[bug] - min_cycle_duration not respected
swingerman opened this issue · 9 comments
Using v0.9.8.beta-5, noticed some undesirable behavior that may be related to this fix.
The program does not seem to respect the min_cycle_duration when switching between "cool" and "fan" states of the fan_hot_tolerance function. While the original issue of gaps between these two states is fixed, now it will cycle back-and-forth between "cool" and "fan" as soon as any temperature change is detected, resulting in the short-cycling that min-cycle-duration is meant to avoid.
This may have been an issue since the v0.9.7.beta-4 release, but I cannot confirm.
Originally posted by @Jeepmb in #218 (comment)
Could you please post your current configuration?
So when the thermostat wants to change from cooling to fan, but the cooler still needs to be cycled out. What do you think we should do?
- don't touch the cooler, and turn the fan on or
- don't change anything until the cooler cycles out?
I think the appropriate action would be the second option. Don't change anything until the cooler cycles out.
Also, I changed my configuration a while back to work around this issue, and can't remember the exact config, so I don't want to post anything that may be inaccurate. But I can try and re-create it for testing if required.
The fix is ready. I will include it in the next release.
0.9.9 still reproduces the same behavior, of by accident someone can get the temperature to a higher setpoint and the heater turns on, and you inmediately change the temperature setpoint back, the min_cycle is not respected and the off signal is sent right away
This is intended behaviour. The built-in thermostat in HA works the same way. Setting target temperatures is considered intentional, and thus, they override every current behaviour.
ok, I thought this would override it. What one wants with the min cycle duration is to respect the heating/cooling device cycle, my AC needs to run a minimum of 1 minute too, to have the oil lubricate the system.
Shouldnt this behavior be changed?
For what it's worth, I am in favor of the current behavior of responding instantly to manual temperature set point changes.
While I expect the normal operational cycling to respect the min cycle duration to reduce wear like @mat1990dj describes, I think it's important for the system to respond to my inputs instantaneously when needed. I believe the intent of those minimum run time requirements for HVAC devices is to prevent damage from very frequent short-cycling, not occational set point changes.