Modbus Master on RS485 with Digital Outputs

You mean like one of these?

https://consteel-electronics.com/modbus-modules-RS485

https://www.csimn.com/CSI_pages/Modbus101.html

I've used several devices like those before, but unless I'm missing something, they are all Modbus SLAVES. It's misleading that they state they "support the MODBUS MASTER protocol" when those devices are slaves. I need a device that is the Master, and will poll the slaves.

I though you said you had a controller with RS485 channel available...? It seems to me you should be able to control everything through the MOD-4DO...

No, our existing controller cannot be modified, but can be a Modbus Slave. Which is why I specifically said (a few times now) "I’m looking for a Modbus RS485 Master".

I don't know how else to say what I've already said.

My device = Modbus Slave.

Device I require = Modbus Master with digital outputs.

I could work with a solution that had a Modbus Master that read from my slave and also drove another slave (e.g. the MOD-4DO), however, I need that master.

MOD-4DO is a Modbus Slave, not a Master.

Would something like this work...?

https://www.pcnation.com/web/details/TE5923/Brainboxes-Ethernet-to-4-Relays-and-8-Digital-Inputs-plus-RS485-Gateway-ED-538-837324009828?utm_source=google&utm_medium=cpc&adpos=&scid=scplpTE5923&sc_intid=TE5923&mkwid=s_dc&pcrid=418067831246&pkw=&pmt=&plc=&gclid=CjwKCAiAy9jyBRA6EiwAeclQhPCmA4UwtobcaBbHtq3kduRr-ai78ugDvoFfvSKvMeUxdfwkF4fg6RoCmfYQAvD_BwE

No. The statement, "The device can be controlled over TCP/IP using the Modbus TCP protocol" means that the module is a Modbus slave that can be controlled, by an Modbus master that does the controlling.

So you need one of these in front of it...

https://www.datexel.com/modbus-rtu-master-digital-io-dat9000io.html

Yes

I do not think the Mod-4DO module is a Modbus master. The devil is in the details.

I would expect the I/O modules to be a Modbus RTU slave with the DO's where the module has a slave ID address and each DO has a Modbus Coil address. The PLC would be the Modbus master and write the coil logic state using either FC05 (Write single coil) or FC15 (hex) (Write multiple coils). But the PLC can't do that, it's not a Modbus Master.

When the I/O module is a Modbus Master then the module has to read one or more coil registers in the PLC to determine what to do with its DO's. The PLC would have to be configured with slave registers to reflect what the I/O module's outputs should be.

And then when Master reads the PLC's coil registers, it has to change the master's DO registers to reflect the read values. That could be done, but I doubt it, because the Mod-4DO module has settings that are typical of a Modbus slave.

The on-line manual is only in Polish. Although I don't read Polish, some of the settings are self evident either through cognates of English words or their location in a table of values that are recognizable as Modbus settings.

Mod-4DO manual in Polish

1. Only Modbus slaves have a slave address, a Modbus RTU master does not have an address. Table 3.4.2 on page 7 has register 40002 that defines a slave node address value. Presumably this is the module's address, and not the target device's slave address.

2. I suspect that table 7.2 on page 9 is the table for the four coil outputs. Which would make sense for a Modbus slave.

3. The home page statement, "Our I/O modules are devices that communicate with the most popular PLC devices and devices that support the MODBUS MASTER protocol." most probably means that their Modbus Slave modules talk to Modbus Master modules, which is true, that's how Modbus works. I do not think that the statement means that the modules are Modbus masters.

4. Only Modbus slaves have register map/tables. Modbus Masters never do. The presence of register tables 3.4.2, 7.1, 7.2 define this device as a Modbus slave.

I think you need a PLC with Modbus master functionality that can do what I described above, or something like Red Lion's Data Station Plus with a DO module (I believe Red Lion has that as an automation controller under a name other than data Station.)

What device are you currently using? If it already has one Modbus slave, can you not just get another slave to add? Most masters will cope with up to 127 slaves, although there may be other limitations depending on data amounts. I don't think that there would be an issue with only four outputs.

I think Iris is on the right track. I would differ only in suggesting the G07C0000 with the GMDIOR00 I/O. The G07C0000 has a multitude of serial protocols, Modbus Master being one of them. Be aware it is expensive for 4 DO and watch out for a delay in updating the I/O. I would caution that if there is anything critical about the sequence of operation for the outputs you need to consider that you can't count on a predictable I/O update and might encounter some unexpected and unwanted operation in the equipment being controlled which would appear random in nature. You do have the advantage of having a display as well. The programming software is offered for free by Red Lion.

I noticed you said a PLC has such a large footprint gives me the impression you are using a rather large footprint PLC currently. You might consider a PLC such as the Panasonic FPX-C14; it's footprint is about 3.5Hx3Wx3D and has 8-in and 6-outputs, relay or transistor, and it is inexpensive. You can parallel the inputs you used on the existing platform that would apply to the 4-outputs you need and then do all of the logic in the FPX. The PLC, last time I check, is about $120.00. The inputs are DC so that has to be factored in with the current inputs you are using. The biggest expense would be the software but it is about $400.00. The combined package is much less than the Red Lion approach.