Verifying Heat Pump’s Energy Efficiency Performance
Dave46
| Posted in Mechanicals on
High energy usage – how to verify efficiency of a heat pump – Mitsubishi ducted central system issues and have a ductless multisplit
I am interested in knowing if it is possible with thermometers & electricity monitoring (have full house but maybe need by circuit?) to verify the performance of VRF heat pump systems as a homeowner and what is the best way to do so? Not going for perfect but would like to know that performance is close to advertised.
My big issue is Mitsubishi reps have said my system is fine but performance seems strange and I want to check the centrally ducted Mitsubishi Smart Multi hyper heat system (MXZ-SM48NAMHZ) with a ducted air handler (PVFY-P54NAMU). I have tried setting the blower fan to high and moving the setpoint way up. The Mitsubishi system then goes to 8kW at any temperature under 50 even though the condenser submittal for a ducted system has a maximum power of 4.8kW at 47f. Mitsubishi reps said this is ok and I still don’t understand why or how it could perform so much worse than the submittal.
Outside temperature yesterday was 40f and I tried the fan on low with a high setpoint and it went to 7.5kW for 15min then held 5.5kW average for next hour before I stopped it. Also power goes up and down 5+ kW on 45-60 min intervals when less than about 30f (maybe defrost but strangely regular and very different from the daikin system).
Installers have come out a couple times and also dealer reps and Mitsubishi reps and I have only ever been told that everything is working as expected but duct work could be improved, really just told to close more registers (static pressure readings in normal range when they measured and have now closed more registers).
Is there any other better or more informative measurement I could take? My daily energy use seems very high but I am not sure what metric is best to evaluate for the ducted system and I have no clue if I can accurately measure at low temperatures with such short heating cycles.
I attached a chart of my energy usage from this morning. Multiple Mitsubishi reps who have come and measured the system have told me the energy cycling is normal and due to defrosting. Seems strange and very different from what I see with the Daikin system. Difference between ducted and ductless? Thanks.
ps. This is an oil to heat pump conversion in New England and I have read all the articles on heat pump retrofit and design load calculations and tracked oil usage for a year prior and I don’t believe sizing is an issue but open to any thoughts. Daikin unit is 3 ton 4MXL36TVJU a couple years old on one side of the house and displaced about 40% of oil usage.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
From the second article you linked to you should be able to get a ballpark estimate of your heating load based on past oil use. Extrapolating from the design temperature should give you a ballpark estimate of your heating load at any temperature.
From the ashp.neep.org website you should be able to get a ballpark estimate of your COP at any temperature.
With the heating load and the COP you should be able to get a ballpark estimate of your electricity usage. That will tell you if you're in the ballpark. Then go from there.
Here's the page for your unit:
https://ashp.neep.org/#!/product/33968/7/25000///0
At 47F, at maximum output it's rated at 52,500 BTU/hr, 5.49 kW, COP of 2.8. So if it's really pulling 5kW at 40F it's producing close to that.
I'd imagine your design temperature is around 10F, so that implies a design heat load of around 100k BTU/hr. That sounds high but not impossible.
I don't know what you mean when you say "displaced about 40% of oil usage." Does that mean it was sized to handle 40% of the load? If so then maxing out at 40F would make sense.
Thanks for the replies.
I have 2 heat pumps, the Daikin was installed first and heats about 40% of house and reduced oil use about 40% over an entire winter when I still had oil. Haven't done a winter with the Mitsubishi ducted unit. The Daikin unit is awesome and energy usage matches that neep page you linked to.
My Mitsubishi unit on the other hand when the thermostat is turned up a few degrees goes up to 8kW at 47F even though max at 47F is 54,000 BTU @ 4.79kW. Mitsubishi reps that checked unit out said that is fine and the Mitsubishi senior engineering who answered email said that shouldn't happen? Not sure/ https://ashp.neep.org/#!/product/57027/7/25000///0
Using the heating load methods from past oil usage I get in the 50,000 BTU range for 10F design. Polar vortex oil fill-ups put me in the 68K BTU range.
Thanks for suggestion on COP calculations. My energy use was close to 90kW from midnight to noon today with temps rising from 20-mid 30 late morning as shown in pic. Using the data from just midnight to 6am and my estimated load of 40,000 btu at 20f I would be getting a total house COP of around 1.7-1.9, which seems a bit low but possible. From data on weekend for just the Mitsubishi (set to low because set to high it goes to 8KW) I calculated COP of 2.6 at 40f using (delta T on blower *1.08 to estimate*CFM from design book to get BTU ) which is possible but also a bit low.
2.6 isn't terrible. At max output NEEP has 3.3 at 47F and 1.84 at 17F, extrapolating between them gives 2.96 at 40F, which is in the ballpark.
At rated output it has 3.3 and 2.5, which gives 3.13.
What is interesting is that I previously calculated a lower COP of 2.5 at 60F. Only difference was I had the blower on high and not low and used the max rated BTU from the engineering docs. Energy use not coming from blower either. Will measure again when warm one day.
Also interesting that their engineering documents for the outdoor unit have much lower energy usage than the submittals (even when including the blower). https://mylinkdrive.com/viewPdf?srcUrl=http://s3.amazonaws.com/enter.mehvac.com/DAMRoot/Original/10006\DB_M-P0855_MXZ-SM_NAM(HZ)-U1_JA22.pdf
Using their own engineering output values for the heat pump the COP at maximum output would be over 4 at 35f or 42F with a 70f return (with just the outdoor unit). Drops some with blower but not that much.
What are you using for a thermostat?
To my eye that graph looks like the unit is cycling on and off. If the outdoor temp is in the 40s at your units minimum speed it may have to cycle to avoid overshooting the set point.
A defrost cycle on the power graph would display as spike to max speed that would last for 3 -5 minutes See my photo 5 heat cycles at min speed and a defrost in the middle of the 4th cycle.
I am using the MHK2 mitsubishi thermostat. It does look like short cycling from overcapacity but this occurs when cold (lower than mid 30s) and not when warm. My power is averaged over 5 min so I miss some granularity but a standard cycle looks like about 5-10 minutes ramp up, 30min run highest power (with about 10 minutes 1-1.5kW) , 15 min low draw.
Is your thermostat cloud connected? I recall a few other posts complaining that the Kumo cloud would not modulate the compressors speed just cycle it on and off.
Walta
Not cloud connected so that is not it. The cycling behavior is super strange to me and behaves like short cycling when cold (off for 10, ramp up delta T slowly (takes 15min), 15-20 min output with steady delta T of mid 30s, then off again). Seems to average about 20-25mins of good output per hour.
Outside temp was high 20s overnight then temp shot up to mid 40s from 8-9. Once in 40s I got a finally got a longer run cycle (@ about 3kW draw).
This could be symptom of low charge
replying to: @freyr_design -- Low temperature behavior does seem similar to symptoms I have seen described for low charge but lines were tested with a multi-day leak test and later the charge was removed / reweighed and verified to be correctly charged to Mitsubishi specs.
Did they check your low pressure switches? I would be tempted to see what pressures you are running at during those temperatures.
No idea if they checked the low pressure switches or what those even are. Contractor / Mitsubishi reps only here when in the 40s last spring. Can get a long runs at that those outdoor temp but energy usage 7-8kW whole time and told high energy because of ducting. Only thing I can think of is liquid line restriction but no response to that suggestion.
replying to: @freyr_design Only tried cooling a few times in summer but low delta T and low power (steady state draw was half of maximum energy compares to specs with very long cycle run times).
I don’t know about Mitsubishi control mechanisms but there is generally a low pressure shut off switch to protect compressor from running without refrigerant. It seems like this could be the issue as at lower temps outdoor your pressures will drop and perhaps trigger this. I wonder if this switch could be faulty. It could also very well be a defrost issue, perhaps there is not enough airflow across outdoor unit.
The short cycling based off only lower temperatures doesn’t really support the oversize theory or blocked ducts in my mind but perhaps there is a mechanism for this. You should be able to test duct resistance as well. You can test the thermometer by seeing if it is sending a signal when is cycles off (voltage meter to see if 24v are being sent, look at wiring diagram to identify which wire on outdoor unit).
Just a bunch of theories….
replying to: @freyr_design Thank you and much appreciate the thoughts and ideas. From what I can find the behavior is not normal and the installer does not seem interested in coming out because Mitsubishi reps have signed off on the install and operation multiple times. I measured duct temp for a cycle a couple times and it is completely off for 15-20min, 10-15 min ramp up, 15-25min high output. Some variation but full on-off cycle always less than an hour. On heat cycles I calculated BTU output I got to be about 50% of peak running BTU per hour. Defrost cycles and airflow for defrost look ok so maybe the pressure switches are the issue. Currently don't have a next step planned so need to figure something out.
No need to close registers - open them all back up.
Next, is there resistance backup heating installed?
From the 11/13 graph, it looks like it basically ran as expected overnight? If it was 18F and your heat load is ~40kbtu at those temps, it looks like it's using an average of 5-6kw after you subtract out the other heat pump. That's what the specs say: https://mylinkdrive.com/viewPdf?srcUrl=http://s3.amazonaws.com/enter.mehvac.com/DAMRoot/Original/10006\M_SUBMITTAL_MXZ-SM48NAMHZ2_en.pdf
Thank you, very helpful as I had not taken an average of the power use overnight and just ignored the shape which does make this run information look better (the shape/ run cycle a bit surprising because it seems like 15min per hour it is not blowing hot air - but will get back to that and measure correctly). I expect a whole house load of about 40K at 20F and with about 25K covered by the Mitsubishi unit (but maybe my calculation are to low or losses higher with the ducting).
Deleted
Typically the Mitsubishi thermostats have the option to use either the local sensor, the air intake sensor or the combination of the two for controls. This is set somewhere in the installer menu.
You can try to change it to see if it improves the modulation. If you don't have supply ducts near the thermostat, the best operation is typically by using the thermostat sensor for controls.
The outdoor unit also has some jumpers for setting max demand which limits the overall power. If you don't need the full power of the outdoor unit, you might want to change that.
There are other jumper setting as well, might be in the install manual that could help.
Thanks for the reply. Just this past week I tried a few different options for local thermostat locations and then tried the air intake sensor. I thought switching to the air intake sensor was helpful to the modulation BUT then the temperature dropped below 35 and the unit returned to the 20 minutes high, 10-15 off, 15 min ramp up and repeat cycle. Looking back through my data I don't think that I cannot find any day where I got more than 30 minutes of continuous steady state running (based on energy draw) when under 35 degrees.
Dave, did you ever get this issue with your Mitsubishi resolved?
Have you been happy with your Daikin?
I have a mitsubishi from 2018 and it was first put in with no external thermostat. I installed the MHK1 20 feet away and not directly in line with the unit. This fixed the short cycling issues, much better performance and heating just the first floor, using 18K. The upstairs was still heated by gas.
Then, a year ago, I got a new mitsubishi unit for the whole house. The whole house unit is 30K and it sends air through ductwork. I was using the PAR40 thermostat and it was set up incorrectly, and would read the temperature from the air intake which is way up in the conditioned attic space. So when we changed it to the thermostat it was better but... not great. The problem was the thermostat location at the top of a poorly insulated stairwell. Thus the operation of the unit created a draft that would shoot cooler air into the thermostat. We had to move the thermostat 2 feet away and behind a bedroom door, problem solved. (Also you can play with dampers on ductwork if you have those to balance the distribution of heat and reduce drafts caused by an imbalance.)
Now we have insulated the poorly insulated stairwell (finished last week). We should be able to move the thermostat back to its original location for the whole house unit. I know this is not your situation but it could be relevant. Since you have two heat pumps it may be best to do testing with one of them off, as the heat from one likely affects the other. We find that when the outdoor temp is 40 degrees we can heat the whole house with the 18K unit and leave the other off. At 20 to 40 degrees we can do the opposite. At below 20 degrees (i.e. 5 at night and 15 during the day) we will need both.
Finally check out the heat geek YouTube channel, where in the UK they are developing a software APP for Mitsubishi designed to make the unit run at a near constant power level during the day, ramping up the temp from around 3 to 6pm so that the unit can use the stored heat to not have to run very hard at night. The software already works on other models.