Frequency Cutoff error in preprocessing?

[kmharrington]

I'm seeing a bunch of observations raising errors that look like this:

1788491409.0804863, obs_1763234806_satp1_1111111, ['ws6', 'f090'], init_pipeline_run_error
        <class 'ValueError'>: lower freq=1 >= upper freq=0.8388319434722693  File "/global/homes/k/kmharrin/.local/soconda/20260812_0.2.5/lib/python3.13/site-packages/sotodlib
/preprocess/preprocess_util.py", line 1449, in preproc_or_load_group
    proc_aman, success = pipe_init.run(aman)
                         ~~~~~~~~~~~~~^^^^^^
  File "/global/homes/k/kmharrin/.local/soconda/20260812_0.2.5/lib/python3.13/site-packages/sotodlib/preprocess/pcore.py", line 567, in run
    aman, proc_aman = process.calc_and_save(aman, proc_aman)
                      ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "/global/homes/k/kmharrin/.local/soconda/20260812_0.2.5/lib/python3.13/site-packages/sotodlib/preprocess/processes.py", line 818, in calc_and_save
    wn_f_high = check_frequency_cutoff(wn_f_low, wn_f_high)
  File "/global/homes/k/kmharrin/.local/soconda/20260812_0.2.5/lib/python3.13/site-packages/sotodlib/preprocess/processes.py", line 792, in check_frequency_cutoff
    raise ValueError(f"lower freq={fmin} >= upper freq={fmax}")

What does this mean?

[msilvafe]

Either the fmin you provided in your config entry for noise is greater than the fmax you provided or greater than frequency cutoff which is computed from the data processing steps preceding noise in the pipeline either from the readout filter on pre-demod/pre-filtered data or by the lpf/bpf in demod for post demod/post filtering.