We are looking to acquire velocity using 2 infrared velocity screens hooked up to a BNC-2090 connector block and a PCIe-6351 DAQ card...I am completely new to Labview so any examples/tutorials/etc that you could point me towards would be great.
Could you please elaborate so that we can better assist you? what type of output do you expect from your infrared screens? Do you have specific models for these? Thanks.
Daniel G. Semiconductor & Wireless National Instruments
The components we are using are the PCIe-6351 DAQ card, a NA BNC-2090 board, a PCB Charge Amp #462B52, PCB piezo transducers, and 2 Oehler #57 screens. In the past we have used Oehler 82's and 85's to get chamber pressures and velocities, but we know that Labview is capable of doing the same thing with more flexibility and accuracy. I am in the process of going through the Core 1,2, and 3 classes online. Any more help or suggestions would be appreciated.
The screen provides a nominal +12 volt pulse of approximately 2 ms duration from a panel mounted BNC connector. The output can be approximated by 50 ohms to +12v in the high state and by 10K ohms to ground in the low state. Rise time is approximately 0.1 microsecond. More than one pulse may occur for the passage of a single projectile.
If this were me I'd condition the signals to 0-5V TTL, then wire them into one of the counters on the 6351 (using the PFI inputs). You can configure the counter to take a two edge separation measurement in which it counts the number of timebase ticks between two external signals. The timebase frequency is known (you would use 100 MHz on X Series) and so the driver uses this information to report the time between the occurrence of the two edges. From there, calculating velocity is easy (assuming you know the distance etween your screens of course). This would be the most accurate way to take the measurement on the 6351.
I opened up NI Max to set up the DAQ card and connector block...I was in the Data Neighborhood tree...does that sound right? It allowed me to set the max/min voltages, but I am kind of lost where you talked about taking a 2 edge separation measurement to count the number of timebase ticks between the 2 external signals. Thanks again for all the help!
MAX gives some basic functionality for configuring pre-made tasks, but for your application you'll want to program the task in LabVIEW (or your ADE of choice) using the DAQmx API. There is a (too) basic example on NI's website which shows two edge separation--you might check the example finder for something more up-to-date if you are using LV 2012 as the DAQmx examples got a revamp with this release (from LabVIEW, go to Help >> Find Examples...).
One caveat when not using buffering is that the call to DAQmx Read is what actually arms the counter. I'd suggest sticking with this approach for now as it is simple, but requires you to specify a timeout on your read function and to fire the shot before the timeout expires (but after the read call is made by the software).
Once you get to this point, if you want to eliminate the awkwardness of having a call block for so long (and avoid timeouts altogether) you can configure buffering and use DAQmx events (or just poll available samples to read) to tell you when a shot occurs. Don't pay any attention to this now though, come back to this later.
Before putting it into code, it might be beneficial to try to understand what it is you are trying to configure in the first place. The X Series user manual gives a great overview of the various counter modes and capabilities. You want to implement the case presented in figure 7-24 (link to manual pdf). To use the counter functionality, the input signals must be TTL voltage levels--don't try to connect the 12V directly to a PFI line.
It might be tempting to try to sample the signals with analog inputs and compute the time difference in software, but this would be less accurate (500 kHz per-channel sample rate with 2 channels gives 2 us resolution compared to 10 ns resolution of the 100 MHz timebase) as well as more complicated to program. You're better off converting the external signal to TTL and using a counter.
I appreciate your time and responses...I have been slammed here at work and am just now getting back to setting up our Labview.
In one of your earlier responses you mentioned conditioning the 12v signals coming from the optical screens to 0-5v TTL. How exactly do I go about doing that? I have been using the Measurement I/O in the DAQ-mx but have not been very succesful.
You have a 12V pulse with 50 output impedance, after some meter of 50 ohm coax cable , and to avoid reflections, you should use a 50 ohm termination, which results in a 6V signal.
One might say fine, ignore that 1V , but you can do better and build a 50 ohm termination with two (SMD) resistors , 33 Ohm and 18 ohms or 30 and 20 ... and you have a ~5V or ~4V signal 🙂
SMD fits nicely in a BNC adaptor , just avoid wire wound resistors 😉
For the pressure sensors: The calibration is done static, but you want to measure a quite dynamic signal .... what uncertainty do you expect?
