Energy efficiency of air-source heat pumps with variable speed compressors

Maybe it will be interesting for us to study the relation between the compressor load and the efficiency of the variable speed heat pumps. Now most of domestic air-source heat pumps are based on inverter technology to maintain the compressor speed, so the major question is how the load is related to the COP/EER factors? Most of the manufacturers put the COP/EER for the rated/nominal power but what is the COP when the machine works on half of the rated power? Is it better or not? What is the effect of oversized inverter heat pumps in energy consumption? These questions are very interesting and important, that is why we develop a special real-time WEB based system for energy monitoring. The system measures the internal and external temperatures, also the electrical energy consumption. Also a full-custom controller was developed to control remotely the air conditioner and to monitor the environment parameters, named HVAC controller. The HVAC controller implements also a WEB interface for monitoring the parameters; also it implements password-protected pages for controlling and setting up. Hope the information and the charts will be useful. The IP address of the HVAC controller is 77.76.18.134, while the charts are available at: energy <dot> bjr-labs <dot> com

Very interesting Boardman.

Perhaps you can tell us more about the system here - a brief summary? Folks interested in the full details can get them off your website.

Interesting to see the outdoor temperature range, over the past month. Gets quite high on some days. Intriguing. :)

The system consists of two modules – HVAC controller (which is hardware module) and external online database real-time information system. The communication between is done via Internet connection. The HVAC module reports the data for the temperatures and energy consumption on every minute to the public database system, which draws the charts and stores the data for future analysis. Soon it will be implemented a tabulated report for energy consumption per week/month and year for 2 tariff (day and night consumption); also total cost for the electrical energy; the average temperatures etc.
Every HVAC controller has its own unique ID, so multiple modules can store data in the database system. But the HVAC module is the more interesting part of the project. It is a small microcontroller based board including 3 processors: one is responsible for remote controlling the air conditioner by implementing a infra-red protocol (the HVAC controller can set the air-conditioner just as like normal remote control of the unit without any external connections); the other is for temperature measurement, also the energy measurement, and the third processor is for the integrated WEB based system and the running TCP/IP stack on it.
The energy measurement is done by a special chip (MCP3905) and it measures the real/active energy of the air-conditioner (or the heat pump). It is quite precise with 0.5% precision of the full range from 0 to 20kW, also the minimal step is 0.2W. So the chip can measure adequately the low power consumption. The energy measurement module includes precise real-time clock for the tariff switching, LCD display, which print the current instant power, also the status of the real-time clock and the total counters for the tariffs. The energy module is battery powered to store the counters after the power fail events.
Temperature measurements are based on DS18B20, which are precise digital thermometers. The maximum error is +/-0.5 degrees in the full scale (-55 to +125 degrees). The communication between the thermometers and the controller is done via digital 2-wire bus that eliminates the error. The connection wire can be long up to 100 meters without affecting a measurement error. Every thermometer implements its own unique hardware address. That helps for the main host to delimit the internal/external temperature samples. The external thermometer is placed nearby the heat pump evaporator. The internal thermometer is placed in the center part of the living room. The HVAC controller also implements two high bright 3-digit LED displays to display the temperatures independently.
In short that is the description of the hardware. There is another hardware developed for the real-time monitoring of the temperature and humidity at:
dsnet <dot> tu-plovdiv <dot> bg/website/thm.jsp
There we can see the relationship between the temperature and the humidity changes. It is also interesting. But for the energy efficiency I made some measurement of oversized heat pump with nominal power of 4.8kW, installed in environment with thermal looses of about 1.7kW. If this is interesting, I’ll post the results also.

Please do continue posting your results. I find this concept very interesting.

is this a self invented en manufactured device, or imported and try to sell it in Europe?

Ice

Hi,

The system is self invented and designed by me. At the moment it is in a test process, also there are some bugs to be fixed. The first idea was to be able to remote controlling the HVAC equipment using simple WEB browser and also to monitor the environment temperatures. The HVAC controller consist of a small board with Ethernet connector and it can do all of the mention functionality.
The second idea was to implement electrical energy meter and to develop external database system. The database monitoring system covers different applications. It is able to monitor the relation between energy consumption and temperatures, so it is suitable for energy efficiency analysis. It is possible to measure the thermal loses of building or rooms. This is done by measurement and logging the energy consumption of electrical stove (heaters) installed in the monitored environment. The coefficient of performance of electrical stove is near 1:1, so when we measure the electrical energy consumption we can easy determine the heat looses of the environment.
Some experimental results show that the coefficient of performance (COP) of heating pump is 4.2 at +7 (external temperature) and +21.6 (the internal temperature). The heat looses measured with conventional electrical heater were 930W; the measured electrical consumption of heat pump was 220W. The official COP for the heating pump in the specification is 3.87 (when heating pump works at its nominal rate at 4.8kW). So the COP is better for part load than the nominal load.

Climacheck made something similar

I'm also busy with something similar.

The person doing the programming for me made something similar for a heat pump manufacturer in The Netherlands.
COP is important but for Europe the SPF will become even more important. The PCB is already foreseen for this.

There's also a standard heatpump with these functionalities already build into it. I can't remember the name right now.

The manufacturer uses this for marketing his products. You log in to his site and you can consult all the different heatpumps they installed. You then can search one in your vicinity and monitor it to see its performance over a longer period. You then can see yourself how well it performs.
While the system is connected to the WWW, it calls also the technician when certain parameters goes beyond preset values. The technician can call into the system and check all the parameters.

Please forgive my ignorance, but how can your determine efficiency on a a variable speed system only using power draw and temperatures. Firstly these types of units have vairable speed indoor and outdoor fans, which will change the working pressures within the refrigeration system, which will effect the actual COP. Certainly the efficiency changes with changes in the process variables, but with these types of units factors are built in which relate to comfort, not efficiency (wind chill factors). Also on heat mode many makes have an off set in the control to allow for low level comfort. (high walls)
Actual efficiency is not the issue, but power draw, in many cases its turn down (reduced heating/cooling effect),

Quote:

---

Originally Posted by mad fridgie

.... but how can your determine efficiency on a a variable speed system only using power draw and temperatures. Firstly these types of units have vairable speed indoor and outdoor fans, which will change the working pressures within the refrigeration system, which will effect the actual COP. ...

---

Agree for 100%, if you don't measure pressures, you're nowhere with some but not all the necessary measurements.

Quote:

---

Originally Posted by mad fridgie

Please forgive my ignorance, but how can your determine efficiency on a a variable speed system only using power draw and temperatures.

---

Actually I use only the power chart on the following manner:

• I use the current environment and heat loss measurement for the current internal and external temperatures. It the environment temperature changes are quite smaller for a long period of time, and then the heat loss of the environment is steady enough.
• I measure the electrical power with an electrical heater for a period of 1 or 2 hours. That time I see the temperature charts to be linear (there must no difference in the building and external temperatures). The electrical heater performance is 1:1, so the electrical power, consumed to support the internal temperature is the same with the heat loss of the building.
• Then I measure the electrical consumption of the heat pump, in the same environment parameters as in (2). No temperature changes, no temperature differences than in (2) will cause the heat pump to work in the same condition as heat loss of the building.
• The electrical energy ratio will show the coefficient of performance for the current environment parameters.

What I mean by efficiency of variable speed system? In the case I use, the average heat loss is about 1kW; the nominal power of the heat pump is about 4.8kW. So, the heat pump works in 1/5 of the nominal power. I do not need to know the pressure in this case, because I measure the real heat loss of the building, so the heat pump compensate this measured heat loss. The temperature charts are important to monitor the difference in the temperatures, so if there is a small difference then the measurements are not correctly enough to be representative.

Under high end test conditions, your method may work (steady state), but heat loss from a building is not linear.
Wind (strength and direction), solar, humidity, rain, thermal mass, stratification and air infiltration. Each of the above can have a large effect of the amount of heat required, you can not just base calculations on temperature difference. The above allows effects the required set point (comfort level) human intervention.
In your test chamber how did you ensure that your heat loss was a constant.

Quote:

---

Originally Posted by mad fridgie

Wind (strength and direction), solar, humidity, rain, thermal mass, stratification and air infiltration. Each of the above can have a large effect of the amount of heat required, you can not just base calculations on temperature difference. The above allows effects the required set point (comfort level) human intervention.
In your test chamber how did you ensure that your heat loss was a constant.

---

You’re absolutely right. Factors like: wind, solar, humidity, rain, air infiltration etc. are important; that is why I wait for a moment when for a long period of time (day or two) the environment parameters are very close, for example I start in night to reduce the solar effect, no wind environment, also the infiltration is the same for the measurement of heat pump and electrical heater (chamber is in the same state for the period of 3-4 hours); that period is too short for serious environment change. First I measure with electrical heater, then I measure the heat pump and finally another experiment is done with electrical heater to ensure the results are close to the first electrical heater measurement. So there is an error, but the error is about 2-3%, which is good enough. But if we use a laboratory insulate chamber

, the system can be used for a precise and accurate measurements.

A fundamental question is what is the COP factor change vs the load of the compressor in the inverter air conditioners (heat pumps). Some manufacturers like Toshiba show an increase when the compressor works under the nominal/rated power. I think that when the compressor works under the nominal, the pressure difference is smallest, although the pressure in the condenser is smaller than in the full power of the compressor. The smaller pressure in the PH diagram will cross much more enthalpy than the biggest pressure. So, the smallest compressor load will reduce the compressor work. The COP must be increase theoretically in lower pressure? What do you think?

Quote:

---

Originally Posted by Peter_1

I'm also busy with something similar.

The person doing the programming for me made something similar for a heat pump manufacturer in The Netherlands.
COP is important but for Europe the SPF will become even more important. The PCB is already foreseen for this.

There's also a standard heatpump with these functionalities already build into it. I can't remember the name right now.

The manufacturer uses this for marketing his products. You log in to his site and you can consult all the different heatpumps they installed. You then can search one in your vicinity and monitor it to see its performance over a longer period. You then can see yourself how well it performs.
While the system is connected to the WWW, it calls also the technician when certain parameters goes beyond preset values. The technician can call into the system and check all the parameters.

---

Hello Peter

I would be appreciative if you would post some details on the manufacturer you referred to above, when you get an opportunity. i.e heat pump manufacturer with remote monitoring of installs.


Thanks

Sólás