Abstract
This study deals with the harmonic analysis of the instantaneous counting rate of a pulse train. This arises from using a fission chamber for reactivity to power transfer function measurements by oscillation methods in reactors. The systematical errors due to the sampling process are computed. The integration carried out when sampling the signal modifies the formulae of the Nyquist theorem on spectrum folding. The statistical errors due to the noise are analysed: it is shown that the bandwidth of the spectral window applied to the noise frequency spectrum is equal to the inverse of the time duration of the experiment. A dead time of 25 per cent of the sampling time does not increase appreciably the bandwidth. A new method is proposed afterwards yielding very approximate results of the Fourier analysis during the experiment. The systematical errors arising from the measuring process are determined, and it is shown that the bandwidth of the corresponding spectral window is still given by the inverse of the time duration of the experiment. (author) [French] Cette etude se rapporte a l'analyse harmonique de la valeur instantanee du taux de comptage d'une suite d'impulsions. On rencontre ce probleme dans l'utilisation de chambres a fission pour
More>>
Freycenon, J
[1]
- Commissariat a l'Energie Atomique, Saclay (France). Centre d'Etudes Nucleaires
Citation Formats
Freycenon, J.
Theoretical study for a digital transfer function analyser; Etude theorique pour un transferometre digital.
France: N. p.,
1964.
Web.
Freycenon, J.
Theoretical study for a digital transfer function analyser; Etude theorique pour un transferometre digital.
France.
Freycenon, J.
1964.
"Theoretical study for a digital transfer function analyser; Etude theorique pour un transferometre digital."
France.
@misc{etde_20793543,
title = {Theoretical study for a digital transfer function analyser; Etude theorique pour un transferometre digital}
author = {Freycenon, J}
abstractNote = {This study deals with the harmonic analysis of the instantaneous counting rate of a pulse train. This arises from using a fission chamber for reactivity to power transfer function measurements by oscillation methods in reactors. The systematical errors due to the sampling process are computed. The integration carried out when sampling the signal modifies the formulae of the Nyquist theorem on spectrum folding. The statistical errors due to the noise are analysed: it is shown that the bandwidth of the spectral window applied to the noise frequency spectrum is equal to the inverse of the time duration of the experiment. A dead time of 25 per cent of the sampling time does not increase appreciably the bandwidth. A new method is proposed afterwards yielding very approximate results of the Fourier analysis during the experiment. The systematical errors arising from the measuring process are determined, and it is shown that the bandwidth of the corresponding spectral window is still given by the inverse of the time duration of the experiment. (author) [French] Cette etude se rapporte a l'analyse harmonique de la valeur instantanee du taux de comptage d'une suite d'impulsions. On rencontre ce probleme dans l'utilisation de chambres a fission pour les mesures de fonction de transfert reactivite-puissance par la methode d'oscillation dans les piles. On calcule l'erreur systematique due au processus d'echantillonnage ou l'integration operee modifie les formules classiques de recouvrement du spectre. On analyse ensuite les erreurs statistiques dues au bruit de fond. On montre que la largeur de bande de la fenetre spectrale appliquee au spectre de puissance du bruit est donnee par l'inverse du temps de mesure. Un temps mort de 25 pour cent du temps de prelevement n'accroit pas sensiblement cette largeur de bande. On propose ensuite un procede simple qui permet d'obtenir, en cours d'experience, des resultats tres approches de l'analyse de Fourier. On determine les erreurs systematiques dues a cette methode, et l'on montre que la largeur de bande de la fenetre spectrale correspondante est encore egale a l'inverse du temps de mesure. (auteur)}
place = {France}
year = {1964}
month = {Jul}
}
title = {Theoretical study for a digital transfer function analyser; Etude theorique pour un transferometre digital}
author = {Freycenon, J}
abstractNote = {This study deals with the harmonic analysis of the instantaneous counting rate of a pulse train. This arises from using a fission chamber for reactivity to power transfer function measurements by oscillation methods in reactors. The systematical errors due to the sampling process are computed. The integration carried out when sampling the signal modifies the formulae of the Nyquist theorem on spectrum folding. The statistical errors due to the noise are analysed: it is shown that the bandwidth of the spectral window applied to the noise frequency spectrum is equal to the inverse of the time duration of the experiment. A dead time of 25 per cent of the sampling time does not increase appreciably the bandwidth. A new method is proposed afterwards yielding very approximate results of the Fourier analysis during the experiment. The systematical errors arising from the measuring process are determined, and it is shown that the bandwidth of the corresponding spectral window is still given by the inverse of the time duration of the experiment. (author) [French] Cette etude se rapporte a l'analyse harmonique de la valeur instantanee du taux de comptage d'une suite d'impulsions. On rencontre ce probleme dans l'utilisation de chambres a fission pour les mesures de fonction de transfert reactivite-puissance par la methode d'oscillation dans les piles. On calcule l'erreur systematique due au processus d'echantillonnage ou l'integration operee modifie les formules classiques de recouvrement du spectre. On analyse ensuite les erreurs statistiques dues au bruit de fond. On montre que la largeur de bande de la fenetre spectrale appliquee au spectre de puissance du bruit est donnee par l'inverse du temps de mesure. Un temps mort de 25 pour cent du temps de prelevement n'accroit pas sensiblement cette largeur de bande. On propose ensuite un procede simple qui permet d'obtenir, en cours d'experience, des resultats tres approches de l'analyse de Fourier. On determine les erreurs systematiques dues a cette methode, et l'on montre que la largeur de bande de la fenetre spectrale correspondante est encore egale a l'inverse du temps de mesure. (auteur)}
place = {France}
year = {1964}
month = {Jul}
}