Warning: pulse time ouside window for NSB pixels

[FrancaCassol]

Hi,
since the charge extractor finds the pulse time by weighting the sample time with its charge, in the case of NSB pixels the charge can be negative and this gives strange pulse times. There is a ctapipe a PR from @watsonjj to exclude negative samples from the calculation cta-observatory/ctapipe#1143, till then we will have some pixels with time outside the expected window (but this should mainly concern pixels without signal)

[moralejo]

On Wed, 22 Jan 2020 at 16:58, Franca Cassol ***@***.***> wrote: since the charge extractor finds the pulse time by weighting the sample time with its charge, in the case of NSB pixels the charge can be negative and this gives strange pulse times. There is a ctapipe a PR from @watsonjj <https://github.com/watsonjj> to exclude negative samples from the calculation cta-observatory/ctapipe#1143 <cta-observatory/ctapipe#1143>, till then we will have some pixels with time outside the expected window (but this should mainly concern pixels without signal)

Note that negative samples will also spoil the timing of pulses of few-p.e. I did not understand what is still pending in the ctapipe change, it seems already reviewed, right?

…

-- Abelardo Moralejo Olaizola Institut de Física d'Altes Energies Tel : +34 931641662 Fax: +34 935811938 Avís - Aviso - Legal Notice - (LOPD) - http://legal.ifae.es

-- Avís - Aviso - Legal Notice - (LOPD) - http://legal.ifae.es <http://legal.ifae.es/>

[vuillaut]

It's not pending, the issue has been addressed and the PR accepted long ago. It will be integrated in lstchain with the next release of ctapipe.
But that should not be a concern after cleaning (and the timing parameters are computed on cleaned images).

[FrancaCassol]

Hi @vuillaut,

It's not pending, the issue has been addressed and the PR accepted long ago. It will be integrated in lstchain with the next release of ctapipe.

you are right, I overlooked it. Perhaps we should slowly start to updated the ctapipe_master branch so to be ready to move to the new ctapipe release when available

But that should not be a concern after cleaning (and the timing parameters are computed on cleaned images).

My comment was related to the pulse time per pixel, which is calculated by the charge extractor at the calibration level before the cleaning

[moralejo]

On Thu, 23 Jan 2020 at 00:43, Thomas Vuillaume ***@***.***> wrote: It's not pending, the issue has been addressed and the PR accepted long ago. It will be integrated in lstchain with the next release of ctapipe. But that should not be a concern after cleaning (and the timing parameters are computed on cleaned images).

Cleaned images may also contain pixels with, say, 4 or 5 p.e., and those in their waveform may occasionally have a negative sample (in the tail of the integrated pulse). Of course, since negative fluctuations are not very large, the effect in the weighted average is not huge.

…

-- Abelardo Moralejo Olaizola Institut de Física d'Altes Energies Tel : +34 931641662 Fax: +34 935811938 Avís - Aviso - Legal Notice - (LOPD) - http://legal.ifae.es

-- Avís - Aviso - Legal Notice - (LOPD) - http://legal.ifae.es <http://legal.ifae.es/>