How are those using GivTCP handling inverter timeouts? I’ve had a few instances where I’ve gone to set my battery reserve to 95% for example but because the inverter has timed out (dropping connection issue), it’s not being set on the inverter as 95%. I get 6-7 minute connection drop gaps fairly frequently.
I have an automation checking the time stamp; if it's more than 5 min. The GivTCP restarts. I used to have a lot of these with older GivTCP.
deanjesper This also happens via the Cloud API actually, we regularly get long multi-second delays between the request and response and then Inverter Timeout messages.
I've shared my automation for detecting GivTCP stopping working in the Predbat documentation https://github.com/springfall2008/batpred/blob/main/docs/output-data.md
It looks for:
And raises an alert. Doesn't do an auto restart as I wanted the opportunity to look at the logs, but easy to add that
The first failure is by far the most common, caused by inverter not being able to connect, but I have certainly seen the last one once when the inverter worked but lost battery connection
You can also look for things with improbable data, such as Inverter Temp 0'C (unlikely) which I believe the direct HA integration 'Givenergy_local' checks for.
Thanks all, I’ll look into setting up an automation based on what you’re doing.
Does anyone know how “Queue Retries” works under the GivTCP config?
geoffreycoan I've added the low inverter temp check to my automation script, and details of how to restart GivTCP as well automatically.
It'll be pushed through in the next PR to the predbat docs, but my fork is https://github.com/gcoan/batpred/blob/main/docs/output-data.md
Ok here’s a real life issue I’m having:
I select Force Export for 45 mins in GivTCP
The inverter goes offline at minute 43 and stays offline for 7 minutes so when GivTCP tries to revert to Normal on minute 45, it can’t and gets stuck
My battery is now stuck exporting
What’s the remedy? Restarting GivTCP won’t do anything (I don’t think?)
deanjesper The inverter goes offline at minute 43 and stays offline for 7 minutes so when GivTCP tries to revert to Normal on minute 45, it can’t and gets stuck
when you say 'inverter goes offline' where are you seeing that?
Is it no longer on your home network, can't ping it and Home Assistant looses connection?
If that's the case then it sounds like its a network issue, is it ethernet connected or Wifi? I ended up putting in a mesh wifi in my house as the powerline adapters I had weren't reliable enough. I've now got a mesh node in the garage and another in the living room, both of which the inverter can "see"
geoffreycoan I have that annoying bug where the inverter goes offline for a few minutes even though it’s still connected to the local network. There’s a patch on my AIO that reboots the “dongle module” if it goes offline but it does mean I get gaps. I’m not sure if others have the same issue but it was patched by Paul at GivEnergy.
deanjesper Oh that's a pain because it means as you say that if GivTCP sends any commands when its offline they get lost.
I suppose you could automate something along the lines of: