View Single Post
  #10   Report Post  
Posted to alt.home.repair,sci.engr.heat-vent-ac,misc.consumers.frugal-living
Rod Speed
 
Posts: n/a
Default prog. therm. and heat pump questions

Logan Shaw wrote:
Stretch wrote:
If you check the data, or measure it yourself, Strips cost 2 to 3.5
times as much per BTU as the heat pump.


That's logical, since the whole purpose of a heat pump is to be more
efficient than resistive heating.

So you set back by shutting
off the cheap heat (Heat Pump Compressor), and recover by using the
expensive strip heat. Setback increases the electric bill if the
heat strips come on during recovery. In addition to warming the air
during recovery, you also have to warm the entire thermal mass of
the house and all it's contents. THAT is ia big part of the added
power consumption.


Also logical, but aren't there programmable thermostats that are
smart enough to deal with this? I have a cheapo programmable
thermostat which is supposed to be able to intelligently determine
how much time it takes to recover. A well-designed programmable
thermostat for a heat pump would be very conservative about this
and would err on the side of recovering way too early (hours early)
rather than having to kick in the resistive heating.

I don't know that such thermostats exist, but it seems like it
would not be all that hard to make one that is smart enough to
avoid having to turn on the resistive heating due to setback.
Machine learning techniques are pretty good these days, and the
thermostat has good information available to it (the average
rate the temperature drops when the system isn't running gives
it an idea of the load, and the duty cycle gives it an idea of
its capacity relative to the load). The worst problem seems
like it would be changing weather (where it's 30F colder one
night than the previous), but even that can be solved if the
system realizes the temperature is dropping a lot faster over
the course of the night than it was the previous night and
knows not to expect last night's data to be a good indicator.


Or just use the inside and outside temps and the measured
performance of the system at that outside temp to know when
it should come off the setback to have the inside temp back to
normal temp at the time specified. Not a shred of rocket science
required at all.