Troubleshooting an RS485 network

We have a wonderware station with an RS232 link from a PC master going out to SEW drive slaves, to a protocol converter, then an RJ-11 interface to another device that is wired up to the rest of the drives. I can't use the utility on the PC to poll the drives any longer to check on information from them and have been tasked with repairing this issue.

Of course, I want to replace the cabling between the PC and the device that connects to the drives to get their status, however due to the fact that this isn't an immediate problem I would like to make a learning experience of this.

What would be a good way to check the traffic leaving the PC itself so I can start tracing out this issue. Also is there a way I could check the data leaving the protocol converter? Perhaps a device I could use?

1) Has this network ever worked, or is this 'commissioning'?

2) What exactly is the problem that you want to solve?

3) An RS-23-to-RS-485 converter is technically not a protocol coverter, but is that what you mean by "protocol converter"?

4) What is "another device" with the RJ-11 that is "wired up to the rest of the drives"?

5) There are hundreds of protocols (rules and language) that talk over RS-485. What protocol are these devices talking? Modbus RTU? Modbus ASCII? Profibus PA? BacNet, LonWorks? vendor proprietary protocol?

1) This network used to work. It is used by a series of SEW drives to communicate and interface with a PC app designed for the purpose.

2) The PC app is no longer able to communicate with the drives. It polls (or "searches") for devices, but will not find any.

3) Yes, this is an rs232 to rs485 converter

4) One side of the converter (from the PC) is a Serial connection, DB9 to be more specific. The other side of the device is an RJ-11 jack and what resembles a phone cable that connects into the next device. This device has an RJ-11 connection and is daisy chained into another cabinet with the drives. It is a BUS type network most certainly, but I'm not sure what the protocol is. I will find out. They are all SEW drives and I'll get more information on them shortly, as I couldn't find documentation at the plant for the drives I'll have to get part numbers and pull it up.

The RS-232 on the PC can be 'loop back' tested with a terminal program and an adapter the connects Tx to Rx to prove it's operating. Google 'loop back' to find a drawing of the adapter.

If the software is 'searching', the RS-232 Tx (transmit) LEDs on the 232-485 converter should blink. Does it? Does the 485 side show any activity?

Does the 232-485 converter use a power supply? Is it still plugged in a operating?

Until you find out what the mystery box is, it's hard to get beyond that point.

The first two tools I would get out are an RS-232 breakout box and an oscilloscope.

The breakout box would go inline between the PC and the 232/485 converter, just to see if there are signal lines changing states at all. If there's no action on the CTS/RTS handshaking line or on the TX line, then the PC's port is damaged or the software isn't even trying.

If the RS-232 side seems to have some action, I would put an oscilloscope between the Data A and Data B lines on the RS-485 side to see if there are any transmissions being seen there.

The very best tools in my toolbox for RS-232 and RS-485 troubleshooting are the EZ-Tap and Versa-Tap from Stratus Engineering. These are an amazing value for money if you work on serial data transmission.

But just to figure out which element in the system is broken, an ordinary scope will be fine.

"Until you find out what the mystery box is, it's hard to get beyond that point."

The mystery box is probably the SEW gateway to convert RS485 to SEW SBUS

The RJ11 connectors and cables are the SBUS network.

The information can be monitored with the help of the software "Movitools".

Ken Roach said:

The first two tools I would get out are an RS-232 breakout box and an oscilloscope.

The breakout box would go inline between the PC and the 232/485 converter, just to see if there are signal lines changing states at all. If there's no action on the CTS/RTS handshaking line or on the TX line, then the PC's port is damaged or the software isn't even trying.

If the RS-232 side seems to have some action, I would put an oscilloscope between the Data A and Data B lines on the RS-485 side to see if there are any transmissions being seen there.

The very best tools in my toolbox for RS-232 and RS-485 troubleshooting are the EZ-Tap and Versa-Tap from Stratus Engineering. These are an amazing value for money if you work on serial data transmission.

But just to figure out which element in the system is broken, an ordinary scope will be fine.

Click to expand...

Thank you Ken, you nailed my question!

nonuke said:

"Until you find out what the mystery box is, it's hard to get beyond that point."

The mystery box is probably the SEW gateway to convert RS485 to SEW SBUS

The RJ11 connectors and cables are the SBUS network.

The information can be monitored with the help of the software "Movitools".

Click to expand...

Also you are right, that is a gateway of some sort. I'm rather ignorant when it comes to all these different types of networking and protocols. I have studied a lot, but I'm just now really getting hands on with it.

We have scopes at work but no leads. I keep telling them we need leads haha!

Thanks for the help guys. I determined the RS232 port on the computer to be defective (it was an addon card) and replaced it and the network is back up and functioning normally.