The screws inside N5ii are all installed with torque screwdrivers and we have very specific guideline on the setting of torque screwdriver when used with different product/model. We need to make sure the components (including PCB and battery mounting) will be fixed securely without affect the maintainability of the device such as stripped thread or broken component mounting due to excessive pressure. Unless these screws are loosed (which will be considered as faulty unit), we do not recommend user to tighten the screws with additional force, let alone replace the our screws with "stronger" screws.
The random pause bug is all about "interrupt". So far we are handling this bug from software optimization perspective: eliminating unintended process interference and interrupt from Android system. I can't say we solved the bug 100% already but we have achieved very good progress so far. Based on the log file we received from our beta tester, we have trace a lot of "triggers" of interrupt and rectify those specific instance promptly. From what we observed (yes, we have reviewed log file submitted after v3.3en), the remaining instance are completely random and cannot be replicated even through system stimulation. Theoretically these type of interference can be triggered by electronic signal during digital audio processing (analogue processing will cause distortion, not interrupt) but so far we haven't identified any of these systematically. In other word, if there are indeed hardware based interrupt, this is unlikely a design issue but hardware issues that escaped our QA process or surfaced after extensive use (hence cannot be detected during QA process before leaving our factory).
Relating installation screws and the pressure applied to tighten the screws to the random pause bug is beyond our technical comprehension.