This issue is described here: https://github.com/rdgao/daily_notebook/blob/master/daily19_04_04_noisefloor.ipynb
In summary, when the aperiodic scale/sd becomes large, oscillations may be missed during the standard fitting procedure in semi-log space. A potential solution was described by @rdgao and my understanding is to remove the aperiodic fit in linear space (spectrum - 10**fm._ap_fit). The fitting procedure after the ap_fit is removed may also need to be updated to work in linear space. There is also the issue of low frequency noise I'm not sure how to account for.
This issue is described here: https://github.com/rdgao/daily_notebook/blob/master/daily19_04_04_noisefloor.ipynb
In summary, when the aperiodic scale/sd becomes large, oscillations may be missed during the standard fitting procedure in semi-log space. A potential solution was described by @rdgao and my understanding is to remove the aperiodic fit in linear space (spectrum - 10**fm._ap_fit). The fitting procedure after the ap_fit is removed may also need to be updated to work in linear space. There is also the issue of low frequency noise I'm not sure how to account for.