rootboy

MrPLC Member
 View profile  See their activity

  • Content count

    11

  • Joined

  • Last visited

Posts posted by rootboy

  1. Posted · Edited by rootboy

    Get yourself a Modbus monitoring program such as: http://www.simplymodbus.ca/RTUslave.htm or: http://www.modbustools.com/download.asp (or try out one of the free ones) and use that to debug your connection. If you are still having problems, paste a few examples of what you are sending out of the PLC with the coresponding results as seen by your Modbus monitoring program. While working for Schneider I had the opportunity to interface several different devices to our ION meters, and the typical "Failure Mode" was: a) The physical connection (this is where com adapters that have blinky lights are a big help. I recommend this one: http://www.usconverters.com/index.php?main_page=product_info&cPath=75&products_id=280). b) The target's interpretation of the Modbus spec (and arguably Schneider's as well). By breaking the problem down into halves, you get to see what you are transmitting, and if you get the Modbus Master as well as the Slave, you can try out different messages to the GE relay to see what works for it. Here's an article concerning memory mapping (and endianess): http://www.controldesign.com/articles/2010/ModbusDevicesINQ110.html We had problems with a steam meter until we figured out that while the documentation said register 4001, it really wanted 40001. Be sure to try this as well. And don't forget the age-old Modbus conundrum of whether the register 4001 is mapped to 4001, or 4000 (some implementations of Modbus start at one, some at zero). If 4000 is the case, you will have to offset everything down by 1.

  5. Posted

    we have one series six left, it's on a mold setter I think. Probably not going to need to work on it anytime soon, but it's good to know that it can be done. thanks!

  10. Posted

    We had this exact same problem with our S6's. Random timers would time out in a fraction of their preset time. The PLC code was stable enough that no one really bothered to look to see if something had ended up being used twice (and we would have caught that sort of mistake in Ladders). What it ended up being was leaky CPUs on the Arithmetic Control Module (according to ATS, who fixed the boards for us). What they did to fix the problem was to replace the four 4-bit CPUs on the board. This did the trick, no more problems. Here is one for sale (I really wanted the link for the photo), the CPUs in question are the four gold-plated ones: http://www.ebay.com/itm/GE-FANUC-IC600CB500A-CONTROL-MODULE-ARITHMETIC-ALU2-SERIES-6-/221045324299 Before you drop $65 bucks on an iffy ACM :) We had a way of determining if the problem was with the CPUs or not. We would cycle power on the PLC and if that cleared up the problem (even temporarily) then the CPUs needed to be replaced. And a further bit of tribal knowledge, never grab the ribbon cable that connects the Arithmetic Control Module to the Logic Control Module by the ribbon to pop it off. They break easily. Always use the pull tab on the connectors.

  11. Posted

    Really? We no longer have many S6's, but near the end we tried to get LM6 up and running on XP. It would boot..., but that was about it. Do you still have your setup? Were you using ANSI.COM? (rather than ANSI.SYS) Thanks, John