I can relate to this quite a bit. Project estimations are very difficult when third-party vendors are involved. During my capstone project, we purchased a strain gauge that ended up being too small for our application. We didn’t know until we tested it, and this was well into development of the project. Testing and validation protocols had to be adjusted, and ultimately we ended up picking a different force sensor. Project estimation on that was definitely underestimated because of issues with third-party equipment. I, too, had anticipated that something like this would happen, so during the development of the project plan, I implemented a contingency plan in case of emergency. Our project is on schedule only because of the contingency plan, but project underestimation is still a significant issue.