Interfacing sensors to measurement equipment, whether a microcontroller ADC input or a high-end data-acquisition system, can cause headaches, particularly when you must use an unusual or unfamiliar sensor. Sensor-interfacing challenges seem to have gotten less onerous, but from what people have told me, some interfacing tasks still cause them fits. Here’s my story:
In grad school–with limited time, talent, and money–a fellow grad student and I faced the job of interfacing a photo-detector array of 256 elements to a data-acquisition (DAQ) “box” (Biomation 610B) sold as a high-speed digital storage unit for oscilloscopes. Somehow we had to drive the detector, a Reticon device, and synchronize its scan rate with the DAQ box. Thankfully we had a schematic for the DAQ box and we had at least some of the multi-phase driver electronics for the array. I can’t remember which one of us figured out how to tap into a clock signal in the box and use it to synchronize the scans of the photodetector array, but I know it took a lot of testing, added circuitry, and many experiments to get the timing just right.
The DAQ box used a flash ADC, so in essence it continuously converted an analog signal into a digital value. As best I recall it didn’t have a start-sample signal we could use directly. The signal from the detector array comprised a series of return-to-zero analog pulses, one per detector element with amplitude proportional to light reaching a detector. So the timing of the ADC had to coincide with the peak amplitude of the pulse from each detector, and we had to grab 256 samples. No data sheets explained how to make this happen, so we solved the interface timing problem through perseverance. The project also involved two computers, optics, and a lot of software, but I’ll leave that story for another time.
What’s your challenging sensor-interfacing story? –Jon Titus