Home Automation Domotica Forum Europe, Bwired Forum

It seems like something is frying your PIC. Either a bad power supply, or some external factor inducing damaging currents into your gateway. Sadly I have no concrete advice to offer on how to remedy that.

You could try to relocate the gateway, move it closer to the thermostat (the boiler interface is optically isolated, so the only way for electric interference to get into the gateway is via the thermostat interface, the power supply, or the air). But if your PIC is fried, that will have to be replaced again.

My cellphone is equiped with a magnetometer, so I held it right next to the boiler at the location of the otgw. After turning on the hot water tap it showed me a highly intense sine wave. So I probably found out what's frying my PICs.

Next I'm going to mount the otgw at some other location; further away from the boiler and without the electrical inteference.

This is pretty much an eye opener to me. Wouldn't have thought about this before (eletronics isn't really my cup of tea...).

Also the cable between the thermostate and otgw is pretty old and unshielded. Would it be useful to replace it by a shielded cable? And what about adding ferrite beads to both ends of the cables?

As it is currently still unclear what exactly is frying your PIC, such measures may help, as long as they don't affect the actual opentherm signal too much.

You could also add a zener diode in parallel with R6 to prevent high voltages getting to the PIC from the thermostat interface. Such a zener diode was actually present in my original design. It was removed in later versions as it didn't seem to be necessary. You are the first person to run into such problems that I heard about in over ten years. Don't you feel special now?

Hey, I've got a similar problem, so I'm just going to follow this thread.

It's completely random when my room setpoint goes to 35degrees. It happened today, but the last time was months ago.

I'm implementing a Homeassistant automation that puts it back to 16 when it rose above 23 degrees as a safeguard. But thats not a really good solution...

It doesn't sound exactly the same. Can you share a log (5 minutes before and 5 minutes after) of an occurrence when the setpoint unexpectedly changed?

After moving the gateway further away from the boiler, I experienced far less problems, but it still froze once in week or so.
Then I replaced the connection between the boiler and gateway by shielded cables and put some ferrite beads on it. And now its running stable for more than a month.

Pretty sure that electrical interference caused the problems. Funny thing is that I didn't replace the PIC for a second time. So, probably my first PIC was't borked at all.

That's good News. Can you provide some specs of the ferrite beads you used, in case someone else wants to try the same solution?

They come in different sizes, so be sure they fit well.

Example:
https://www.reichelt.com/de/en/split-fe ... 25738.html

hvxl wrote:It doesn't sound exactly the same. Can you share a log (5 minutes before and 5 minutes after) of an occurrence when the setpoint unexpectedly changed?

I didn't have a log back then. And it has not occurred ever since my last message. Could be the electronic interference might be the cause as mentioned here. The OTGW was first positioned on top of the solar collector boiler. Perhaps thats giving some interference. I'll keep it away from the collector.