We need a USB connected "something" to timestamp (ms resolution is good) receiving a trigger which is only 10 microsec long and with amplitude 3 or 5 V. Those triggers would be coming in 1-3 sec intervals. What would be our "best" solution? I just see that if we are to do trigger detection "in software", we would need high sampling rate available only in expensive T7 series. But I wondered if anything could be done with the cheaper devices?
Thanks in advance for ideas
.png%3Fitok=PHGBxf6M "EEV Blog Reviews LabJack T7-Pro & Software Tools")
.jpg%3Fitok=B1YcuDFK)
.png%3Fitok=Ybt-OlgU "LabJack + JEDI One Can Create a Powerful Distributed DAQ Ecosystem")
Use a high-speed hardware counter to count the pulses, as it will not miss a 10us pulse.
How accurate do you want the timestamps to be? Are you more concerned with the relative time between pulses, or is it important to know the time of each pulse according to the system time on the host computer?
Need to know the time stamp (count is not important per se). But we do not need super-high precision -- 1 ms precision would already be high enough (although sure thing the higher the better ;-), ideally not less than 10ms). One of the overall needs later is to use those time stamps to estimate offset/drift on the external hardware which sends those pulses, and which is "offline" so we could not use ntp client on it to maintain its internal clock in sync.
With the lower cost U3 (USB only) or T4 (USB & Ethernet) you should be able to get host system timestamps of about 1 ms. Software simply polls the counter value as fast as possible in command-response mode (typically about once per 1 ms or better) and whenever it sees that the value has incremented it notes the host system time. This would be slickest with some simple custom software. Are you able to do a little programming?
https://labjack.com/support/t4-software-options
Programming -- shouldn't be a problem.
Sounds like a plan. The only remaining question s-- e.g. with the U3,
- how short/weak of a (square form) pulse it could detect to increment the counter? Couldn't find at its spec sheet, just wanted to make sure that it should be capable of detecting our short (10 microsec) pulse
- we should get HV version since pulse is at 5V and LV talks about "0-3.6 V", correct?
thank you for all the clarifications!
See note #16 here:
https://labjack.com/support/datasheets/u3/appendix-a
The hardware counters on the U3 support up to an 8 MHz square wave which is a high time of 62.5 ns and a low time of 62.5 ns, so it will count pulses down to 62.5 ns.
LV or HV is fine. The counter uses a digital input so look at the DIO specs. All digital inputs are 5-volt tolerant meaning that they are happy with 5 volt signals:
https://labjack.com/support/datasheets/u3/hardware-description/dio