I also had a similar problem when decided to cut off the first minute of the pre-show music, and then shifting the tracks back. When I exported it after moving them, they summed together a bit differently and a couple peaks suddenly were clipped.
That seems to me like you are summing multiple channels and may have cut a slightly different length from the start of each in making the shift. Not enough that you would visually notice a shift or such that they sound obviously misaligned afterward, but just enough that some high momentary peaks sum differently. This can happen depending on how you specify making selections and cuts, which is somewhat different for each editor. For example, if you have,
select between zero-crossings (or the equivalent) specified in your editor, the cuts to each separate object are made at whatever zero-crossings are closest to the edit points you specify, which is likely to differ by a few samples for each object. If you then shift both cut objects in time to the beginning of the time-line such that both align with the zero-point of the timeline, they will be misaligned by the same amount.
Not saying that's definitely the case, just outlining one potential way it could have occured. And to be clear- specifying
select on zero crossings is often what you want when a cut occurs within the piece without a fade, as it will eliminate a potential click at the edit point.
An easy way to avoid that is to trim the heads and tails (the unwanted pre-show and after-show portions) prior to alignment and editing, or to simply leave them there and start your tracking not from the very start of the raw audio file trimmed to length, but from wherever you want the initial fade-in for track 1 to occur. Otherwise you need to be careful in how you select, cut, and move the objects to keep them aligned with each other.
In audacity, it shows you with red vertical lines where the clipped bits. I just edited those bits again and export again, keeping it clean. But I can't quite explain it. I kind of just assumed that the dither (which I believe is a low-level noise of some kind used to mask the truncated frequencies) added some fraction of a decibel to those bits and thus pushed them over the edge. Not much mind you, but still it was showing as being clipped. Luckily after all the editing it worked out and sounds really good. Took some work but I feel it was well worth it.
The dither is way too low in level to add in that way. It is completely swamped by the far higher signal in both sources. You're only going to get increased peaks where both sources have aligned peaks with similar levels.
I'd love hear some of your recordings using your omni-setup to hear the possible difference with those verses the condensers I used. Or maybe have you take a listen to mine and tell me what I should possibly tweak or try, or if things sound pretty good as is. Sometimes listening helps more than just talking about it.
Definitely does! Happy to give a listen. Most of my stuff is not posted, but I'm working on shifting towards doing so, mostly for folks here. I think most productive is a hand-in-hand combination of discussion and listening to really get a good grip on what's going on.