Currently I am working on single photon source. To demonstrate the anti bunching I have to do the HBT experiment. I am facing some problems in the coincidence measurements. I used TAC (without MCA) for coincidence and scanned for all SCA window. Some how I am not able to find any variation in the coincidence. It looks like accidental coincidence only. If there is somebody who has already worked in this area please give some suggestions.
Do, you know the intensity correlation time (the time that it takes for intensity correlation function to decay to ) ? If you count for a time long compared with the intensity correlation time then you will only record random coincidences. For example, the correlation time for thermal light (filtered light from, say mercury arc lamp) is in the nanosecond range. So if your counting interval is greater than the correlation time you will only see random coincidences. I do not know if that is what is happening but this is something I recommend that you look into.
Do, you know the intensity correlation time (the time that it takes for intensity correlation function to decay to ) ? If you count for a time long compared with the intensity correlation time then you will only record random coincidences. For example, the correlation time for thermal light (filtered light from, say mercury arc lamp) is in the nanosecond range. So if your counting interval is greater than the correlation time you will only see random coincidences. I do not know if that is what is happening but this is something I recommend that you look into.
It's a long shot but TAC/SCA devices often require a minimum START to STOP delay. Two detectors with the same response time connected by cables of the same length, it typically doesn't work. More details needed...
I agree with the previous comments: you should have idea of the characteristic time of your transition (which should be an isolated two level system or something like that) and make sure that the the instrumental response function is short enough to measure it. The time delay between "start" and "stop" should be shifted as far as possible from zero (otherwise you risk to miss the "good" part of the g2 or have problems with the ADC channels at short delay). Moreover, the count rate of your source must be much larger than the dark counts.
I think Miroslav Jezek is right. If you connect the same detector to both start and stop you will be able to characterize the "internal" delay of your TAC because this should give you a peak at \Delta t = 0. If you do not find this peak within the scan range you need to add additional delay by using longer cables in front of either start or stop. Internal delays should be mentioned in a manual. Otherwise its a bit of trial and error. Delays of up to around 100ns may be necessary.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Optics