Avanta 28C/OpentermGW 4.0a3/Celcia 20 - boiler comms

[dirkx@webweaving.org]

Folks,

On a freshly build openterm gw; USB comms are fine; voltages sane; etc.
- A Celcia 20 directly connected to a Remeha Avanta 28C boiler works fine.
- Same Celcia 20 connected to a 4.0a3 firmware OGW reports T00.. reports at regular intervals (provided IT=1 is set).

However - its display gives an Error 203.
- No B reports are seen.
- Both GW=0 and GW=1 tested.
- When just the boilder is connected - will see 'R' messages appearing after a while.
- IT=1 needs to be set in onder to clear/avoid Error 3 from the OGW.

Any suggestions as how to trouble shoot this. I have the usual measuring equipment; including an oscilloscope.
Thanks,
Dw.

[dirkx@webweaving.org]

Upgraded to

1. "OpenTherm Gateway 4.0b1" on the gateway
2. 4.0b2 on the client (tried windows under wine and the Rasp-Pi version -- identical results).

That has made all Error 01/03 go away. Although interestingly enough it does not recognize the Celcia (PR=R reports 'D').

Issue still the same - 'T00..' reports seen from the Thermostat - but no boiler comms.

Dw.

[hvxl]

There appears to be a problem with the gateway interface towards the boiler. The R messages are what the gateway sends when there is no thermostat connected. You never seem to get anything back from the boiler.

Have you gone through the troubleshooting page on the web site? Especially the tests in the "Opentherm interfaces" part might shed some light on this issue. Test #4 and #5 of the diagnostic firmware may also produce useful data points. Run test #5 in three configurations: on a disconnected gateway, with X1 looped to X2, and with the thermostat and boiler connected.

Someone else had a similar problem recently. Unfortunately he stopped responding several weeks ago. I don't know if the problem is solved, he lost interest, or just hasn't had time to look into it. In any case you find some more suggestions on what to try in that topic.

[hvxl]

Although interestingly enough it does not recognize the Celcia (PR=R reports 'D').

Normal communications are required for the gateway to be able to properly identify the thermostat type. But the thermostat type is not yet relevant at this point.

[dirkx@webweaving.org]

Ok - Diagnostic 4 (loopback) and 5 (voltage) show no obvious issues; and suggest normal working/tolerance of the electronics.

However occasionally the voltages go haywire. So checking for ground loop and related fun and game next.

Dw

[dirkx@webweaving.org]

Thanks for the help.

Oscilloscope revealed a tremendous amount of 50hz noise and a slowly undulating 15 volt bias on the Boiler wire X3.

Replacing a cap with worn-rubber && cleaning the dustly/water-leakage remains from the main board in the Avanta28C fixed this.

We now have functional comms; and the 4.01b firmware nicely goes in PR=R -> C mode.

Couple of odd values in the log - and somehow no return water temperature ever seen - but functional sofar.

So on to finding a way to remove the thermostat completely - and having the zwave GW control the boiler directly (though am wondering if it is the Celcia that does the modulation - and perhaps wether it is better to 'fool' that device than to control directly).

Thanks,

Dw.

[hvxl]

We now have functional comms; and the 4.01b firmware nicely goes in PR=R -> C mode.

Couple of odd values in the log - and somehow no return water temperature ever seen - but functional sofar.

Great that you figured out what the problem was and thanks for reporting back that the issue is now solved.

The Celcia doesn't request the return water temperature. If you are interested in that, instruct the gateway to inquire about it: AA=28.