Tests for ADC Cards

There are mainly two aspects which are tested out in the ADC Cards, the first being the conversion per se and second, the DC offset of the analog signal. These are the basic tests that are performed to clear a card. There are a few other tests one of which is to test the coupling between the channels, which have been performed on some cards mainly to ascertain the design limitations. These tests are done only on few cards and are not done as a matter of routine certification of cards for use.

The conversion is tested by injecting a zero mean sine wave from a function generator and following it through the entire path-the amplifiers, the ADC, the lathes and the TTL-ECL converters. This tests out all the components and also the performance of the card. A tests jig is made with a ADC backpanel mounted on a a sub-rack with the power supply and clock connected. The output is acquired from either of these sources can be read on a PC where a FFT analysis shows the performance of the card in terms of the cards in terms of the spectral dynamic range obtained.

The signal after passing through the amplifier may have a DC offset, a large value of which can degrade the performance of the FFTs due to spectral leakage. This offset is tuned out in two stages. The first stage is during the ADC card testing where the potentiometers in the power supply section and in the reference generation section are tuned appropriately to make the input signal appear as a zero mean signal to ADC. This is achieved by tuning the absolute reference voltages at the two ends of the ADC reference ladder. The final voltages obtained at the two ends of the ladder are, in general, asymmetric to compensate for the DC offset of the signal. This stage of tuning is done using an oscilloscope by observing the MSB of the ADC output. For a zero mean signal, the MSB should be a square wave( i.e. a duty cycle of 50% ). The second stage of tuning is done so as to minimise the DC channel amplitude. This is done, since it has been empirically determined to be both, more sensitive as well as easier ( as compared to the oscilloscope method ).

GO BACK TO ADC SUBSYSTEM
GOTO DIGITAL BACKEND