PLC-5 and Mutliple RSView32 PCs - Occasional Wireframe of Tag Data

I have been working on and off with a customer who has a PLC-5/80E that is communicating with five different PCs that are running RSView32 v 7.50.00 CPR 9 SR1. Most are desktop PCs running Windows 7, and one is an old Advantech PC running windows XP. There is one PC running Windows 7 that has issues from time to time where some or all of the tags on one or more screens will "wireframe" and not update information. If the PC is rebooted, the problem will go away for a day or maybe a week. It eventually returns, and it is not always the same screen.

All of the PCs run the same project, with only some differences to account for running on a dual monitor or single monitor system. (Different macros are called up to allow screens to be put on the right or left on a dual monitor system). Other than this, it is the same project. There are two systems that have dual monitors, and three PCs with one display. That is probably not so important.

The communications issues seem to be only occurring on one of the two dual monitor systems, and it happens to be the one PC that is used most by the operators. The node comms for all RSView32 PCs is configured to use "Channel 2 - TCP/IP" and PLC-5 Enhanced for the type. The project uses the Direct Driver. I increased the timeout on the node comms to 5 seconds. This did not help. I did not design this, and it looks like they may have had these issues for awhile. I thought that is I am using the Direct Driver, that RSLinx is not a factor? I might be wrong about that. I have only worked with RSView32 a few times, and it was with a ControlLogix PLC, not the PLC-5/80E.

I can look at the TCP/IP diagnostics of the PLC in a web browser.

I am looking through some of the activity logs to see if I can find anything that might be related. Currently the problem RSView32 PC is updating everything fine. Unfortunately the activity log files over the last few weeks are gone, but I am going to look through any new ones for more info.

I have tried Rockwell tech support a couple of times, but have only had some suggestions about re-installing the software.

Curious if anyone has seen anything similar or had a solution. This facility is planning to upgrade and move away from the PLC-5 and RSView32 platform, but this will not occur for awhile.

I apologize. I see I had a similar thread some time ago. I did try to delete this one.
One thing that someone suggested was to go to Windows and look at the local area connection in Control Panel. Under Local Area Network Properties if I look at the Configure button I should be able to find info on how the PC is connected to the Ethernet network. (I have no info on switches, etc. or how this was ever setup way back when...) When I do this, I can look under the Link Speed and I see that on the PC for the Intel 82579LM Gigabit network connection properties the speed is 1 Gbps, and the speed and duplex is set at Auto Negotiation on one of the machines. Now, the machine I just got this from is one that the operators do not use and is more of a maintenance PC for running trends, etc. I was going to go and look at this info on the operator machine, but I wanted to see how this was done before I bothered the operators. I am pretty sure the PLC-5/80 in question can only do 10 MBps, and mybe this is causing some of the problems on the high use PC?

I thinking along your lines now. You really have to dig in and see how much data is being served. I wouldn't rule out a physical problem with that computer's network connection either. In fact, I would likely start there by testing the cable, seeing if you can ping the plc or other systems on the network.

As you go along take notes on the network architecture, of course.

Most PLC-5E controllers had a AUI transceiver port that was only capable of half-duplex 10 Mb/s operation. Series F controllers made in about the past ten years have a 10/100 auto-negotiating port with a conventional RJ45 jack, and should be labeled "10/100".

In general, switched Ethernet networks handle different link speeds easily. The most likely place it would be a problem is between the PLC-5E itself and the switch it is plugged into If you've got an ordinary 10 Mb/s PLC, you should have auto-negotiate turned off on the associated switch.

The RSView32 Direct Drivers still use RSLinx, but each of your computers is independently reading data from the PLC-5. Because all the applications are similar, you're ostensibly putting 4x as much traffic onto the PLC-5 as you need to.

If I was going to do an upgrade to improve system performance, I would choose one of those computers (or install a separate one just for this purpose) to run RSLinx Classic Gateway, so that all the RSView32 instances could be OPC Clients. The RSView32 stations would get their data fast from a powerful PC over OPC, while the PLC-5E itself only has to talk to the one RSLinx Gateway station.

The downside of that, of course, is that you have to get all the computers configured so that they have the appropriate security and networking settings to allow OPC data to flow between them.

For troubleshooting, start by checking to be sure there's nothing profoundly wrong with the interfaces on the PLC and computer, then start looking at the RSLinx performance and logs on the affected PC.