Just for background information here is some basic info on programming our digitizers. This is by no means a complete guide to all the possibilities, but should give you the basic idea of the process:
The first step is to call the "init" command which initializes the digitizer into a default state and returns a task handle that will be used for all following command calls. The next step is to call the functions that select the settings you will want for you acquisition. These settings include things such as input impedance, input voltage range, sample rate, trigger settings, etc. Once your settings are defined you arm the acquisition by calling the "start" command and then the digitizer waits for various triggers to occur which progress the digitizer though the various stages of an acquisition. At this point in the process the digitizer is running and acquiring data according to your specification on its own. Your software can poll the digitizer's status to find out when the acquisiton is complete. When there is data that has been captured you can call the "fetch" commands to transfer the data from digitizer's onboard memory to the PC's memory. When you are done with the acquisition task you call the "stop" command, which halts the acquisiton process on the board. Finally you call the "close" command to clean up the process in the driver.
There are many trigger events that can be defined such as "start", "reference", and "advance". You should read the NI Scope programming manual to understand what each of these does but most of them pertain to the idea of a multi-record acquisition. In a multi record acquisition you can configure the digitizer to store more than one record into the board's internal memory whenever a trigger event occurs. A record consists of a variable number of pre-trigger and post-trigger samples along with a time stamp that indicates when the trigger event occurred relative to the triggers of other records. In the most basic situations your max record size will be limited by the amount of on board memory on your digitizer. For a single record acquisition your max record size is usually equal to (Totally Memory per channel)/(sample resolution). Multi-record acquisitions can have a small amount of overhead for each record which can slightly reduce the total number of samples that can be stored. You should consult the specification sheet for the digitizer you want to use to learn the specific capabilities of your device. There is a special mode that is supported by some of our digitizers that can allow you to capture more than the size of onboard memory called "Continuous Acquisition Mode". In this mode the digitizer is storing data into it's onboard memory at the same time you are fetching data from the onboard memory and moving it into the computer's system memory. There are many things that need to be considered when using Continuous acquisiton mode. Depending on what type of bus your digitizer is using (PCI/PXIexpress/USB etc.) you may not have enough transfer bandwidth to move data out of the onboard memory before it overflows. Also many computer busses share bandwidth with other boards in the system so your overall transfer rate may not be the maximum allowed by the bus.
---------
The 5132 does not support the ability to trim the internal oscillator frequency. This feature is supported on many of our PXI and PCI digitizers such as the 5122 and 5114.
For the sake of making sure we are all on the same page I'm going to make a few definitions. Lets call the moment in time that you need to begin acquiring sampled data t=0s. From the information you have given me it sounds like you will be generating a digital trigger from your system 500 ms before this time (i.e.. t = -500ms). You need to capture 100ms worth of data which means that your record will end at t=100ms. With our digitizers you can specify a trigger hold-off time which will delay the start of the acquisition from the time the trigger occurs. This means you won't have to start storing data at t = -500ms and can instead start storing at t=0. As long as all of this sounds correct to you then I think the 5132 can store the amount of data you need. From your earlier post I wasn't quite sure if you were saying that the trigger you could generate would occur before the even you needed to capture (t = -500ms) or after the event (t = 500ms). If it does occur after the event then the problem is that you can no longer use the trigger hold-off method and instead need to use "pre-trigger samples". When you specify the acquisition settings you can tell the digitizer how many samples it should store that occurred immediately before the trigger event. The number of pre-trigger samples you can store is defined by the amount of onboard memory. The 5132 is currently only available in a 4MB/ch version. If you are sampling at 20MS/s this means you can store a maximum of 200ms in internal memory but you would need 500ms. Again, this is only an issue if your trigger signal is after the data you want to capture.
As for using an external clock at the same time as using an external trigger it is true that the 5132 only has a single input (PFI1) that is used for the external clock input and digital trigger input. I couldn't tell from your earlier posts if you were intending on capturing data on both ch0 and ch1 or if you would only be using ch0. If you are only using one of the channels then you could use the other one as your trigger input and use the PFI line for your external clock. If you did use ch1 as your trigger line the trigger signal would not need to be full scale. Basically you can set any trigger threshold level you want within the full scale signal range you select for ch1 and as long as your trigger signal has sufficient amplitude to guarantee that noise does not generate false trigger threshold crossings you should be fine. Also keep in mind that the PFI line uses 3.3V CMOS levels while the CH1 input is much more flexible as far as voltage levels so in some situations using CH1 could be preferable. Another potential option could be that if your volumes were high enough we could make a special version of the 5132 that has a higher accuracy internal clock. This would free up the PFI line for use as a trigger.
One last issue for you to consider involves the isolation method you were referring to. All of the USB isolators I have seen are limited to USB 2.0 Full Speed (12Mbits/s). The 5132 supports USB 2.0 High Speed (480MBits/s). This doesn't mean the isolator won't work but it does mean that you will have reduced transfer rates between the PC and digitizer. If you have found a High Speed isolator I would be very interested in taking a look at it.
If you have any more questions please let us know, hopefully this information is helpful to you in making your decision.
-Matt
