RS485 issues

[rmorella]

We have an RS485 based communications issue

We are using a Unitronics V430-J-RH6-N to communicate to a network of up to 4 slaves, consisting of Ziehl-Abegg Modbus modules.  The ZA module wiring requires that, in addition to the data lines, the ground connections also be daisy chained together.  This is what I am used to.  The Unitronics communications connector does not specify a ground connection on the 6 pin RS12 connector provided. They only mention using pins 1 and 6 for Data A and B. 

1. Where at the PLC do I terminate the ground for the RS485 circuit?  Currently we have it tied to chassis.

2.  Is there a possibility that tying the V0 ground at the PLC to Chassis, as suggested in the Unitronics documentation, could be causing us issues?  The ZA socuentation specifically mentions NOT to connect the shield on the cable.

 

 

 

[Ausman]

Hi rmorella,

you don't actually say what your issue is!  Does it work at all? etc?  How are you communicating with the 430 whilst trying to use the 485 port?  Pls describe the issue with more detail.

1).   I normally tie it to the earth terminal on the plc, which is tied as close as possible to the chassis ground.

2).  Are you saying that you have 0V tied to earth?  There are many different views on doing this.  But for starters I would only be using the dedicated terminals available for their specifically labelled connection.

It is curious that ZA say not to connect the shield...I'm assuming tying it to earth.  I guess they are relying on the supposed "self-regulating" nature of 485 to interference.  But having the shield connected at one end (PLC) is a must for me.   This was discussed here: 

Also don't forget termination resistors and thus any jumper settings needed on the 430 for this and 485 setting .

I have just put a link in a post to the old dedicated communication pdf.   Go here, download and have a good read.

cheers,

Aus

[rmorella]

Aus,

Thanks for the reply.  I assume that ZA is likely trying to prevent people from using the shield for signal ground, since typically the shield is grounded to chassis at one or two points (I agree with you, in that I always connect the shield only at the chassis of the PLC).  I assume their biggest concern is ground loops, and differences in ground potential, should someone get it wrong.  This is why, I question where connecting the RS485 signal ground to the chassis, which would be the case id if follow the Unitronics documentation.

My biggest issue is that there is no accommodation for the signal ground on the Unitronics modular connector used for RS485. These seems like an oversight on the part of Unitronics, since all of the RS485 enabled devices I have encountered provide a connection for RS485 signal ground.  So, my question is...   

How have others dealt with RS485 signal and chassis grounding of the Unitronics PLC with respect to reliable RS485 operation, considering that they do not offer an obvious signal ground terminal for RS485 communications in their documentation? 

I have attached a diagram of how our contractor has advised us to connect this.  The drawing shows 1 slave (there can be up to 4 slave units).   There is also a connection between 0V and chassis ground which is made at the 24VDC power supply.

Modbus-485 Connection Diagram.pdf

[Joe Tauser]

On 10/20/2017 at 4:14 PM, Ausman said:

  Does it work at all?

Well, does it?

Unitronics has never had an SG (signal ground) connection.  I'm not saying it's right, but we have to play the cards we're dealt.

If a device has an SG terminal, I connect it to 0V.  It seems to work.  But then again, I usually ground the negative of my DC power supplies.

16 hours ago, rmorella said:

(I agree with you, in that I always connect the shield only at the chassis of the PLC

I also do this, connecting to the earth ground of the system.  The "GND" terminal on your device is not clear to me, as they may or may not connect the circuitry common point (AKA their own 0V) to metal parts in the device (real and true ground).

Joe T.

[rmorella]

Thanks again for the response.  The current problem, as described by our field service tech is as follows.  On a unit with 3 slaves, only two out of the three will communicate with the master. It works in one motor mode, two motor mode, but not three motor mode.   When a 3rd slave is included, it doesn't work.   

I'm sure you are asking yourself "why the very specific focus on grounding?"  The original engineer left the RS485 signal ground off the design.  It is a very short network with all three slaves mounted in the same machine.  From what I have been told, it has worked fine this way for all installations (numbering in the many dozens).  However, recently,  we have a few installations where one or more of the slave module's communication ports are repeatedly failing.  Since the engineer is no longer there, I was called in to consult.  I immediately noticed the lack of a signal ground wire and suggested that it be added to the design, if for no other reason than a precautionary measure.  Now that the changes have been made, the field tech is experiencing the issue described above.  I do not have any experience with the Unitronics RS485 communications and it's grounding, so I am wondering if my suggested design changes have not introduced other potential problems.

Also, I am curious as to why Unitronics does not want us to use the same signal ground  pin for RS485 as the one in the RS232 pinout.  However, I will save that discussion for another day, and assume that internally they are the same as any other PLC with a non-isolated RS485 port.

 

 

[Ausman]

7 hours ago, rmorella said:

When a 3rd slave is included, it doesn't work. 

So did it work before?  Is the program allowing enough buffer time b/n requests?  What speeds are you trying to comm at?  Ziel defaults into 19200 , 8,  even.  Slow it up (edit, I just realised how confusing that turn of phrase is!  Slow to 9600) and change to none.

Changing the slave modules with failed comm ports brings them back on line OK?  Could the slaves be at a MTBF?

Is correct termination in place only at ends?

What is the cable in use for 485?

Physically check grounding and commoning points for continuity/voltages during use to determine their connections.

Going on my knowledge of Ziel bus stuff, I think you are over-thinking it.  See attached scrnshot from a manual, noting the simple wiring shown.  Do the comms as plainly as possible, then check other things.  Comms/cables may not be the real issue.

cheers,

Aus