09-12-2013 12:48 PM - edited 09-12-2013 12:57 PM
Hi, I have a digital waveform and at certain times I have transistions and glinches , how can I programmatically detect and record the time stamps of the transitions
Thanks
09-12-2013 01:01 PM
It depends, how are you acquiring the digital waveform?
Mike...
09-12-2013 01:53 PM
I use DAQ 6363 OEM and use of one the digital channels to acquire the signal. So as I mentioend my input is a digital waveform and I should monitor bit n. What I would like to do is monitor this bit and when a transition happens I record the timestamp.
09-12-2013 03:50 PM
If the period is short, you can tie the digital input to a counter channel and the counter will return an array of counts between transitions. If the time between edges is long, you can create a DAQ event that will trigger an event structure with each transition. You can use the time between events to measure the period.
Mike...
09-12-2013 04:28 PM
I forgot to mentioen that , when I acquire the data I do some processing and then I monitor the processed signal in my consumer loop. I would like to monitor bit n transition time of that digital waveform.
Let me make it simpler. I have a digital waveform which continously read 14 bits. I would like to detect the transition times in bit 11( when 0 become 1 and when 1 become 0)
Thank you for your help
09-12-2013 11:37 PM - edited 09-12-2013 11:40 PM
Let me see if I understand you. You are reading 14 bits in parallel and you want to know how long between transitions of just one of those bits - say bit 11. Is this bit number constant or can the number of the bit in which you are interested change?
Well, you could XOR the value from each read of the digital bits with the 14 bits you got from the previous read. If the resulting value of bit 11 is true, the state of bit 11 has changed. Use that flag to read/restart an interval timer. Of course the resolution of your measurement will be dependent upon the rate at which you are reading the 14 bit input. If you are only sampling the 14 bits every 500 msec, you obviously won't be able to measure intervals of less than 1/2 sec.
Mike...