finite pulse train generation?

[Paco 565]

i am trying to generate a very specific pulse with my tio-6602 in labwindows. after a 20us delay, i want the board to generate 20 pulses, with a width of 1 us and an interval of 5us. ideally, this would be retriggerable at 100 Hz. any suggestions on what would be the best way to approach this problem?

[Geneva L.]

Paco,

You can use a gated pulse train, which is similar to finite pulse train generation. With finite pulse train generation, you have one pulse for the gate of a counter set to generate a pulse train. The pulse train will only be output based on the gate, so if one pulse appears, you have a finite pulse train. Now, if you make the gate be a pulse train itself, then you can continuously output bursts of pulses. For instance, you can set it up such that the pulses are output when the gate is high. When the gate is low, the counter pulses would pause and when the gate resumes to the high states, your pulses would continue. I've attached a code excerpt for using NI-DAQ function calls to generate a finite pulse train that is in our database. I will see abo
ut getting it put onto the http://www.ni.com/support pages. You can also find more information about the function calls used by referencing the NI-DAQ Help file installed with the NI-DAQ driver software and with the NI-DAQ User Manual, which is available at the web pages.

Regards,
Geneva L.
Applications Engineering
National Instruments
http://www.ni.com/ask

[Paco 565]

Geneva,

thank you for the documentation on finite pulse train generation. i was able to create the 20 pulses needed. however, i have been unsuccessful so far in creating the 20 us delay. it seems that i would need another counter to do this (first counter to trigger the entire sequence, second counter to create the delay, third counter to generate the 20 pulses), but when i try to do this, the squarewave becomes unstable...seems like the counters are unable to sync?

Paco

[ejk]

I tried the code, but the first pulse of the pulse train wanders through the gate pulse. Because of this, the output of the pulse train sometimes stays high when the gate ends on a pulse. This is a problem for me. Is there something more I need to do in order to syncronize the pulse train with the gate?