next up previous contents
Next: APV25 Magnetic Field Test Up: APVMUX Test (February 2001) Previous: Calibration and Noise   Contents

Crosstalk

Figure: Positive (bottom) and negative (top) outputs of the APVMUX after amplification by the repeater. A group of channels on the even APV is internally pulsed.
\begin{figure}\centerline{\epsfig{file=mux_scope.eps,height=8cm}} \protect \protect\end{figure}

Figure: Zoomed view of the APVMUX switching between pulsed and non-pulsed APV channels.
\begin{figure}\centerline{\epsfig{file=mux_scope_zoomed.eps,height=8cm}} \protect \protect\end{figure}

The crucial characteristic of the APVMUX is its switching speed. If the switches were too slow, a signal in one chip would drop its shadow in the second chip data. To test the APVMUX switching behavior, the even APV25 was pulsed with internal calibration, while the odd chip, connected to the same multiplexer channel, remained at pedestal level (see section [*], p. [*], for APVMUX details). Fig. [*] shows the corresponding output of the APVMUX after amplification by the repeater board. This oscilloscope screenshot is the measured equivalent of fig. [*], p. [*]. Zooming into the region of the pulsed channels, it is obvious that the switching procedure is extremely fast and precise (fig. [*]). A rise time of approximately $3\,\rm ns$ has been measured, including the amplifier on the repeater board and the oscilloscope, which actually account for the major fraction of this value.

The crosstalk of signals on the even chip onto the odd has also been measured and quantified with the APV readout system. As expected from the oscilloscope screenshots, virtually no crosstalk was observed with this measurement. Fig. [*] shows the internal calibration waveforms in deconvolution mode for both pulsed (black) and adjacent (red) channels.

Figure: Averaged internal calibration waveform of the even APV25S1 chip (black) and the negligible crosstalk in the neighboring channels of the odd chip (red).
\begin{figure}\centerline{\epsfig{file=apvmux_xtalk.eps,height=8cm}} \protect \protect\end{figure}

The relative crosstalk amplitudes (tab. [*]) are about $1\%$ and thus negligible, since the noise level is about one order of magnitude higher. For comparison, a crosstalk in the order of $5\%$ was obtained with the $50\,\rm MHz$ bandwidth ADC version because of slower transients.

Table: Relative crosstalk amplitudes observed in the odd APV when pulsing the even chip.
Mode Relative crosstalk
Peak $0.7\%$
Deconvolution $1.2\%$


next up previous contents
Next: APV25 Magnetic Field Test Up: APVMUX Test (February 2001) Previous: Calibration and Noise   Contents
Markus Friedl 2001-07-14