Sample sizes for piloting new components

I'm not exactly sure which folder this belongs here in the Quality Forum so I'm posting it here. We've recently started involving QA (me) in piloting components prior to using them in our packaging process and I'm a little uncertain about the right sample size to use. We manufacture and package dietary supplements. The components I'm talking about are new label stock or new bottle lids, that sort of thing. Let's say we're going to test a new lid to see if it seals better on our jars than our current lids. In the past, I've designated a sample size of 30 for both the control and test and then compared the results using hypothesis testing (z stat to a 95% confidence level). Because we don't have a lot size, it's a little less straightforward than simply using the ANSI standard. What do you do?