mighoser
Aerospace
- Jul 10, 2006
- 160
Folks,
I just received the final report from the lab. The failed link is referred to as Link #79 Some conclusions from the lab:
1. Link #79 exhibited a shallow linear surface discontinuity from the original strip rolling process. The seam of unfused material was approximately .001" deep by approximately the width of one leg of the annular end of the link and was oriented at approximately 20 degrees with respect to the rolling direction of the parent strip material.
2. It is probable that the stress concentration resulting from the surface discontinuity coupled with the low tensile ductility exhibited by link # 79 made it susceptible to failure under the transient start up load of the test firing.
3. The alloy chemistry from Link #79 exceeded the maximum carbon limit per AMS 5528F, paragraph 3.1, with the remaing alloy elements within specification for type 17-7PH stainless steel.
4. The hardness of Link #79 was 46 HRC (487 HK 500 gf)
The shallow linear discontinuity along the edge of the fracture in Link #79 was the only such anomaly observed among all of the hardware examined at the M&P Lab. The seam probably originated during the rolling operation which produced the nominal .032 inch sheet stock. Due to coining of the fracture edge containing the seam, the original depth of the flaw cannot be determined, but it was probably greater than the .001 depth documented in the report. Surface flaws of this type can arise from scabs or gouges in the thicker gage coil or hot band surface, which subsequently gets rolled back into the sheet. The opposite half of this feature was not recovered from the wreck debris. It is likely that the entire discontinuity in original strips would have been rolled more or less closed at the surface, forming a tight crack-like stress concentrator with a sharp root radius. Given the lack of tensile fracture ductility exhibited by this link, the fracture toughness and crack tolerance of this material was probably low relative to material with higher ductility and lower hardness. The combination of a stress concentration the seam and reduced toughness because of the heat treatment condition during the transient loading during startup of the conveyor mechanism is thought to have caused Link # 79 to fail catastrophically.
Additional thoughts: There has been some confusion on what the hardness of the parts should be with respect to AMS 5528F and AMS 2759/3. AMS 5528F describes response to heat treat should be 38 to 46 HRC whereas AMS 2759/3 requires hardness to be 38-44. During tensile tests of parts, we found those parts which were equal or less than 44 HRC exhibited higher failure loads and elongation at failure in a very consistent manner. Those parts which failed prematurely at lower failure loads and elongation were 44-47 HRC. I suspect that our fine blanking/straightening process which occurs when material is in the TH1050 condition introduces some normally acceptable micro cracking. Higher hardness and therefore increased notch sensitivity probably can explain the results of our tensile testing. It is important to note that though the higher hardness link failed at low loads, these load are below worst case operational loads.
Questions: Is there a better choice in terms of alloy or HT which would minimize our risk to notch sensitivly from varitions in processing?? I believe 17-7 TH1050 one the least forgiving alloy/heat treat with respect to processing variations.
Any comments on our labs defect hypothesis. Is it plausable?
