit with no image. Normal backlight voltage is around 48-50v as measured on CP860. If you have 48-50v on CP860, you likely will have backlight.

Differential Diagnosis for no display backlight on the 820-02016 Logic Board

•Failed display TCON board.

•Failed display LED strip.

•Short to ground on backlight output caused by a failed output decoupling capacitor. (Most common)

•Failed backlight driver (UP800)

•Backlight driver circuit fault (resistors, traces, capacitors, MOSFETs.)

➤Diagnostic Steps

1. Check for a short to ground on backlight output (PPVOUT_LCDBKLT). CP860 or adjacent capacitors in the series is an appropriate area to measure, since it does not require removal of the system board. If a short to ground is measured, unplug the display connector to rule out a potential short to ground within the TCON board or LED strip of the display. If the short is persistent with the screen unplugged, proceed to the "Backlight output shorted to ground" repair steps below.

2. If no short to ground is measured, and there is no signs or history of infiltrative liquid damage to the system board, test the device with a known working display. If the device has backlight with a known working display, proceed to the repair steps below for "Device has backlight output on a known good display, but not on the originally installed display" below.

The following section will focus on determining if UP800 is the cause of the backlight issue.

Failure of UP800 without the presence of liquid damage is uncommon, however if you have no backlight with a known good display, and do not have a short on backlight output, UP800 may be bad or you may have another circuit fault such as a bad resistor or broken trace within the UP800/backlight generation circuit.

If UP800 is corroded, replace it along with any other corroded resistors or capacitors adjacent to it.

It is also common for current sensing pins (pins 9 and 10) of UP800 to be very corroded with corresponding pad/trace damage. If this is the case, gently scrape the coating off of the traces going to pins 9 and 10 of UP800 and run jumper wires from the traces to pins 9 and 10 of UP800. 0.02mm jumper wire is recommended.

①. Ensure that the backlight circuit has its input voltage (VIN) by measuring on BOTH SIDES of FP800, which is the backlight fuse. Voltage on both sides of the fuse should be equivalent to PPBUS_AON. If voltage is normal on both sides of the backlight fuse, proceed to the next step. The backlight fuse can blow in response to a previous short to ground in the backlight circuit, or due to the screen being plugged in without the battery disconnected. Rarely, the backlight fuse can become open (blown) due to development of sulfur deposits on the fuses' filament , or due to micro manufacturing defects on the fuses' filament. If the backlight fuse is blown, check for a short on backlight output (PPVOUT_LCDBKLT) and on backlight input (PPVIN_LCDBKLT_F). If the fuse is blown, proceed to the "Backlight fuse (FP800) blown in the absence of a short to ground" repair steps below.

② Check that the backlight circuit is being enabled. The screen will need to be connected and flipped open for the enable signal to be sent out. With the board on your bench, with a known good screen connected, measure voltage on RP842. BKLT_EN_R / BL_PWR_EN should measure around 1.8v. If backlight EN is missing, be sure that you do not have a parts issue with your display or logic board to display cable. BKLT_EN missing on this board due to a board issue is very rare, but may be cause by a failed PMIC (U8100) or failed iGPU in the M1 SoC itself.

③ Ensure you have continuity between pin 1 of RP831 (LCDBKLT_FB_XWR) and PPVOUT_LCDBKLT. Although rare on this board, it is worth checking to see if the backlight feedback trace is blown. Without feedback, the backlight driver will not produce its output.

④ Check the 5v input to UP800. (PP5v_BKLT_A and PP5v_BKLT_D). RP845 or RP844 may blow if UP800 fails. Ensure you have 5v on both of the above listed lines. If low, replace the blown resistor(s) and UP800.

If you have VIN (PPVIN_LCDBKLT_F), Enable (BKLT_EN), and both 5v lines (PP5v_BKLT_A and PP5v_BKLT_D), and no backlight output with the absence of a short to ground, replace the backlight driver (UP800).

➤Repair Steps

Backlight output shorted to ground:

- Find and replace shorted component.

Ensure the short is not caused by the display. Be sure to measure with and without the display connected.

Most commonly, a short on backlight output is caused by a shorted capacitor. If you have a short on backlight output, solder a wire to a sturdy component that can handle a good amount of current, usually one of the backlight output capacitors (CP860/CP867 or similar is usually a good spot.) Inject 10v at 5 amps and thermally image the system board or deploy comparable short detection methods. For more information on short circuit detection, please see the following page - [6]

Device has backlight output on a known good display, but not on the originally installed display

- Replace the display assembly. Use only a display for a A2337 MacBook Air. Earlier Intel MacBook displays will not work.

Backlight fuse (FP800) blown in the absence of a short to ground.

- Replace the backlight fuse with a fuse from a donor board or a compatible replacement. (0603 package size, 3 amp, 32v fuse. Brand does not matter.)

Before applying power be sure to double check for a short to ground. Check for a short to ground on backlight output and backlight input.

PP5v_BKLT_A or PP5v_BKLT_D absent or low

Ensure PP5v_S2 is present to both resistors prior to replacing any components

- If PP5v_BKLT_A is absent or low, replace RP845 and UP800.

- If PP5v_BKLT_D is absent or low, replace RP844 and UP800.

Case 1: 5V Cycling between 0.0 and 0,150A

Case 2: 5V 0.4 - 0.5A