PXI-8433 Transmitting Negative Numbers

[JordanJMcBain]

I have a serial device to which I am attempting to send negative commands but I am having difficulties.  The device receives positive commands and processes them in a correct fashion but cannot interpret negative commands.  The protocol consists of 5 bytes - one lead character (one byte), one command character (one byte), two bytes for the command payload (data high followed by data low), and a checksum.  The payload requires two's complement and I have this coded.

Can anyone think of any reasons off the top of his/her head why I might be having these problems? (e.g. big endianness vs. little endianness?)

Regards,

Jordan

[eaguero]

When you say the device that cannot work with negative commands, do you mean the third party or the PXI-8433?

Which device do you have?

[JordanJMcBain]

I am certain at this point that the problem is with LabVIEW/PXI.  I can restate the question as follows: If I write a byte of 254 by itself to the serial port, the correct value is written to the serial line (as indicated by Versa-Tap serial spy device from Stratus Engineering).  If I write a U8 array with a value of 254 (FE) in it, the serial spy device indicates that 126 (7E) has been written to it.

This irregular behaviour only happens when I write a value over 127 as an element of an array to the VISA port.  I am not sure how I could write one byte at a time to my device and have it accept it but I am looking into it right now.

BTW: I am using LabVIEW 2009 SP1.

[JordanJMcBain]

If I put "ASRL End IN" and "ASRL End Out" to "none" (rather than Last bit), I can successfully write an array of bytes with a value like 254 in it.  My device, however, does not seem to recognize this.  The problem is with LabVIEW/PXI and not my device though - I feel fairly certain.

[JordanJMcBain · Accepted solution]

With "ASRL End Out" set to "None" rather than "Last Bit," I was able to write a byte array having values greater than 127.  I consider this very unusual behaviour and I suspect that it is a bug.

I would not have been able to resolve this issue without the Versa Tap serial spy device: http://www.stratusengineering.com/VersaTap.html