IBM Intel Xeon L5420 44E5137 User Manual

Product codes
44E5137
Page of 100
Heatsink Clip Load Methodology
86
Quad-Core Intel® Xeon® Processor 5400 Series TMDG
C.2.2
Typical Test Equipment
For the heatsink clip load measurement, use equivalent test equipment to the one 
liste
Notes:
1.
Select load range depending on expected load level. It is usually better, whenever possible, to operate in 
the high end of the load cell capability. Check with your load cell vendor for further information.
2.
Since the load cells are calibrated in terms of mV/V, a data logger or scanner is required to supply 5 volts 
DC excitation and read the mV response. An automated model will take the sensitivity calibration of the 
load cells and convert the mV output into pounds.
3.
With the test equipment listed above, it is possible to automate data recording and control with a 6101-PCI 
card (GPIB) added to the scanner, allowing it to be connected to a PC running LabVIEW* or Vishay's 
StrainSmart* software. 
4.
IMPORTANT: In addition to just a zeroing of the force reading at no applied load, it is important to 
calibrate the load cells against known loads. Load cells tend to drift. Contact your load cell vendor for 
calibration tools and procedure information.
5.
When measuring loads under thermal stress (bake for example), load cell thermal capability must be 
checked, and the test setup must integrate any hardware used along with the load cell. For example, the 
Model 13 load cells are temperature compensated up to 71 °C, as long as the compensation package 
(spliced into the load cell's wiring) is also placed in the temperature chamber. The load cells can handle up 
to 121 °C (operating), but their uncertainty increases according to 0.02% rdg/°F. 
C.2.3
Test Procedure Examples
The following sections give two examples of load measurement. However, this is not 
meant to be used in mechanical shock and vibration testing.
Any mechanical device used along with the heatsink attach mechanism will need to be 
included in the test setup (i.e., back plate, attach to chassis, etc.).
Prior to any test, make sure that the load cell has been calibrated against known loads, 
following load cell vendor’s instructions.
C.2.4
Time-Zero, Room Temperature Preload Measurement
1. Pre-assemble mechanical components on the board as needed prior to mounting 
the motherboard on an appropriate support fixture that replicate the board attach 
to a target chassis.
For example: If the attach mechanism includes fixtures on the back side of the 
board, those must be included, as the goal of the test is to measure the load 
provided by the actual heatsink mechanism.
2. Install the test vehicle in the socket.
3. Assemble the heatsink reworked with the load cells to motherboard as shown for 
the Quad-Core Intel® Xeon® Processor 5400 Series CEK-reference heatsink 
example in 
and actuate attach mechanism.
4. Collect continuous load cell data at 1 Hz for the duration of the test. A minimum 
time to allow the load cell to settle is generally specified by the load cell vendors 
Table C-1.  Typical Test Equipment
Item
Description
Part Number (Model)
Load cell
Notes: 1, 5
Honeywell*-Sensotec* Model 13 subminiature load cells, 
compression only
Select a load range depending on load level being tested.
 
AL322BL
Data Logger 
(or scanner)
Notes: 2, 3, 4
Vishay* Measurements Group Model 6100 scanner with a 
6010A strain card (one card required per channel).
Model 6100