I am working with a Micro GC 490 with a CP-Sil 5 CB 8 meter channel. However, I am encountering something I do not like at all when using our natural gas samples: negative peaks. The occassional spikes where the TCD value falls below the baseline are bad enough, but the worst problem I'm seeing is a whole region - roughly ten seconds or so in duration - where the baseline drops significanly before returning. This region moves about between runs, and can cause serious problems when it happens in the first three minutes - right where the compounds I'm looking for occure. This always throws the area counts off for compounds in the region. Even baking the column out doesn't seem to offer a permanent solution: the negative region always seems to return.

Currently the column is being used at a constant 150 C and 21 psi.

Does anyone have any suggestions as to what could be causing this, or at minimum how to fix this issue?