Monday, July 25, 2016

insulated refrigerator freezer result negative?

INDEX to the series
This graph from last night shows the electrical energy use of my newly insulated refrigerator freezer over about three hours with virtually nothing else on in the house.

I was surprised to find that the energy used last night was greater with the insulation I added yesterday! Until I realized that the last time July 10 I looked carefully at refrigerator usage, the temperature in the house was MUCH cooler. We are having a bit of a heat wave here so the temperature this morning in the house is 21C. You can see the little thermometer in the top right of the graph. Last time it was 12C. I would expect the fridge to be working harder when the ambient temperature is warmer. So I don't know the effect of the insulation yet. I have changed two variables and you can't do that. I'll have to repeat the measurement when the overnight temperature is cooler.

The temperature is measured outside by the Blueline Innovations device on my electricity utility power meter. Since I have the windows open, I would expect the inside and outside temperatures to be about the same. I also had the windows open on July 10.

So last night, I unplugged my small chest freezer downstairs about 10pm and plugged it back in about 5am this morning so it was not running during the recording. The fridge and the freezer have similar signatures so if I want to see one or the other, I unplug the other one for a time.

Everything in the downstairs chest freezer is still frozen solid although most of the light frost is gone from the inside walls and the plastic bags of food. There are still big chucks of ice in there on the walls after seven hours off. I have been watching this fairly carefully since I have been keeping my fridge and freezer OFF with timers during my peak period. They must keep at a safe temperature even after six to eight hours off. If you have a house full of teenagers using the fridge often, this might not work for you.

Other than the refrigerator freezer upstairs, only the network is running (which is how I get the data from my Blueline energy monitor), a couple of night lights and a phone charger. These make up the baseline load of about 70 watts. The base load is a bit lower than the previous tests because I not only turned off the computer but switched off its power bar.

Here is the graph annotated with guidelines and measurements added with CorelDraw. You can use your own favorite graphics program (or a ruler on a printout) to do the measurements and calculation. The Blueline software does not do the appliance calculation.

What I want to know is the area under the curve. The curve in this case is the square wave which represents the time (horizontal axis) that the refrigerator is running. The vertical axis shows the power demand at any particular time. The power times the time is the energy used in kWh.

Looking first at the vertical, when the fridge is ON, the house is drawing about 220 watts. I am eyeballing this by drawing a line through the middle of the tops of the square waves. It's an average, but you can see the fridge is a pretty consistent load when it is ON. When the fridge is not running, the house still draws energy (base load). Again, I am eyeballing this through the troughs of the square wave at about 70 watts. So the fridge is using the difference (220-70)= 150 watts, the same as before.

I would not expect the insulation to affect the amount of power at any time since the motor will draw the same current as before. What should happen is the motor will run less often.

Lookng at the horizontal time axis. It is hard to read exactly the Blueline graph so I am going to use some rulers. CorelDraw lets me lay down these rulers and tells me the distance. I have labeled one hour as being 3.12cm. The units of length don't matter as long as my units are all the same. I am going to use distance to compute time.

Across the top, I see that the fridge is ON an average of (0.92+0.89+0.91/3) 0.90cm or (0.9/3.12) 0.3 of an hour (about 20 minutes). I see the cycle time averages 2.15cm or (2.15/3.12) 0.7 of an hour (about 42 minutes). So I can say that the refrigerator is ON (0.3/0.7x100) 42.8% of the time.

So at this rate, based on these samples, over a whole day my fridge would use (power x time x duty cycle) (0.15x24x0.428) 1.54kWh or (1.54x364) 562kWh per year.

There is more to the story. As I was adding the insulation on the outside of the freezer compartment, I was wondering if I would have to re-balance the controls for the fridge and the freezer. At one point, several hours after finishing, I did adjust the fridge by one notch but then moved it back. I can't honestly be certain that I put it back to the same exact place. Both the fridge and the freezer controls are in the middle of their ranges.

What did surprise me this morning though, was that I had been playing with the "Exterior Moisture Control" and had mistakenly left it in the ON position. There are strip heaters around the door openings that are intended to drive off condensation on the cold door edges. I did not mean to have this on for last nights recording. This would definitely have used more power. I am not sure when the heaters come on but likely they are turned on and off the same as the compressor. But I still see the same 150 watts? Puzzling.

So THREE variables!

Thanks for your interest
George Plhak
Lion's Head, Ontario, Canada

INDEX to the series

Sunday, July 24, 2016

superinsulating my refrigerator freezer

INDEX to the series
I've spent the better part of the day adding more insulation to the freezer compartment of my refrigerator. It may seem ugly but some experiments are ugly. (click any picture to enlarge)

I am doing this to 1) save energy, 2) have the freezer stay cold longer when I inhibit it's coming on during my peak period and 3) prolong the appliance life since it won't hopefully be working as hard (as often).

My fridge is a major user of energy. It is older and not as well insulated as those available today. Mine uses only about 150 watts but it runs about one third of the time, 24/7 if I let it. I am intentionally keeping it off during my peak time with a plug-in timer. I've calculated that it currently uses about 412kWh per year (before today). Please see

I've used "superinsulating" in the title in the sense of being on top of or superior to what was there before. Not in the sense of some new technologically advanced stuff. In fact, I am using quite ordinary materials.

The best insulator is a vacuum but we can't economically use vacuum as an insulator for a large appliance like a refrigerator. The result would be heavy and expensive, making a suitable vacuum container to surround the cold space. That would not be practical.

The next best insulator is air that does not move around (convect). Foam is a very inexpensive and lightweight material which traps air and keeps it from moving around. The more foam thickness, the better the insulation.

That's what the manufacturer uses - foam. An 18 cu foot refrigerator like mine in a showroom today is larger by a couple of inches. This is how they have made refrigerators more efficient - thicker insulation. Compressor technology has not changed all that much. The controls might have improved efficiency slightly but the best control improvement I can provide is to keep the fridge from coming on during my peak time. That is a feature not available on today's refrigerators, even the most expensive models.

So I have added insulation on the outside. Adding on the inside would make the space smaller of course but would accomplish the same thing. If I open the door of a current fridge the space on the inside is about the same as mine (18 cu.ft.) but the wall is noticeably thicker and the overall size is larger.

I decided to add insulation on the outside and to check the effect on my old fridge. I can get a new 18 cu ft fridge for $800-2500 and probably have wonderful new features that I don't need except for the increased insulation thickness. I have now added extra wall thickness around the critical freezer section of my refrigerator for about C$20.

I will have to replace mine eventually of course but why NOW when I can do this fine experiment? I have tried to apply the insulation in a workmanlike manner that will be solid for at least a year or two, possibly longer.

For the top and sides, I have used found material, clean and intact surplus styrofoam packaging material that conveniently was in 2" thick block form about a foot square each. I would have used nearly a full 4'x8' sheet if I'd had to buy (Styrofoam SM sheet cost in Canada about C$35).

It might have been easier in some ways to cut from a full sheet but the price was right and the reuse aspect was a plus.

More surplus styrofoam in the form of trays I found at the local recycling depot probably recently carried a large appliance like a microwave oven. They were just slightly smaller than the door and the back of the freezer section.

I filled the trays with a layer of rock wool (Roxul) that I had on hand for the house. I had to split the rock wool down to about 1.5 inch thickness to fit into the trays. The trays filled with rock wool were then attached to the front of the door and the back of the freezer.

The outside is covered with a reflective bubble wrap like material which was really the only thing I bought specifically for this project. A roll of Reflectix Staple Tab Insulation 16"x25'/0.41mx7.6m) cost about $20. I used virtually the whole roll.

Plus lots and lots of packing tape (probably close to a full roll), a few strips of wood and about 10 wood screws to hold the wood to the metal skin of the fridge in a few key spots.

This is not meant to be a complete how to description but rather an overview of what and how I did it. If you would like to know more, please comment below.

I hope to have some preliminary results in a couple of days.

Thank you for your interest.
George Plhak,
Lion's Head, Ontario, Canada

INDEX to the series

Tuesday, July 12, 2016


INDEX to the series
This is a graph of my freezer's energy use last night. To isolate the freezer from the refrigerator since they look similar in power draw, I unplugged the fridge from about 22:00 to 3:00 so it is not running during the time on the graph. I zoomed on the freezer cycling, not including the fridge startup to get the max zoom on the vertical axis, the energy use of the freezer.

My freezer is an old smaller Woods chest freezer sitting on the cold basement floor. It must be from the 1990s. Sometimes I notice condensation of the top of the door so I don't think it is very well insulated. But it is quiet and does it's job.

I can see on the chart that the freezer follows a regular cycle like the refrigerator, drawing less power about 100watts but for longer, about an hour every couple of hours. This is going to vary a bit, mostly with the room temperature in the basement. These days that is about 23C. In the winter much lower. I don't open the freezer door very often, perhaps once every couple of days.

The fridge and the freezer are like dripping taps. They don't use much power but they are always on. What energy they use goes in cycles on and off. At any moment they might be on or off. Digital? More like Binary?

The shape of the freezer energy signature is interesting as it ramps down slightly as it runs. The spike on the first peak doesn't repeat on the other two. This is probably the freezer compressor motor starting up. Most household motors start with a heavy burst of current known as the "locked rotor current" but that only lasts for a second or less and then the motor current drops dramatically.

I think a limitation of the Blueline equipment is that the samples are taken every minute or so. Blueline may be limited by the hydro meter sample frequency since it gets its data from the hydro meter on the outside of my house. Some guy named Nyquist figured out that the sampling rate had to be at least twice the data rate but thats another story... you can only see slow data with a slow system.

Here is the longer view this morning looking at the past 12 hours.

The freezer you can now see relative to some other appliances. D is the dishwasher. W is the water heater which comes on with the dishwasher and also about 4:30am. You can see that the water heater runs much longer when triggered by the dishwasher. There was no hot water used after the dishwasher. F is the refrigerator being plugged back in. S is me out in the shop with some of the big overhead lights on, doing some painting. T is the toaster oven warming up a bagel.

Here I am in CorelDraw taking a more detailed look at the freezer data and figuring out a yearly consumption estimate from last night's data.

You can use your own favorite graphics program to help with the visualization and calculation. You can be as detailed as you want to be or just eyeball the data to get a sense of what is going on. After I do a couple of calculations, I think I see more in the data. As I said previously, I am disappointed that the Blueline software doesn't do more of this calculation for me but I don't mind doing this for the knowledge I am getting about my actual use real time thought whole house monitoring.

I realize now that I did the calculation incorrectly for the duty cyle for the refrigerator yesterday so I'll go back to correct that.

Looking first at the vertical energy axis I have placed three numbers in watts reading off the rather bad axis which has no tick marks, I see that my baseline last night was a little lower than I expected, about 90 watts. This is probably because I didn't leave any lights on by mistake. The baseline is the computers, the network (which sends Blueline my data from the actual power company meter), a few CO2 detectors and some miscellaneous chargers and power bars.

Next I estimated the peak ppower draw of the freezer at 215 watts, (215-90)= 125 watts not including the baseline. When the freezer has run for a while just before it turns off, the power draw falls to 195 watts or (195-90)= 105 watts. Since the droop is approximately triangular, I will use the rough midpoint of 115 watts for the calculation. About half the time, the freezer uses more, half the time less, so it should be about right.

Now I calculate the time duty cycle, in other words, what percentage of the time the freezer runs. On the chart, 1.21cm = 1 hour. The numbers at the top (0.91, 0.86, 0.85 are samples of run times shown in cm. At the bottom are the lengths of the cycles (2.03, 1.99). I can't find the length of the third cycle since the rising edge of the next ON is off the right of the chart so I'll use only the first two samples. Using the ratio of the lenght to reepresent time, I can say that for the first cycle, the freezer was on (0.91/2.03X100)= 44.8% of the time, drawing 105 watts. During the second cycle, the freezer was on 43.2% of the time. I'll use the midpoint of 44%.

Based on only these two samples, for a whole day, the freezer would use (0.105x0.44x24)= 1.1088kWh or (1.1088x365)= 404.7kWh per year or virtually the same as my refrigerator (412.2kWh)!

Thanks for your interest,
George Plhak
Lion's Head, Ontario, Canada

INDEX to the series