MNT Reform Motherboard Repair Logs

MNT Reform
1

As I’ve been doing a number of board level repairs this year, I’ve decided to collect my notes here as they might help others to troubleshoot or to give to repair companies as a guide of where to look.

3 Likes
2

MNT Reform Repair (First EU Batch/R-2)

Date: 2021-06-30

Leftmost column of keys nonfunctional.

Probable cause: Short between power jack and keyboard switch.

Replaced keyboard with new keyboard (QWERTY). Kept old for refurbishing (MCU change).

i.MX8MQ SOM was damaged – no boot. Maybe only voltage regulators burned out. Stored for future repair.

Replaced i.MX8MQ SOM with new one.

Performed safety modification to prevent future problems: MNT Reform: First EU Batch? Do this safety modification

Tested all features with R2Test SD Card.

Cleaned up and put back into sleeve that it came in.

2 Likes
3

Repair Log: MNT Reform Beta (D-3)

Date: 2021-06-30

Cleaned keyboard and exchanged keycaps to MBK model (QWERTY).

Cleaned unit with compressed air.

Replaced missing screws.

Replaced all 8 battery cells (7/8 were dead).

Replaced discharge resistor R29 (it was exploded). 10 Ohms.

Performed charge control modification with a bodge wire (see MNT Reform R-2 Motherboard Update 2021-04-19).

Performed safety modification on power jack (see MNT Reform: First EU Batch? Do this safety modification).

Upgraded motherboard firmware to R2/20210419

Upgraded keyboard firmware to R1/20210419

Had 2 bad measurements on LTC sense resistors:

upload_a7261547134b4a4a19ea7d448d8e92fc
upload_a7261547134b4a4a19ea7d448d8e92fc1013×1089 197 KB

Decided to desolder and exchange LTC6803-4 chip. This fixed the problem. The chip was faulty.

Performed charging test. Cell voltages nominal, charger correctly finished at top end.

3 Likes
4

MNT Reform Repair (R-3)

Date: 2022-10-26

Symptom: Internal display doesn’t work after a US repair company, PSI, originally fixed a problem with the power circuit. Then, they exchanged U10 with a new one (my suggestion). This did not fix the problem. Customer sent the board to us for inspection.

  • The problem is that regulator U13 does not work. It supplies 1.2V for the eDP converter and the USB hub. The repair company did not check for/find this problem.
  • U13 outputs only 0.025V
  • LPCs 1V2_PWR_EN line is low, so it doesn’t enable U13 or U13 pulls the line low very hard
  • U13 is the culprit. I removed the 0R on its enable and it is now high on the LPC side.
  • Replacing U13 fixed the display (and i guess USB).
  • I have also tested both pcie slots, ethernet, audio, battery charging and discharging. Looks all good.
2 Likes
5

MNT Reform Repair (R-3)

Date: 2022-08-03

Symptom: Internal display stopped working. In dmesg it looks like eDP link failed.

  • The problem was a faulty U10 (SN65DSI86IPAPQ1). This is rare. Maybe the eDP connector was reversed or there was a static discharge.
  • Swapping U10 with a new one fixed the problem.
2 Likes
6

MNT Reform Repair (R-3)

Date: 2022-08-04

Symptom: Motherboard burns a lot of power and generates heat immediately when on wall power.

  • It turns out Diode D5 was shot. This causes some of the buck/boost MOSFETs to get very hot.
  • Removing D5 fixes the problem. According to LTC4020 manual, this diode is optional, so leaving it off.
2 Likes
7

Motherboard repair on 2022-11-03

This one was complicated, took at least 4 hours.

Symptom: system uses 12W of wall power when powered off (no CPU module inserted). Q17 gets very hot, turns some sprinkled IPA into steam. owner already removed D5, a usual culprit, but it did not fix the issue.

initial repair:

  • swapped Q17
  • swapped U2
  • swapped Q2
  • removed D7, D6, D4. short is gone, but LTC (U2) not operational
  • put D6, D4 back in → short is back
  • measured 2ohms across Q2, that’s bad
  • inducing power at 30V_GATE works fine, LPC is functional
  • so the problem is close to buck/boost of LTC
  • inducing 12V at BAT1FUSED also works fine, LPC detected on USB
  • so the problem is close to the input stage of LTC
  • candidates: Q15, Q1
  • 120 ohms to gnd on gate of Q17. unlike on reference board
  • replaced Q17 again. it was shot again! problem is gone.
  • put in new D7. still fine.
  • repaired HDMI connector with dental tools.
  • replaced partially melted connector J23.

tests:

  • SD Card boot OK
  • eDP display OK
  • HDMI OK
  • PCIe 1+2 OK
  • Ethernet OK
  • RTC OK
  • UART 1 OK
  • USB OK
  • SYSCTL OK

battery tests:

  • highest cell (BAT1->2) measures 0. drop of 3V on F1.
  • fuse F1 is shot!
  • replaced with new fuse.
  • now another cell measures 0. R14 measures 6V, R15 measures 6.5V, which is odd.
  • this is fine, it was just the protected BB closing off one cell. opens after applying charge.
  • cells are charging normally.
  • LPC FW updated to latest version.
  • repair concluded.
2 Likes