KenE

I've got it. I talked with someone at my distributor and after confirming the settings (as you all did) he mentioned to just try a crossover cable to the module. This worked. To make a long story short, the problem seems to be related to something on our corporate LAN that was not liking the IP address I was using. I had temporarily used an IP address that was reserved on our DHCP server (ie. static) that was currently assigned to the MAC address for a spare piece of equipment (it wasn't plugged in at the time). I have done this in the past with other things and had no issues. Some piece of equipment somewhere must have been not liking the MAC address of hte mitsubishi device or something like that because it would all work when I put everything on a local switch in my office but unplugged everything from our LAN. Thank you all for your help. Ken Note: I now have a new static IP reserved for this device and it works flawlessly plugged into the LAN.

I don't have access to GX IEC Dev at the moment with this laptop (64 bit win7 machine). I've been in touch with my distributor but they haven't been able to help me out yet. Am I right in thinking that once you set the basic settings like IP address that these cards should respond to a basic ping request?I would think you don't even need to set up a "connection" setting just to get the thing to respond. In the last go around I had just a melsec connection set up but to no success. I've done quite a bit of TCP/IP programming with Windows and Linux and I just can't figure out why this is so difficult. I am definitely leaning toward something being just plain wrong. Maybe I'll see if I can get IEC developer installed in a virtual XP machine just to make sure its not the software.... Thanks, Ken

Found it, but it didn't make a difference after I formatted, re-downloaded, and then reset the processor. I started a new project with the bare minimum requirements to send to the card and no luck.

They are on the same subnet which for me is 192.168.169.xx and they are plugged into the same unmanaged switch in my office. Ughhh.... I'm starting to wonder if I have the very rare malfunctioned Mitsubishi PLC card. I only have one here for development at the moment so I can't easily swap out the hardware. I may have to get the distributor in here to verify I'm not doing anything wrong and try another piece of hardware that they bring. BTW, I couldn't find format PLC, but I did do the delete from the download screen and then re-upload. One fishy thing is that the dates on my download are for last friday morning even though I've been editing, saving, and compiling since then (and changing parameters to try different things). I'm not sure why this is that way. Maybe I'll start a new project to see if the project is somehow corrupt. thanks, Ken

I have screen shots of open setting and operation setting in the first post. They are the first two images. I have done both reset and full power up several times but that didn't work. I will now try to format the memory, it can't hurt. Thanks, Ken

As to Crossbows reply, I am using an empty spare rack on my desk, so that is why it is in the first slot. There is nothing else installed except the PLC. In reply to kaare"t, the setup of the card is as follows: slot: 0(0-0) type: Intelligent points: 32 startxy: 0000 errorresponse: clear plc operation at error: stop switch settings: all blank The quick start I had didn't say anything about switch settings so I didn't even think to go in there. I'll double check with the big manual just to make sure. I know that for modbus card you have to enter the IP address, but the setup for this card seems to be much better detailed (assuming I get it to work, at least). Thanks for the help. If you guys have any other ideas I'd appreciate it. Ken

I agree with kaare_t, this setting is just for keeping the connection open. I've seen this on TCP programming in windows and Linux. Thank you though, Ken

Hi guys, Just got a QJ71E71-100 ethernet card recently for a Q02H system, plugged it in and went through the quick start guide I have. So far I can't even ping. All I'm trying to do is set up the card as a TCP/IP server to send and receive ascii data to a PC... nothing extraordinary. Here is what I have set up currently: I cant seem to ping the thing nor can I connect to it with GXworks2 or Putty (at port 4000). The LED status is as such: RUN:ON INIT:ON OPEN:OFF SD:OFF ERR:OFF COMERR:OFF 100M:ON RD:FLASHING (activity) I'm not even sure what to try next, any ideas? Thanks, Ken

Thanks WKla, I'll have to carry around the USB-serial cable but this sounds like a good solution. For some reason I like the small rectangular displays for these small machines. Doing the pass through seems easier than routing out two USB cables or something like that. I'll touch base with our Mitsubishi distributor to double check before we buy, but this makes sense. Thanks again, KenE

Hello, I've been evaluating some simple and small HMIs for communicating to the FX3G PLC. It looks like the GOT1030 is a good fit for the applications of our small/simple machines. I've been digging in the manuals and talking to our distributor about our best options but it has been confusing. I'd like to go all USB programming but the 1030 doesn't have a USB port, so the only option would be if the FX3G had some kind pass through operation to program the HMI through the USB to the RS422 port on the CPU. Alternatively could we plug into the RS232 mini DIN on the GOT and pass through the RS422 port to program the FX3G using serial (At least this way we only need one cable to the machine)? Anyone know if either option is possible, or is there a better way? I'm sort of baffled because all of their newer equipment has had USB ports. We are thinking of upgrading to the 1040GOT, but it is much more screen than we need for our applications and the form factor is tougher to package into the enclosures of our tiny machines. Thanks, KenE

You can use the rising edge: I've been using this for a little while. It used to not be in IEC developer. It is more handy than using PLS_M in some cases. I'm using GXWorks2 1.64S ... not sure if this feature just came out or not. KenE

I'm not interested in being IEC compliant, I want the higher level of programming provided by the function blocks. Mitsubishi offers the 10ms and 100ms timer function blocks, all I want is the equivalent 1ms timer function block. If for some reason the IEC TON and TOFF worked properly I would use them, but they truncate to 100ms... kind of makes it tough to do a 20ms delay doesn't it? Ken

I think we already hashed this out except for the IEC TON, TOFF, etc. I just looked at the TON and they compiled a 1ms time value but it didn't realy work. I got a warning that it truncated it to 0ms. For 123ms input it truncated it to 100ms. So they are using 100ms timers somewhere internally. I think kaare_t's suggestion for generating a 16 bit timer clock with rollover is the closest solution. Why mitsubishi can't do this on their own software is beyond me. thanks, Ken

Hello, Yes, I have very similar code as you posted in that picture and for what I need for the 1ms timers it was fine. It doesn't help with auto assigned timer coils though. Do you know off hand if they have that 1ms free running auto rollover timer that you mentioned? I suppose I could implement that with one of the 1ms timers and some logic to reset it at the rollover if you can tolerate a 1 scan blip at rollover (I probably can). It would be nicer if this was in the system already though. I'll take a look next time I get a chance. Thanks, Ken

The only thing I know for sure is that the FX device manual states that timers TC256 and up on the FX3g and FX3U are 1ms. FX3G has 64 and FX3U has 256 I believe. I've used them on the FX3G with the out_t instruction and tested with an O-scope, so I know the hardware is there. The problem is that I can't readily use the out_t instruction *inside* a function block and have "n" instances of the function block running because I would have to hard code the timer coil number inside the FB. The problem in the USA is that most folks here are used to hard declaration of the timer numbers because they aren't used to dealing with auto-assigned device numbers like "structured ladder" (i.e. from IEC developer). I know that you, kaare_t, know what I'm talking about from your prior posts. If there is a 1ms free running timer that rolls over automatically I could accomplish the same thing in a different way as you suggest. Since FBs can easily have 16 or 32 bit local storage to save the "start time" I could do some subtraction and be done. I will look into this as it is very clean and would be universally applicable if I can pull it off. Why Mitsubishi can't use the same software core on all their processors will never stop aggravating the heck out of me. Oh well, it is what it is as they say. Thanks, Ken