Hi.
Below link to complete tutorial how to connect the S11 extension using node-red
https://youtu.be/HJJRrV9tV7U

Dobrý den,

váš předchozí příspěvek prověřím a případně doplním.

Na aktuální otázku je odpověď: ANO, lze funce I/O nastavit i jinak
Pro nastavení u rozšiřujících modulů Unipi můžete použít např. program mbpoll v linuxu (po připojení přes SSH) a komunikovat se zařízením přes modbus, kde si jednotlivé funkce nastavíte dle modbus tabulky.

Přeji hezký den
Antonín

Hello,

We always (even for Unipi 1.1) recommend using a secondary external relay and using the relay on the unit only as a signal relay.

All available technical parameters for relays on Unipi 1.1 can be found in this article, for Unipi Neuron in this article and for Unipi Extensions in this article.

It is better to deal with the details of the electrical installation with an installation or inspection technician according to the legislation of your country.

Best regards,
Antonin

@martin_triska Thank you for this answer :)

As you can see the addresses of both MCP23008 are visible.

0x20 for the MCP23008 on the Unipi 1.1 board
0x22 for the MCP23008 of the EMO-R8 expansion board

And both boards have their own power supplies purchased from Unipi (CLW-1005-W2E-EB) of 5V in 2A.

And with or without, it doesn't matter...

So I made a new RJ11 cable and it worked :)

Life is simple especially with a good connection....

Hello @sknobloch,
accessing the Unipi Extensions over ModbusTCP is not possible at the moment, although we are thinking about it.

You need to access the devices via ModbusRTU over the corresponding serial device. Here you can find the mapping of the physical RS485 ports: https://kb.unipi.technology/en:hw:02-neuron:portmap

Best regards,
Martin

Hi @CKB,
the Unipi Extensions are generic ModbusRTU slaves. Since they communicate over RS485 and Unipi 1.1 doesn't have any, you need to use USB/RS485 converter.

Best regards,
Martin

@martin-kudláček said in Neuron slow communication with extension modules:

mid-length

Hello Martin,
Thanks for the answer. I've already thought of something like that.
According to the DIP switch, the xS11 can only have a maximum baudrate of 19200. An increase has led to a loss of communication.

I thought about swapping my M303 controller for an L303 to have enough inputs on the controller to bypass the Modbus. I then only use the extension modules to use outputs and inputs where speed is not an issue.
Is it correct that the processing of the internal inputs on a controller (e.g. L303) is a faster method than Modbus?
Is my approach understandable and sensible, or do I have a mistake in thinking?

With kind regards
Walter

@tomas_hora all good, it just needed a good apt-upgrade :)

Hi @Martin-Kudláček THX

Regards

For the people who encounter the same problem or just stumble on this post by coincidence:

UniPi tech wrote: “We know the firmware update, especially of the extensions, is a sort of Achilles heel and we are thinking about some concept of dual firmware to which you can revert to if the flashing of the new firmware fails.”

I also would like to mention the excellent support from UniPi! I’ve received a very fast and meaningful response from both the technical as commercial team, and they are moving heaven and earth to provide me the fastest possible solution.

I didn’t experience such a personal approach from an international company since a long time.

Hello @marko2276,
the Direct switch is indeed supported on xS10 and xS40 as well.

Best regards,
Martin

Hi @alydersen,
it is hard to tell what's wrong without the access to the setup. Would it be possible to arrange SSH access to the PLC? We can go over the details via email: [email protected]

Best regards,
Martin

Apologies, I had taken time off for personal reasons for the last two days. I will be here over the next two weeks every day.

hi tomas,

2.05a seems to have fixed this error too - the extensions are fully populated and the send events now (the old version didnt send events too as i had to discover).

thx for helping,
wbr,tja...

hi tomas,

the new version (2.05a) seems to be fixed regarding that strange message.

thx for helping,
wbr,tja...

I assume that this problem has now been solved, based on your responses in other topics? It's hard to see the connections when they overlaid on top of each other like that.

Thank you Tomas, it was very helpful. Now it working. I only switch off parity on external modules and communication started :-)

@sidoor It might work but it always depends on in what environment the installation is done. You can just cut the RS45 connectors and do the wiring in the standardized clamps

@mauri Correct

@christophebler Ok, than the serial line works well. But I still recommend using our opensource image because the parameters that can be set using the github version is limited and might cause some troubles which are fixed in the opensource image.

Byt the way, what do you want to do by 'Set Adress to 0 and rebooted both devices'? You can set address only at the slave which is the xS50 and if you set it to 0, than the SW configuration is used (address 15, 19200bps, no parity) and you are opening different device...

Address 0 is not allowed on Modbus, thus the behavior like I described.