Q1 ripps are 1- 3 bytes larger.

If I understand this discussion correctly, you're talking about the encoder analysis quality setting, not a bit rate setting. So I would not expect the resulting files to be any different in size at all.

I don't have the lame parameter documentation in front of me at the moment, so I can't check what this parameter is supposed to do. But if it's a quality setting as I suspect, then you should always use the highest available setting when encoding. It will take longer to encode, but it won't make the files any bigger and the result will be that the encoded file sounds more like the original than if you did it at the faster setting.