One would think that if you want to make a small string section sound bigger, you could just record it twice and the sound will double. Unfortunately, this does not work too well. There will be a tinny or fuzzy sheen to the sound. Listen to the strings on Coldplay and Christina Aguilera songs and you will hear what I mean.
Even though there are two separate takes, players have a unique talent for being able to play the same phrase pretty much the same way every time. This will lead to phasing as the waveforms of the two takes occasionally line up. Even if you ask the players to vary it, they have a hard time pulling it off. This issue is most evident in passages with long notes. Riffs of short notes are not as big of a problem as they contain more variations in the sound by default, so the chances of the waveforms lining up exactly are slimmer, the but resultant sound will never be as good as if you just had more players in the first place.
The larger the section, the less this becomes a problem. But by the time the section is big enough to avoid the problem, there is probably no point stacking.
If you do have to do work with a small group and stack it up, here are some tricks –
1. Swap the violin parts. Have the seconds now play first. As it is the top parts the we notice phasing the most, this helps mitigate that problem.
2. Change the seating. Make sure no one sits in the same chair on the second pass. This means they hit the microphones differently. Also, sitting next to different people means they will blend a little differently.
3. Do a second pass with mutes on. This will build up the ‘mass’ of sound, and the change the mute makes to the sound is enough to avoid any bad phasing.
A better way to use a small group for a larger sound is to write new parts. An example would be to have them all play chords or riffs then play the melody on another pass. We do this for movies and video games quite often.
Whatever tricks you try, though, remember that nothing beats using a large section.