Aller au contenu principal

Install a rear B200 GPU

Follow instructions in this section to install a rear B200 GPU. The procedure must be executed by a trained technician.

About this task

Attention
  • Read Installation Guidelines and Safety inspection checklist to ensure that you work safely.
  • Touch the static-protective package that contains the component to any unpainted metal surface on the server; then, remove it from the package and place it on a static-protective surface.
  • A torque screwdriver is available for request if you do not have one at hand.
Note
Make sure you have the required tools listed below available to properly replace the component:
  • Torx T15 head screwdriver
  • 2 x Torx T15 200mm extension bit
  • Phillips #1 head screwdriver
  • Phillips #2 head screwdriver
  • Alcohol cleaning pad
  • B200 PCM
  • B200 SXM6 PAD-1
  • B200 SXM6 PAD-2
  • B200 GPU Service Kit
B200 (GPU & Retimer NVSwitch) (service & shipping bkt) Kit are reusable and mandatory when servicing GPUs and GPU cold plate modules. It is recommended to keep them at the facility where the server operates for future replacement needs.
Important
Putty pad/phase change material (PCM) replacement guidelines
  • Before replacing the putty pad/PCM, gently clean the hardware surface with an alcohol cleaning pad.
  • Hold the putty pad/PCM carefully to avoid deformation. Make sure no screw hole or opening is blocked by the putty pad/PCM.
  • Do not use expired putty pad/PCM. Check the expiry date on putty pad/PCM package. If the putty pads/PCM are expired, acquire new ones to properly replace them.
The following illustration shows the B200 GPU numbering and corresponding slot numbering in XCC.
Figure 1. B200 GPU numbering
B200 GPU numbering
Physical GPU socketSlot numbering in XCCLogical number in nvidia-smi

GPU 1

Slot 21

4

GPU 2

Slot 24

7

GPU 3

Slot 22

5

GPU 4

Slot 23

6

GPU 5

Slot 17

0

GPU 6

Slot 20

3

GPU 7

Slot 18

1

GPU 8

Slot 19

2

Procedure

  1. (Optional) For new GPU, remove the connector covers at the bottom.
    Figure 2. Removing connector covers
    Removing connector covers
  2. Install the GPU.
    1. Install the two GPU screw handles diagonally. Align the screw handles to the cold plate screw holes; then fasten the screw handles by hand.
      Figure 3. Installing the GPU screw handles
      Installing the GPU screw handles
    2. Hold the GPU screw handles to carefully place the GPU onto the GPU baseboard.
      Figure 4. Installing the GPU
      Installing the GPU
    3. Remove the two screw handles by loosening them by hand.
      Figure 5. Removing the GPU screw handles
      Removing the GPU screw handles
    4. Attach the two Torx T15 200mm extension bit to two torque screwdrivers. Simultaneously fasten the two diagonal Torx T15 screws with the screwdriver set to the proper torque.
    5. First set the torque screwdriver to 0.11±0.011 newton-meters, 0.97±0.097 inch-pounds to simultaneously fasten the two diagonal screws; then, simultaneously fasten the two diagonal screws.
    6. Then set the torque screwdriver to 0.6±0.024 newton-meters, 5.3±0.212 inch-pounds to simultaneously fasten the two diagonal screws; then, simultaneously fasten the two diagonal screws.
      Figure 6. Installing the GPU
      Installing the GPU
  3. Replace the Phase Change Material (PCM) on the GPU cold plate.
    1. Apply the PCM jig to the GPU cold plate.
    2. Remove the liner from one side of the pad. Align the PCM with the jig and place it onto the cold plate. Remove the jig; then, apply finger pressure across the entire surface area of the PCM to remove any trapped air and allow 1-2 minutes dwell time until it is firmly attached. Carefully remove the remaining top liner.
      Attention
      • PCM cannot be reused. PCM must be replaced with new ones every time the water loop is removed.

      Figure 7. PCM application

      PCM application
  4. Replace the putty pads (x10) on the GPU.
    1. Follow the B200 GPU application instructions to apply the putty pads.
      Note
      • Apply the putty pads from B200 SXM6 PAD-1 to the six locations marked with number 1
      • Apply the putty pads from B200 SXM6 PAD-1 to the two locations marked with number 2
      • Apply the putty pads from B200 SXM6 PAD-2 to the two locations on the GPU VR marked with number 3 and 4 (gray color)
      Figure 8. GPU putty pads instructions
      GPU putty pads instructions
    2. Remove the liner from one side of the pad.
    3. Make sure to align the two gray colored putty pads to the GPU VR (1) and the markings; then, place the pads to cover the GPU VR as illustrated and apply light finger pressure across the entire surface area of the pads to ensure adhesion. Carefully remove the remaining top liner.
    4. Align the putty pads to the markings on the GPU; then, place the pads onto the GPU and apply light finger pressure across the entire surface area of the pads to ensure adhesion. Carefully remove the remaining top liner.
      Attention
      • Putty pad cannot be reused. Putty pad must be replaced with new ones every time the water loop is removed.
      Figure 9. GPU putty pads replacement
      GPU putty pads replacement
      1 GPU VR (Cover the GPU VR with putty pad)
  5. Remove the service bracket and GPU cold plate assembly from the manifold.
    1. Loosen the captive screw that secure the service bracket to the manifold.
    2. Lift the service bracket and GPU cold plate assembly away from the manifold to remove it.
      Figure 10. Removing the service bracket and the GPU cold plate assembly
      Removing the service bracket and the GPU cold plate assembly
  6. Place the GPU cold plate onto the GPU.
    1. Flip over the service bracket and GPU cold plate assembly; then, gently place the cold plate onto the GPU.
    2. Adjust the GPU cold plate until the two guide pins are seated in the guide holes on the GPU.
      Figure 11. Placing the GPU cold plate
      Placing the GPU cold plate
  7. Loosen the two captive screws to remove the service bracket from the cold plate.
    1. Loosen the two captive screws that secure the service bracket to the GPU cold plate.
    2. Lift the service bracket away from the GPU cold plate to remove it.
      Figure 12. Removing the service bracket
      Removing the service bracket
  8. Fasten the screws by 360 degrees following the screw installation sequence:, and repeat to fully tighten the four Torx T15 screws with the screwdriver set to the proper torque.
    Note
    • (Except for the brand-new cold plate module) Ensure the TIM breaker screw is loosened to its initial position before tightening the cold plate screws.

    • Loosen the TIM breaker screw to return it to its initial position.

    • Close the lid. If the lid cannot be closed, the TIM breaker screw needs to be further loosened.

    1. First set the torque screwdriver to 1.0±0.1 inch-pounds, 0.112±0.0112 newton-meters to fasten the screws for a few rounds. Then set the torque screwdriver to 5.3±0.212 inch-pounds, 0.6±0.024 newton-meters to fully fasten the screws.
      Note
      • Make sure to follow screw sequence to prevent cold plate tilting.
    2. Repeat until all the screws on the four GPU cold plates are fully tightened.
    Figure 13. GPU cold plate screw tightening sequence
    GPU cold plate screw tightening sequence
  9. Route the leakage sensor module cable back onto the GPU cold plate and into the cable clips.
    Figure 14. Installing the leakage sensor module cable
    Installing the leakage sensor module cable
  10. Install the rear fan cage support bracket.
    1. Align the rear fan cage support bracket with the corresponding screw holes; then, install the rear fan cage support bracket on top of hose holder B/C as illustrated.
    2. Fasten the four M3 screws (PH2, 4 x M3, 0.5 newton-meters, 4.3 inch-pounds) to secure the rear fan cage support bracket to the fan cage.
    3. Fasten the eight M3 screws (PH2, 8 x M3, 0.5 newton-meters, 4.3 inch-pounds) to secure the rear fan cage support bracket to the chassis.
      Figure 15. Installing the rear fan cage support bracket
      Installing the rear fan cage support bracket

After you finish

  1. Reconnect all the cables that were disconnected. See Internal cable routing.
  2. Reinstall the power complex. See Install the power complex.
  3. Reinstall the CPU complex. See Install the CPU complex.
  4. Reinstall the fan cage. See Install the fan cage (trained technician only).
  5. Reinstall the rear top cover. See Install the rear top cover.
  6. Reinstall the front top cover. See Install the front top cover.
  7. Complete the parts replacement. See Complete the parts replacement.