Modbus data source setup

patd2020

Hi Woody, thank you for taking the time to reply. the modbus test software is generic not supplied with the device. i checked the comm port settings and they match. So I wonder if my serial to ethernet converters servers for some reason are not playing nice with Mango. Because they work with the test software.
Basically each modbus device is connector to a serial to ethernet converter see drawing (thetop part is description bottom part just icons of the same) and a virtual comm port is created for each device.

They also support IP so I tried MODBUS IP and get the same no response error. but if I change the ip port from the real one to a random ip port I get a timeout error. Which makes me think the connection settings are right and the request to the slave is getting through but just not getting a response from the devi0_1466476518946_ConnectionDrawingt.pdf ce.

Any thoughts?

JoelHaggar

Try checking the Override timing box and leaving the settings as 0. This leaves the timing up to the adapter and can work much better. We are changing the way this works in the next release.

JoelHaggar

You probably need a longer timeout setting if this is going over the internet. I would try up to 5000 ms.

patd2020

Thanks for your suggestion Joel but still getting the no response error

JoelHaggar

I think you'll need to run a Modbus packet capture software on the end of the virtual com port to verify what communication is or isn't coming out.

The error you are seeing in Mango just means Mango sent out a message and didn't get any response back from the device.

patd2020

any recommendations
for this type of software?

JoelHaggar

Here is a good free one: http://www.isquared.ca/English/downloads_CC.htm RS-485 spy

phildunlap

Hi Patd,

Some other thoughts as I'm also reading...

Is it possible the slave IDs are flipped / mistaken? I see COM1 trying to talk to slave 2, and COM2 trying to talk to slave 1. Sometimes operating systems can switch their identifier on you at a reboot and such.

You can try to make equivalent requests from your test software and Mango through Modbus IP to netcat, then compare (or if you're feeling serial you can cat -v /dev/ttyUSBN)...

[sudo] nc -l 900 > testSoftware
*run query from test software pointed to this machines ip, port 900*
[sudo] nc -l 900 > testMango
*run query from Mango*
hexdump -C testSoftware
hexdump -C testMango

But alas you are probably using Windows with names like "COM1" :P You can do it if you have a Linux system on hand.

patd2020

Thanks Joel & Phil. yes I see your point about the non matching slave/com port. I should have matched it to avoid confusion but that is actually how they are setup/ Slave 1 on Com2 and slave2 on COM1. Yes I am on Windows. I will attempt to run your suggested tests and report back

patd2020

here is the RS485 Modbus guide for the Flow m0_1466480777062_FlowRS485.doc

patd2020

Here is an update. I now can read the registers of both devices. the second device which had Modbsus ASCII as default I had to change to RTU. Now I need to figure out how to set up the data points

patd2020

Sorry I meant to say I change the data source to MODBUS IP because I could not get the COM to to work