When I build an executable of a working vi that uses serial I/O, it crashes if the serial I/O comes from a USB-RS232 converter. It works fine on a hardware serial port, and it works fine when running the vi within LabVIEW. The exe doesn't crash right away, it crashes in about 10 sec if I use the default VISA timeout. If I increase the timeout, I can make the exe run for as long as 3min47sec, but not longer.
Baud rate, #of bytes to read, etc... don't seem to make a difference. If no serial data is sent to the vi through the serial port, it doesn't appear to crash. The bytecount in the buffer never exceeds 200 and no serial data ever seems to be lost before the crash.
I have seen this on two different model Lenovo (IBM) laptops, running WinXP. I have noticed that on a Dell Desktop PC an a Dell Laptop, the exe seems to run fine, even when using the same USB-Serial converter that causes problems on the IBM laptop. I have tried two different brands of USB-Serial converters but they exhibit the same problems. I'm running LabVIEW 6.1, but I'm told that this is unlikely to be resolved if I switch to 8.2.
Does anyone have ideas of what could be going on? Are there workarounds or fixes?
I use a USB-RS232 converter regularly on my HP laptop, with no problems whatsoever. Can you post your code, so others can test it? It sounds like an IBM issue with the USB port, not a LabVIEW problem.
