Increasing heating period while output decreasing?
This is an awesome integration, and after some days I think parameters are almost "perfect" in my case, using PID and external compensation.
Anyway, on my daily control, looking for final check of parameters, I've found something I could not explain for the moment. That's why I put the case below.
On the below image, at 22:38, heating mode is ON until 22:49. Heating period has increased while output has decreased. Is this a correct way of working?
Am I missing anything, or perhaps could be a malfunction (I don't think so, but I'm wondering why)?
Additional info:
pwm = 10 minutes
set point = 21
kp = 700
ki = 0,033
kd = 1500
ke = 2
Thanks for taking a look and explain if necessary.
This may be due to PWM stretching, in case you set minimum on or off duration in settings, the thermostat may change PWM duration to respect it.
Indeed yes. I have a minimum of 1 minute. Your comment clarifies. Thanks. I need to observe a bit more to have a fully understanding...
Back again, with new observations: I am very often observing behavior that surprises me. I would like to know whether this is normal, as I have the impression that it introduces an undesirable effect into the PID control. I apologize because the texts are in Spanish, but I think it is easily understood.
It has been reported already, but I have no time to investigate at the moment.
OK. I didn't know. We are open to whatever we can do to help...
I've continued to observe the operation of the PID, and I've seen strange things that I've related to the Tuya integration (the switch is Tuya). Consequently, I have changed the Tuya switch to another Shelly. From that moment on, I have not observed strange behavior in the PID again. For now, I dare to say that the strange behavior observed have been due to errors in the Tuya integration, which does not seem to work correctly 100% of the time. In any case, I continue to observe the behavior, and if I detect any anomaly, I will share it again.
I already had some issues with Tuya cloud integration. The link is not reliable and sometimes it misses some commands, so with this thermostat that may need to control the heater very frequently, it can become critical if the commands are missed or delayed.
I personally switched to Local Tuya integration to fix these issues.
Mmmm... I have also Local Tuya, that was the one I used first, but I had so many problems with it that I decided to change to Tuya. Reload of integrations was needed frequently. I hope Shelly integration will work much better. At least, for the moment, is this way.
Did you assign fixed IP addresses to your devices in your router's DHCP settings? It's important for Local Tuya as it needs to know where to find the devices on the network. I'm using this fork of the integration which receives more updates, and I have no issue.
Actually, no. Anyway, my router assigns IPs that it usually respects. But anyway, my problems came from having to reload integrations, both Local Tuya and Tuya, too often. I have read this in numerous complaints from other users. So far, the switch to Shelly has been a wise move, but I take into account your advice on integration that you suggest (although it is what I have used without much success). I'm already in the fine-tuning phase of the PID, with a goal of achieving a maximum oscillation from the set point of +/- 0.02 ºC (and I'm pretty close).