ABB ACH550 RS485 trunk issue

[EDGAR]

Hello everyone! I have a little issue with a single drive in a 485 network. Basically I have a nano working as a modbus master polling information from each drive. My issue is that one of the drive gives me a UART error even with the nano communication turn off. Yesterday I remove that drive from the network and locally it work using an u-485 converter. But as soon as I connect one of the network wires it starts counting on UART error.

My question is, have anyone dealt with an anomaly like this? Also, what's on the U485 that's not in the nano 485.

One note, the drive in question is the 3rd node on the network at of 4 drives and the last node works fine.

Thanks, any help will be greatly appreciated...
Ed

[support]

U-485 has some biasing on the RS485 port. Nano-10 does not supply the biasing voltage. This might makes a difference to the drive?

You can try to add in a 1K resistor to RS485+ to pull up to +5V and another 1K resistor connect to RS485(-) terminal to pull-down to 0V and see if it would improve communication.

[EDGAR]

Thanks for your reply!

Do you want me to add the 1k resistor at the drive terminal or at the nano terminal?

Thanks,
Ed

[garysdickinson]

Ed,

I have not used the ABB drive but I have worked with Hitachi and TECO.

The RS485 interface for both the Hitachi and TECO VFDs is electrically isolated (floating). On these drives you must add a 3rd wire to tie the VFD signal ground to the 0Volt input to the Nano-10 PLC.

The documentation for the ABB Ach550 rs485 indicates the the RS485 interface is electrically isolated. If you do not add this 3rd wire between signal ground on the ABB unit and the this Nano-10 then Rs485 communications will not work reliably.

The ABB Ach550 signal ground is named "AGND" and it is on the X1 connector, terminal #31.

If you have already done this, then kindly ignore my meddling.

Gary D*ickinson

[support]

Thanks for your reply!

Do you want me to add the 1k resistor at the drive terminal or at the nano terminal?

Thanks,
Ed

You can add it near the drive terminal since Nano-10 does not require biasing.

Gary's point is also valid and could explain the problem you encountered. All the RS485 devices must share a common ground. So if you have mulitiple drive connected together then the 3rd wire should be used to tire up all their common ground and the 0V on the Nano-10.

[EDGAR]

Thank You Both for the input! The network is already setup as you guys mentioned, the shielded wire on each drive is connected to the AGND terminal and at the Nano is connected to the 0V. All but Drive2 is not communicating at all, with UART Errors.

My next step is to enable to built in bias terminator in the last drive (Drive4) to see if that works, if it doesn't work I will disable it and then add the 1K by the Nano and try again as you mentioned.

[---------------RS 485 Network--25' max-------------]
 ====\/=======\/======\/======\/
      [Drive1]     [Drive2]     [Drive3]     [Drive4]

I have also reached out to ABB to see if there's anything else I can do, so far still waiting....

My very last attempt to fix this id everything else doesn't work, I will try to use the Auto485 and connect it to the drive via RS232 back on the network with via 485 , I haven't try this yet... Keep you all posted.


Thanks,
Ed

[garysdickinson]

Ed,

Since you have 3 out of 4 drives working, I would focus on the drive that is not behaving.

I don't think that termination resistors will be useful as your total cable length is very short and the BAUD rate is very low.  

The bias resistors are probably not the problem, either. If this was an issue you would be experiencing communication error on the other drives.

These are my guesses as to what is wrong with the problem drive:
 

• The A and B signal wires are switched on the failing drive.
• The communication parameters are set incorrectly on this drive
• One of the signal or ground wires slipped out of the terminal block on the failing drive.
• The RS485 transceiver is damaged on the failing drive

I would connect your laptop at the PLC and monitor/test the communications. Use all the tools that you have.

Gary D*ickinson

[EDGAR]

Thank You!
I'm leaning more to bullet #4, but the drive worked when connected to the laptop with the U-485. Comm settings are all the same for all drives.

ABB just sent me some troubleshooting documents but nothing that I haven't tried. I was thinking on using the Auto485 but I can't because the com port on the drive is used by the keypad.

Thanks!!
Ed

[garysdickinson]

Ed,

I am leaning towards bullet #1 (cable wiring problem).  

Bullet #4 will give you the same symptoms as bullet #1.  It is possible to destroy the RS485 transceiver in the drive, but you have to work at it.

If the A/B signal wires have been swapped then the the serial data will be inverted.  This will confuse the UART in the ABB drive. The probability of the drive receiving a valid Modbus command is 0.  Modbus slaves do not respond to garbled messages. The PLC will time out waiting for a response from the drive.

If the drive will communicate when wired directly to your PC but goes deaf when wired in with the other 3 drives makes me think "cabling problem".

This stuff is a bear to debug.  

Gary d

[EDGAR]

The wiring is correct there's not much to it, is working on the other drives on the daisy chain. ABB suggest to check the impendence across the + and - terminal. Measurement needs to be 164ohms with +/- 5%. They also asked me to remove the shielded from the ground, and turn on the bus terminator on the drive.

I will be at the plant on Monday working on this, I will keep everyone posted.

Thanks Gary!!
Ed

[EDGAR]

I forgot to mention that I get the error even when I swap the wires, when connecting just +ve wire, when connecting just the -ve wire. Even when there's no communication being broadcast I still get the error and the drop is not responsive.

I will follow the ABB suggestions and let everyone know the outcome.

Ed

[EDGAR]

Hi, just to update everyone. After testing the impendence on the drive it was well above 164Ohms (34K) the transceiver was deemed blown by ABB. Next step is to change the entire board, or just use the built-in AO to read the desired analog signal. However, just because it works with the laptop, I'm researching on using an RS485 extender ie. RS485 to RS485 just to interface with the VFD to see if it works.  

Any ideas will be greatly appreciated...
Ed