Set and forget or programmed minisplits. Which is more efficient?
Hi everyone.
Just came across this website. Amazing and knowledgeable resources here.
I have a question around the set and forget vs programmed mini split heating functions. I have a two story with finished basement that has hot water radiation. I installed a two head Panasonic mini split in July and was really hoping for lower heating bills. I just got my December bill and the actual kilowatt usage was higher this December vs last year.
I have the main floor 18000 BTU programmed to come on at 6:30 am and shuts off at 9:00 am. As well, it comes on again at 2:30 pm and shuts off at 11:20 pm. The second head is in a relatively large bedroom and we only use it as needed. The remaining bedrooms are one a seperate hot water zone as well as the basement.
My question is should I leave the minsplits on all the time with a comfortable temperature or continue with the programmable features. A December bill of over $600.00 is something that I am trying to avoid in the future.
I live on the East Coast of Canada where the temperature ranges from 5 C to -10 C in the coldest months of winter.
Thanks for taking my question.
Ed
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Most important thing is to reduce the set point at night - where the COP is very low in your climate zone. If you can do that you can get substantial savings.
Increasing the deadband also makes a big difference when you are sleeping and don't care abou the exact temperature.
What do you use for heating the hot water radiation? Was there a reduction in the cost of oil/gas? Is that system off while using the heat pump?
I believe in setting and forgetting with heat pumps. They run more efficient when they are modulating/maintaining.
What Panasonic models do you have? A quick google search of 18k Panasonic units dont really show any with decent cold weather performance.
I am seeing the 18k North American units have a hspf of 8.5 and cut out at -4 which isn't that great. My Gree sapphire for comparision has a hspf almost double that at 15 and a -22 cutoff.
A quick check of heating degree days in a random Quebec province showed this past December was 5% warmer than last year, so keep that in mind as well.
To lower your energy bill you have 2 options:
1) Significantly drop your thermostat setting at all times
2) Aggressively air seal your house and come up with a plan to have nominal insulation values of R-40 walls and R-60 attic. The basement slab needs to be at least R-10 if there's nothing underneath it.
Common low hanging fruit is air tightness, and unfinished attics could be air sealed and increased to R60 in a weekend, unfinished basement walls could have foam glued to them in a day.
>"I have the main floor 18000 BTU programmed to come on at 6:30 am and shuts off at 9:00 am."
At 6:30 AM it's usually close to the coldest daily outdoor temperature, the temperature at which the mini-split is operating at it's lowest efficiency.
That is being compounded by starting the temperature recovery ramp at that hour it's also forcing the mini-split to run at it's maximum speed, which is also it's least-efficient speed.
So by running it at the worst speed at the worst outdoor temperature it's chewing through any "savings" from the lower house temperature pretty quickly.
For most (but not all) cases it will use less electricity with a "set and forget" approach.
But not all models are the same- a model number and a better location information (for weather data purposes) would be useful in this discussion, but your reported winter temps are fairly temperate, and even mini-splits not designed for cold temperatures should still do better with set & forget than the setback approach.
Also, what size head is in the bedroom? Most bedrooms don't have a heat load high enough to warrant it's own head, and if JUST the bedroom zone is calling for heat it could be short-cycling the compressor of a 2-zone system into low efficiency too.
With both zones operating in a "set and forget" mode the system as a whole will run more efficiently even if the heads are oversized for the zones, allowing the compressor to operate over a wider modulation range with fewer short-cycles, and spending far less time running full-on at lousy efficiency.
A mini-split with the lowest specified output in the table at -4F/-20C isn't going to be a problem in your climate, and will still be putting out useful heat at temperatures below -20C, even though that heat rate isn't specified.
Thanks Dana et al.
The panasonic model number is Cu-3E19rBU and the head in the master bedroom is 12K BTU which is covering off a 16' x 26' room. The HSPF rating is 10.5. We have the heat on 17C in the bedroom during the night as we find it better for sleeping.
I live in St. John's, Newfoundland and the weather fluctuates as we are near the ocean. The coldest temperature here in the winter is usually -10C to -13C and that only lasts for a few weeks.
Our hot water radiation is heated by an electric furnace. We have R-60 in the attic as I just topped it up last fall with blown in insulation. Our basement is half finished and the section not completed has R-20 on the walls.
Can you clarify the JUST the bedroom comment? We are only using the mini split in our bedroom for heating as we assume it has little impact on the additional three bedrooms, kids bath and laundry room. Those rooms are on the hot water radiation zone. Should we keep our bedroom door open or closed for the size of the unit in there? If it remained open will it be too much of a draw on that unit.
Yes, St. John's has very temperate maritime winters:
https://weatherspark.com/m/147565/2/Average-Weather-in-February-at-St.-John&%2339;s-International-Airport-Canada#Sections-Temperature
The Cu-3E19rBU has pretty reasonable capacity and efficiency all the way down to 5F/-15C, but you'll notice that the efficiency at it's minimum modulation levels are considerably higher that at higher speeds, at any listed temperature:
https://neep-ashp-prod.herokuapp.com/#!/product/26410
The minimum modulated output of it's compressor at +47F/+8C is a pretty hefty 5500 BTU/hr, so unless multiple heads are operating at once it's going to be doing more cycling than modulating, which isn't terrible if cycling on/off at it's slowest, most efficient speed, but it's still an efficiency hit.
When it's +17F/-8C outdoors and recovering from a deep setback, the 18,000 BTU/hr head in the main zone running on it's own would be running the compressor at above half it's range, probably hitting the full maximum output delivering a COP efficiency of about 2.14 . If it's allowed to modulate and just maintain temperatures rather than setting back it looks like the Cu-3E19rBU would be delivering a COP of between 2.5 and 3.35 . That's potentially more than 50% more heat delivered per kwh. There is no WAY that you're saving enough energy in the setbacks to make up for that large an efficiency difference between running full-out during the recovery ramp vs. just maintaining temp at some low to intermediate speed.
>"Can you clarify the JUST the bedroom comment? We are only using the mini split in our bedroom for heating as we assume it has little impact on the additional three bedrooms, kids bath and laundry room."
Panasonic has more than one 12,000 BTU/hr head and they're not all the same. But there is no way a bedroom needs a 1-ton head. Most half-ton heads are sub-optimally oversized for bedroom loads. When coupled to a dedicated single zone compressor the one-ton CS-XE12SKUA can modulate down to 3000 BTU/hr @ +47F (still a lot for a bedroom), but the compressor serving it can't:
https://neep-ashp-prod.herokuapp.com/#!/product/26406
The one-ton CS-E12RKUAW can only modulate down to 4100 BTU/hr @ +47F which is even more oversized for the average load, but still can't go under 5500 BTU/hr when it's the only zone operating on the Cu-3E19rBU , so the limiting modulation factor is still going to be the compressor when "...JUST the bedroom..." zone is active.
If BOTH the bedroom zone AND the 1.5 ton head in the main area are running at the same time, under most conditiones the combined output of the heads would be above the minimum output of the compressor, allowing the 12K bedroom head (and 18K main room head) to modulate down to IT's minimum output rather than be forcing it to cycle on/off at the COMPRESSOR's minimum output. That will ease the efficiency hit due to cycling by quite a bit.
At +17F/-8C ...
the minimum modulated output of the CS-E18RKUAW head is 3400 BTU/hr...
https://neep-ashp-prod.herokuapp.com/#!/product/26404
... the minimum output of the CS-XE12SKUA is 2100 BTU/hr...
https://neep-ashp-prod.herokuapp.com/#!/product/26411
...and the minimum output of the CU-3E19RBU is 4100 BTU/hr.
So when it's -8C outside, as long as there is at least 4100 BTU/hr of load on the zone served by the 18K head (there probably is that much) the oversized bedroom head can cycle on/off if need be without dramatically affecting the overall system efficiency, since the compressor would only need to shift from 5500 BTU/hr (both heads running) and 4100 BTU/hr or whatever the load is on the 18K zone, with just the 18K head running. By not having to spin up the compressor from zero and just stepping up/down in smaller increments as the bedroom head cycles it will stay in a comparatively lower modulation level (= higher-efficiency mode.)
At -8C at any compressor speed the mini-split is at least twice as efficient as heating with an electric boiler or electric furnace, so even if running the ductless continuously maintaining a higher indoor temperature, the heat coming through partition walls/floors/ceilings from the ductless zones offsets some of that power use from lower efficiency system.
One thing I have experimented with is a slooow ramp up from the set back: 1°F every 1/2 or full hour. Trying to keep the speed low.
Too soon to know if it's any more efficient than set and forget.
I also keep the downstairs units at normal temp, and just reduce the bedroom temp. Then there's less of a ramp in the morning.
The best answer will come from measurement, not guessing.