The isolation is achieved by means of a series of dual input and-gates with tri-state outputs, one gate for each bit. The and–gate only allows the data on the input to appear at the output on reception of a signal from the computer. An and–gate is a device, the output of which will be identical with the logic state of the input on applying a logic 1 voltage to the second input (i.e. the output of an and–gate will only go high when both inputs are high). The operation of the and-gate system is shown in figure 37. In the upper diagram the digital output from the A/D converter appears on one input of each and-gate and, as the second inputs are inactive, the output of all gates are logical zero. On activation of the second input of each gate with a logical 1 voltage from the computer, the logic state of the output of the gates are exactly the same as their first inputs. As soon as the data has been read into the computer, the activating voltage (the enabling pulse) is removed. The output of the gates is zero and the digital output from the A/D converter is again isolated from the computer. The data now located in the accumulator of the computer is then transferred either to the computer random access memory (RAM), to a buffer store or to disc.
The parallel data transfer mode is simple to use and allows for greater flexibility than the serial mode. However, it requires numerous connections between the PIA and the A/D converter, and is, therefore limited to those cases where the converter and computer are relatively close. This type of transmission was common in chromatographs that have a built-in dedicated computer.
When the separation is complete, all the chromatographic information is stored in the computer memory as a series of binary words. Each binary word represents the detector output at a given time and the period between each consecutive word represents the same interval of time. Consequently, the computer records the concentration of solute passing through the detector at regular intervals of time throughout the separation. Thus, the chromatogram can be reconstructed by simply plotting detector reading against time (which in fact is the reconstructed chromatogram). If the chromatogram is registered on a potentiometric recorder, as opposed to a printer, the output must be passed through a D/A converter to recover the original analog signal. Typically, one chromatographic data point will occupy 2 words in memory (which for an 8 bit machine would require 16 bits). Consequently, a chromatogram 20 minutes long acquired at 5 data points a second would occupy 12,000 words. Part or all of the chromatogram can be reconstructed or, if so desired, certain portions can be expanded to permit more detailed examination of those areas of particular interest. Once the data is in memory, many different types of calculations can be carried out to help identify the components or carry out quantitative analysis.
