Specifications

12 CHAPTER 3. PERFORMANCE OF THE DACS
3.3 Discussion
Once again it is stressed that the experiments described above are only mend to give a first im-
pression. No hard conclusion can be drawn, merely some guesses of what to do to improve the per-
formance and accuracy, because the DACS is not completely finished. Any change to the DACS will
have an effect on the performance and accuracy. So a detailed survey is only useful after completion.
Still it may be clear that the errors are considerable and increasing with the amplitude of the
signals. First of all, the direct connection of the input and output as depicted in figure 3.1 may cause
ground-loops. This could be the reason for the large amount of noise on even small amplitude signals.
During the building process of the CBox, there was some discussion about connecting the ports
single-ended or differential. The connector pcb in the CBox was soldered three times because of this.
As a result the connector pcb short-circuited twice already and is expected to do so in the future. Of
course this can have a negative influence on the signals as well, but not on all signals.
Finally, the PowerDAQ PD2-MF-16-500/16H was manufactured on 01-Jan-2003 and calibrated
on 21-Nov-2003 and PowerDAQ PD2-AO-16/16 respectively on 01-Oct-2003 and 24-Nov-2003. The
manual of the PowerDAQ PD2-AO-16/16 board recommends a calibration every year. There is a Pow-
erDAQ software tool under windows to recalibrate the board, however it requires a high-precision
voltage source. Another option could be to adjust the equation which calculates the voltage from the
rawbuffer. Of course before any of this takes place the errors have to be identified as being determin-
istic.