doing fft of data aacquired

[Uglam]

Hi, I am trying to develop a system that will acquire analogue data, digitize them through NI 5752 and do fft upon them by Flex RIO fpga (PXIe 7965R).

For doing that I tried to combine two examples, 'NI 5752 finite acquisition with external clock' and 'FFT Co-processor'. The newly built project (FFT Co-processor for External Signal.lvproj) with host (FFT Co-processor for External Signal (Host).vi) and fpga (FFT Co-processor for External Signal (FPGA).vi) vi are attached here. I am not using the other vi in the project file. Hence not attaching that one here.

When I run the programme, an error message comes, which tells "Error-50400 occurred at Invoke Method: FIFO.Read in FFT Co-processor(Host).vi.....Possible reason(s): NI platform services: The transfer did not complete within the timeout period or within the specified numbers of retries.

I have already run both the examples separately on the pc, and they worked fine. I have tried different ways to solve the problem or find the specific reason behind the problem but did not find a clue. As they are lot in number, I am not writing them here.

Thanks in advance for any body attempting to help me.

[Andy.G]

Hey Uglam,

I have had a look at the project you have attached and it would appear both VI’s are blank? I am familiar with both the aforementioned examples and have an idea of what you are trying to achieve.

For the application you are trying to develop the combination of these two examples would not be the ideal solution and in fact makes very little sense. As essentially what you would be doing is acquiring data and reading it on the host pc and then writing that data to another card and finally writing the processed data back to the host. This would slow down the process and would involve a lot of unnecessarily complication to the overall design.

You would be better to make a physical connection between the cards and route the acquired and processed signal from one card to the other directly. This way you would only have a single DMA from the card to the host.

I would like to re-enforce that the examples are not designed to be solutions, they a tailored to achieve a specific purpose and therefore modifying them or “bolting” them on to one another is not always the best approach to take.