The M14 x 1.5 x 90 Double Ended Studs involved in the customer’s request were assembled by initially installing them onto a drive socket and then driving them into a threaded hole in the hub. The stud is driven until the shoulder of the stud contacts the hub and a torque of 75 ft-lb (100 N-m) is achieved. The contact between the shoulder of the stud and the hub causes the first thread in the hub to distort creating a locking action between the stud and the top thread of the hub.The assembly tool automatically shuts off when 75 ft-lb torque is achieved. The operator then switches the tool to reverse to disengage the socket from the stud (see Figure 1).

The customer said that in some cases, the socket would not disengage from the thread of the stud when the assembly tool was reversed. This would cause the stud to back out of the hub. The customer’s review of the assembly procedure indicated the cause of the studs backing out was the
result of the studs not being made to the requirements of the drawings. These discrepancies were noted on the studs:
• The ends of the studs were concave and not flat or rounded as required by IFI-528.
• The studs that backed out did not meet the decarburization limits specified in ASTM F568 (as referenced in ISO 898-1).
• The “Total Thread Length” exceeded the maximumlengths as defined by IFI-528.
• The studs were not marked for strength level as specified by ASTM F568.

The customer said the studs’ excessive decarburization in combination with the concave end was causing the drive socket to lock onto the stud creating a disengagement torque that exceeded
the torque required to back the stud out of the hub.

Test Parameters

Three samples of studs that had backed out of hubs were submitted to Earnest Machine Products for testing. The samples were inspected for compliance to the customer’s drawing for dimensional conformance. The studs were also tested for compliance to ASTM F568 for Property Class 10.9 for Rockwell Hardness and Decarburization. Testing was also done on stock samples to learn the percentage of the lot that exhibited the nonconformances identified by the customer.

Test Results
The three samples that had backed out during assembly were inspected to the requirements of the customer’s drawing and IFI-528.

Dimensional Requirements are seen in Table 1. The dimensions that are underlined in Table 1 do not meet the requirements specified.

Table 1. Dimensional Requirements.

	Specified	Sample #1	Sample #2	Sample#3
Overall Length	90+/-1.5	90.33	90.26	90.28
Body Diameter	14.0/13.68	13.82	13.83	13.81
Total Thread Length*
Damaged End	20 +3/-0	23.2	24.1	23.9
Undamaged End	20 +3/-0	23.3	23.5	23.4
Thread Gain (undamaged)
Go	6g	Pass	Pass	Pass
No Go	6g	Pass	Pass	Pass
Grade Marking	10.9	Concave None	Concave None	Concave Non

Dimenstions that are underlined do not meet the requirements specified.
*IFI 528 specifies the “total thread length” as the distance from the end of the stud to the top of the extrustion angle.
*IFI 528 specifies that the points are to be flat or rounded (oval). Inspection of the studs from the lot showed that they are all concave.

Hardness Requirements. Core hardness readings were taken by sectioning the damaged thread end of the stud at one diameter from the end. Hardness readings were taken on the Rockwell C scale at the mid-radius and the average of four readings were recorded.

Surface hardness readings were taken by initially lightly sanding the surface of the shoulder to remove any oxides or scale and measuring the hardness on the Rockwell 30N scale. These are seen in Table 2. Hardness testing showed that the core hardness meets the requirements specified in ASTM F568 (and ISO 898-1). The surface hardness also meets the requirements of ASTM F568
that the maximum hardness does not exceed R30N 59. The surface hardness does fall well below the minimum core hardness specified for a 10.9 strength level product.

Table 2. Hardness Requirements.

	Specified	Sample #1	Sample #2	Sample #3
Core Hardness	Rc 33/39	Rc 35	Rc 33	Rc 34
Surface Hardness	R30N 59 max	R30N 40	R30N 28	R30N 24
(Corresponding Rc Hardness)	(Rc 39)	(below scale)	(below scale)	(below scale)

A sample was tested for “partial decarburization” per the requirements of ISO 898-1 (Hardness Method). The sample was sectioned in the undamaged end of the thread. Knoop Hardness readings were taken at the required three locations on the thread. Per the requirements of ISO 898-1, the reading measured at location 2 must not be less than 30 hardness points
of the reading taken at position 1 (see Table 3).

Table 3. “Partial Decarburization”

Location	Hardness
#1	353.3
#2	306.3
#3	319.1

Testing showed that stud does not meet the decarburization requirements of ISO 898-1. The difference in the hardness readings was 47 parts between points 1 and 2 on the thread. Figure 2 shows decarburization in the shoulder and decarburization in the threads after the thread rolling operation.

Fig. 2 – Decarburization in the shoulder and decarburization in threads after the thread rolling operation.

Evaluation of Stock

Testing was then conducted on stock samples of the stud to determine the percentage that show signs of decarburization.The samples were tested by measuring the surface hardness on the shoulder of the studs. The hardness was measured using the Rc scale.

This testing showed that approximately 15% of the stock had surface hardness readings that measured below Rc 30 (with the majority reading in the Rc 5 to 10 range).

Testing was then conducted to compare the tensile strength of product that exhibited decarburization to samples that did not show decarburization.

The samples were tested by pulling them in the tensile tester to failure. The maximum load required to cause failure was recorded and the tensile strength was calculated based on the stress area of the thread. The results were recorded in pounds and converted to metric units (see Table 4). Testing showed that all samples exceeded the minimum ultimate tensile strength for a Property Class 10.9 Stud of 1040 MPa.

Table 4. Evaluation of Stock.

	Max Load	Tensile Strength	Surface Hardness (converted to Rockwell C scale)
Samples Showing decarb
1.	33011 lbs	170400 psi (1175 MPa)	Rc 3
2.	3285 lbs	169600 psi (1169 MPa)	Rc 3
3.	33420 lbs	172500 psi (1189 MPa)	Rc 5
Samples that did not show decarb
1.	34716 lbs	179200 psi (1235 MPa)	Rc 31
2.	33897 lbs	175000 psi (1207 MPa)	Rc 31
3.	35053 lbs	174600 psi (1201 MPa)	Rc 31

Micro-hardness readings were taken on one sample that showed decarb and on one sample that did not show decarb based on the surface hardness testing. The depth of the decarb was determined by taking micro-hardness readings every 0.002″ from the surface of the shoulder into the core. The samples were tested using Knoop hardness and then converted to the corresponding Rb or Rc hardness (see Table 5).

Table 5. Microhardness Readings.

Sample with Surface Hardness of Rc 3
Depth	Hardness	Depth	Hardness	Depth	Hardness
.002″	Rb 78.6	.008″	Rb 88.9	.014″	Rc 32.7
.004″	Rb 80.6	.010	Rb 98.2	.016″	Rc 34.3
.006″	Rb 84.4	.012	Rc 29.7	.018″	Rc 34.7
Sample with Surface Hardness of Rc 31
Depth	Hardness	Depth	Hardness	Depth	Hardness
.002″	Rc 34.1	.008″	Rc 35.9	.014″	Rc 37.4
.004″	Rc 35.7	.010	Rc 37.0	.016″	Rc 38.0
.006″	Rc 35.8	.012	Rc 37.1	.018″	Rc 36.9

Testing showed that the sample that measured low on the surface hardness has partial decarburization to a depth of 0.012″. The sample that did not measure low for surface hardness did not show decarburization.

Conclusions

Review of the nonconforming characteristics identified by the customer did show that the studs do not meet the drawing requirements specified in drawing 6.C09201-12, as follows:
• The ends of the studs are concave.
• Approximately 15% of the studs do not meet the decarburization limits specified in ASTM F568
(as referenced in ISO 898-1).
• The “Total Thread Length” exceeds the maximum lengths as defined by IFI-528.
• The studs are not marked for strength level as specified by ASTM F568.

The customer has also indicated that the two contributing nonconformances that are resulting in the studs’ backing out are the following:
• The points not being flat (or rounded).
• The presence of a decarburized surface.

This combination results in the drive socket becoming locked onto the stud, and then the resulting disengagement torque exceeds the toque that is required to back the stud out of the hub.