ControlLogix--Prosoft--Modbus--JUMO reciever?

[sirius0 1]

Hi All, I am trying to setup a JUMO wireless temperature probe receiver to get data into a ControlLogix PLC via a ProSoft card. When I checked status on the ProSoft card (MVI56-MCM), via hyperterm, I found that I have the card visible (node 20). Also I seem to have all the connections settings correct. But I believe I am having device address and/or Internal address issues. Has anyone set one of these up before? I would really be interested in a sample list of what was entered against the MCM tags to get it reading. Thank you. Chris

[sirius0 1]

Ok sixty three views so maybe some are interested but have nothing to say. So i will journal my progress here. I now have the JUMO receiver set up and i get probe IDs transferred to the PLC. The temperature of a probe comes across as well, however, it is a 32bit float. This gets dumped into two 16bit consecutive tags on the MCM tag list. The second tags MSB is the signing bit, the next eight bits are the exponent followed by a 23bit (24th bit is inferred, this is a IEEE754 32 bit float) mantissa. I need to extract a decimal to 2 places after the decimal point.. Any suggestions as to what commands to use? Regards Chris

[y2kmfic 3]

Getting two integers from your MCM.ReadData in to a float is pretty simple. Create a COP instruction, where the source starting address is the first integer, and copy that in to a Real tag. Since the length follows the destination, your length would only be 1, since you only have 1 float/real tag. Once you do the copy, and your data appears to be way off, you may need to experiment with the swap code to swap the bytes, words or both. CH Edited 4 Sep 2013 by y2kmfic

[sirius0 1]

Yes you are correct. Sorry everyone but I had made progress and not posted. I have actually used the CPS command and pointed it to the first 16 bit word, I set the Length to 32 bits so that the second word would be picked up the destination is a REAL. The REAL in Allen Bradley is a IEEE754 as well so it worked straight up. i used CPS instead of COP as the bit pattern with COP can be updated during the read. In the IEEE754 the eight bits after the sign bit are actually a two's compliment (or 127 bias) exponent AND the value is normalized. This all means that the smallest change in data is a large shuffle to the bit pattern. If the mantissa has been copied whilst the exponent has changed to suit another mantissa then there is a big change to the value if not an error. CPS freezes the data whilst reading and is this instance a lot more reliable. Chris