Home Automation Domotica Forum Europe, Bwired Forum

I'm currently working on version 4.0 of the opentherm gateway firmware. Since I don't use all of the available features myself and I only have one boiler and thermostat, it would be helpful if other people could test the firmware with their equipment to make sure the new firmware will work with as many different types of devices as possible.

The firmware contains a number of new features requested by users on this forum, so it may well be worth your while to check it out.

The firmware is available at a special opentherm gateway alpha/beta firmware page. The new features are also listed on that page.

Please report here if you find any bugs or other problems with the firmware.

Cool! I'm not dependent on the OTGW yet, so I can give it a shot

have set TQ=21 and will watch if it gets reset by scheduled program in half an hour...

The gateway will autonomously generate opentherm messages towards the boiler if no opentherm compliant thermostat is connected.

So the problem with the Isense thermostat could be fixed, to disconnect the Isense and control the boiler only with the OTGW?
In an earlier firmware that was not possible.

Rutger wrote:In an earlier firmware that was not possible.

Yes, I generally make new versions of the firmware different from previous versions

I have just published firmware version 4.0a3. It includes the feature requested in this topic.

Nice!
Will soon test too

Schelte,

It looks like the CH water pressure (MsgID=18), the DHW temperature (MsgID=26) and the Return water temperature (MsgID=28) value in the PS=1 response are corrupt.
Output I get with version 4.0a3:

PF
00000001/00000000,6.00,00000011/00000000,100.00,0/0,15.00,0.00,1?3.:0,19.82,52.50,1?1.93,8.50,1<3.75,0/0,0/0,60.00,75.01,0,65535,64478,0,0,65535,65535,0
OK

Can you have a look at this?

I have the Celcia 20 Thermostat, I don't see negative impact the firmware doesn't send the Celcia setpoint in DataID 9 until now, maybe you can give more specific info what I should see?

Hi,

I tested the CS command and it seems to work ok. The only problem is that the CS command sent by the thermostat is also transmitted to the boiler.
So if I set the Control setpoint to 50 and the thermostat sets it to 10, both commands are sent to the boiler.
I feel that if the control setpoint is set by OTG the control setpoint by the thermostat should not be sent to the boiler.

Regards,
Bas.

Bernard,

I moved the storage of these values to a different memory area that doesn't get initialized on start-up. I guess your boiler doesn't provide the values for the DataID's you call corrupt. As a result you just get the uninitialized data, which may be outside of the range the code is able to handle (0-199). I will add initialization of the memory area in a next version. Thanks for reporting.

Thank you as well for the information that the Celcia setpoint still works. In earlier versions of the firmware, the same data value would be sent in DataID 4 and 9, but it wasn't clear if the value in DataID 9 was relevant. But using the Celcia specific value in DataID 9 would set the setpoint of other thermostats very high. To avoid such problems I prefer to send the proper setpoint value in DataID 9, which you have hereby confirmed also works.

wbvreeuwijk wrote:So if I set the Control setpoint to 50 and the thermostat sets it to 10, both commands are sent to the boiler.

That doesn't sound right. Are you sure that is what is happening? Can you show some logs?

Hi Schelte,

Upon re-examining the logs I found that it IS working correctly. I was confused by the different letters. I normally have my gateway in monitoring mode so I expect only T and B messages, but in gateway mode it is also putting in the A and R messages.

Another question that comes from this. Is it possible to cancel the CS override? The CS=T does not seem to work.

Regards,
Bas.

You can cancel the override with the command CS=0.

Opentherm gateway firmware version 4.0a5 is now available. This version adds support for attaching a DS18B20 or DS18S20 sensor to the gateway to measure the outside temperature. A measurement is done once every 5 seconds. I have only tested this new feature with the sensor connected directly to the gateway. I don't know how well it works if there's a long cable between the gateway and the sensor. Please report back if you try that.

For now the power for the sensor is supplied by an I/O pin. That way a short circuit on the cable to the sensor does not stop the rest of the gateway from functioning properly. If it turns out that a long cable doesn't work well, the power I/O pin may be needed to drive a more advanced interface with slew-rate control.

Someone requested to provide some kind of visible indication on power-up of the gateway. So it now briefly flashes all four LEDs about a second after power is applied.

The experimental code to update a remote override setpoint on the iSense has been removed because unfortunately that idea didn't pan out.

I also fixed a few bugs: The relocated data value storage area is now again being initialized, which should fix the problem Bernard reported. The clock was repeatedly being reset to midnight on thermostats that support this feature. That has been fixed as well.

This pretty much exhausts my list of outstanding feature requests. If you still miss a feature, now is the time to speak up. If I don't get any more requests, this version (possibly minus some bugs) will become the final version 4.0.

Downloaded it right away, installed it yesterday evening.. this morning when I came down the iSense said F203(not contact with the boiler) (for the first time ever). Not sure if it is related to the firmware upgrade but it sure is a funny coincidence.