Tuesday, July 02, 2019

improving cell data with a refrigerator

I have been experimenting with my cell phone as the hot spot for my home internet. The pic is my Rogers cell based internet after placing my phone in the best location which I discovered here yesterday, on a box on top of my fridge.

Rogers is offering an unlimited data plan which was a $5/mo upgrade so I took it. I intend to drop my land line based internet to save by having cell only.


My current ADSL (phone line based) internet is the least expensive plan from the only landline provider in my area, Eastlink. I enjoy slow but solid internet, 5 Mbps download, 1 Mbps upload, unlimited for $75/mo. Rogers will now cost me $80/month for their cheapest wireless unlimited plan. (see their website and the pic above)

Before yesterday, my Rogers cell data and reception were not good inside the house. And highly variable, from un-usable to outstanding. Today, reception has not changed except that I found the best location from which to access my available signal. I am pleased with what I found but there are a couple of caveats (read on).

Where I live has two negatives for cell data. My house is low in elevation relative to the two area cell towers so much of the cell signal is blocked by surrounding hills. Second, my house is covered with a metal mesh based cement that effectively shields electromagnetic radiation, like cell signals. Two negatives.

I looked at 4G cell repeaters/boosters. Hundreds of dollars and not all work with Rogers. I haven't tried one yet.

I learned about cellular booster reflectors from old satellite dishes.

Instead, I loaded an app to measure cell signal strength. I am using Network Cell Info Lite. This one is for Android.

The gauge display shows main and next best cell signal strength in dBm. The gauge updates every second or so.

Most people don't think much about where they put down their phone. Or maybe your carry yours? Our phones are marvelous little radios which are sensitive to orientation and surroundings! Especially when we need them to work their best.

I have been exploring my house for my BEST signal. Signal inside my house varies over a huge range from -100dBm to -125dBm. The dBm scale is logarithmic. Lower numbers in this case are hugely better.

I can see that I am using band 12. This changes sometimes to band 4. Both of these are 800-900 Mhz. In rural areas like mine, lower frequencies like these will travel further and have adequate data rates to impress me when working well!

I don't get to pick a tower or frequency. Those are selected auto-magically between my phone and my provider.

I can see LTE on the display. This might mean 4G but it's complicated in Canada.

Yesterday I drove to the two towers that serve my area. look up your local cell towers, operators and frequencies Each is about 6 km distant, one is to the north and one is south. Using my app phone signal strength meter I measured the signal about 1 km from the towers at about -60dBm! But nobody lives there, well not many anyway...

By clicking once on a tower in the map, I learn that at the northern tower, Rogers uses 700MHz, 850 MHz and 1900MHz, Bell uses 700 and 850. At the southern tower, Rogers uses only 700 and 850. Bell is not on the southern tower. More specific information on all of the cell transceivers at a tower comes up in a table if you double click on that tower.

Back at my house, quite by accident, I found TWO spots where the signal is best. Both are on top of large grounded metal objects: my refrigerator and my clothes dryer. The metal case of both appliances is attached through their electrical cords to earth ground.

With my phone flat directly on top center of either the fridge or the dryer, my signal strength is the best inside my house. The fridge is slightly better at -97dBm.

When I check download data rate using speedtest.net from either of those spots, it is noticeably higher (twice as high!) than most anywhere else in the house!

But Upload data rate is much worse when the phone is placed directly on top of these large metal objects! It slows to less than 1 Mbps. But - if I lift the phone up about 5 inches high and place it on an empty cardboard box (about 1/4 wavelength, my phone is using 850MHz) the download data is still good and the upload data improves enormously!

Is it possible that the appliances are making a ground plane for the phone, helping with receive and transmit?

I tried different height boxes (two, three, five and eight) and the five inch seemed to be best.

I tried changing the phone's orientation, on either side, on the ends, rotating it slowly in different planes while holding it approximately 5 inches above the fridge and trying to watch the display. But placed down flat on the phone's back on the 5 inch box seemed the best (lowest) dBm reading. Minor changes if rotated slightly and slowly on the box.

I tried putting the phone over a large un-grounded metal plate on the kitchen table, with and without the 5" cardboard spacer box. The metal plate made no difference.

I rarely use my phone for voice so parking it on top of a box on the fridge is no problem for me. I can hear it ring anywhere in the house.

When I go out, the phone comes with me so the home network stops. I don't have any "smart home" devices that need always internet so this is not a problem.

If there were two phones in the house, one could set up sharing of two cell links with two hot spots (with bridging?) to further improve data rate?

If you try this, please let me know.

This experience has caused me to clean the top of the fridge. It was pretty dirty up there but cleaning had no effect on the signal strength.

Thanks for your interest.

George Plhak
Lions Head, Ontario, Canada

Update Jul 22 2019 Eastlink has been OFF since I wrote this and I have been accessing the web with Rogers cell only. I frankly haven't noticed the change, except for generally better responsiveness (plus!) and having to retrieve my cell phone off the fridge when I go out and then putting it back (negative). The actual internet data service is far superior to what I was getting with Eastlink (5mbs down 1 mps up) as this speedtest just now shows. I was worried about the throttling back "feature" but that does not seem to have happened yet. I get my latest full Rogers bill with usage and dollars on the new plan tomorrow.

Monday, June 24, 2019

hot wire foam cutter

I built a simple hot wire foam cutter to make parts for a couple of projects.

I wanted to slit a full sheet of 1.5 inch insulating foam into strips for a mooney wall. Another project, a water heater insulator, required large foam discs.

What started as quick and dirty ended up a very useful tool.


Using scrap wood, I began by adding a slot and a "fence" to a partial sheet of particleboard to make the table. The board was what I had, about 3 foot by six foot supported on two saw horses. I added a strip of wood lengthwise opposite the fence to support the front edge. The table is hung on the wall when not in use. [click any pic to enlarge]

Using particle board was problematic. When fed across the table, large pieces of foam would catch on the tiny sharp fibers sticking out of the particleboard. I later sanded and painted the top surface with several coats of latex paint which helped. It would have been better to use a smoother surfaced board.

The cutter arm is U shaped to support the cutter wire and slides in a groove under the table to adjust. It is then fixed in place for the cut.

I thought to make a variation with the hot wire slide-able to make cross cuts but haven't done that yet.

The slotted hole in the table is for the hot wire. I didn't want to cut through the fence on the back of the table. There is a minimum width that the wire can be set to, about 6 inches and a maximum of about 18 inches. The width of the slots for the mooney wall are 14.5 inches so this is a good amount of adjustment. Varying the dimensions could give other capacities.

The hot wire on the bench with power supply testing cutting action!

It turns out that the wire does not need to be red hot as shown, but just slightly less, sort of a dull red. I have an adjustable DC supply giving the wire about 4 amps at 8 volts. The voltage and current will depend on the wire type, gauge, length and the desired temperature so the adjustable lab supply is a good fit for a hot wire cutter. Normally the supply sits under the table and is connected to the two ends of the hot wire with test leads.

The wire I am using is heater wire (nichrome), typically used in toaster ovens. The first samples of wire I used were harvested from an old appliance. Later I found that I could easily buy small amounts of nichrome wire on Ebay from China so ordered 0.3, 0.4 and 0.5 mm coils.

I found that 0.4mm nichrome gave good cutting life and resistance to breaking. Getting the right temperature and the feed rate requires a bit of experimentation.

Nichrome wire expands when heated, as much as 5%, so there is a spring in the mount to put the wire under tension. When at the operating temperature, there should still be a bit of tension from the spring to get a straight cut.

The wire is simply looped through the eye bolt and wound around itself a few turns. The nuts on the eye bolts are adjusted to put the wire under tension. Normally I form, adjust and test the wire on the bench out in the open, then remove it from the arm, mount the arm on the table and then remount the wire through the slot in the table. It's easier.

Shows the smooth cut on blue type styrofoam. The edge obtained is very smooth.

Cutting foam discs with a pin through the center into the table as a pivot.

Using a thin template to guide the wire to produce a cutout in one of the discs.

Again using a template to make cutouts in one of the discs which has been already cut in half with the wire.

One of the many eight foot by 14.5 inch slabs for the mooney wall that was cut with the wire cutter.











Thanks for your interest.

George Plhak
Lions Head, Ontario, Canada

Detail under the table showing the mount for the arm made from scrap wood. The arm (with the wire removed) is inserted in a slot from the back and fixed in place with the screws that protrude to the right. Crude but adequate.

Thursday, June 20, 2019

water heater update

INDEX to the series

This graph [click to enlarge] shows the big reduction in standby energy use between my old (blue) and new (red) water heaters at different basement temperatures. The new uses much less electricity, about 25% less. Same capacity, same size heat elements. 25 year age difference.

What's changed?

The new heater has better insulation, a DIY insulated mounting base and is time of use controlled so it does not use peak priced electricity.


I will show the key physical differences between the new and old. The old was produced in 1994.

My old water heater had been quietly making hot water in the basement here for 25 years. It was not leaking and did not require replacement. It had no maintenance during the last five years with me. It worked well but could have been more efficient.

I pay tribute to this fine Canadian product by Giant Factories Inc. It owed me nothing.

To become more efficient, I purchased a competitor's product from GSW (division of A.O.Smith). Although Fergus is on the nameplate, the GSW factory in Fergus hasn't made anything in years.

I might have made a better comparison by getting a replacement Giant. Local stock product was competitive in price.

In any case, I will compare old product to new product. Both are standard electric domestic hot water heaters, now called "Automatic Storage Water Heater". Both are the same size (give or take a cm or so, I lost my actual measurements). About C$350 cost now. The technology of even the most simple commodity hot water heater has improved dramatically. Much of the improvement is federally mandated to improve efficiency. My old and new have the same capacity and the same elements but are 25 years apart in manufacture.

In other words, GSW's Fergus product of 1994 is probably much like the Giant of 1994.

So if you haven't thought about improving your trusty old hot water heater, you might save significant electricity if you do. Here is what I found:

The top of the old with the cover removed. I can see the steel tank where there is no insulation.

I am holding the anode which does not help efficiency but adds life expectancy. This anode rod (aluminum or magnesium usually about the size of my little finger) is completely corroded. An anode rod is replaceable but who ever does that?

The anode protects the inside of the steel tank by being sacrificially dissolved itself as a dissimilar metal, like a zinc anode on the hull of a sailboat. When it is gone, the tank starts to corrode so this one is well past its best before date. 25 years old! Probably never been changed.

Note the iron pipe fittings out the top for inlet, outlet and pressure relief. Copper lines were attached to these fittings with no thermal break (insulated coupling).

These UN-INSULATED metal pipes had been heat radiators into my cold basement. The pipe from the hot water outlet (the one on the left with the pressure relief valve) was always HOT to the touch. Lost heat.

Note also that the insulation inside is a glass fiber wrap with lots of empty space/air! Particularly over the very top center. Perhaps I pulled some of it off with the top? It could not have been this bad could it?

I didn't determine how the steel tank was supported inside the steel can but there must be some metal support structure from the tank to the outside which would probably cause some heat loss.

The new heater has about 2.5in / 6-7cm of foamed in place insulation at the top near the center. I am measuring to the steel tank through the anode access.

In fact I had to dig out some insulation to get access. I wanted to look at the installed anode. The top of the anode rod bolt will be my measuring point for water tank temperature once I insert a sensor and cover/insulate the opening I have made in the insulation.

The anode access has a cute cover plug but once removed, you need to dig out the foam down to the hex head on the anode bolt.

Note the insulated fittings for hot and cold on the top of the tank. The pipes that lead to these fittings are no longer warm to the touch.

Likely the foamed in place insulation completely supports the tank mechanically (unlike glass fiber) so no need for a bridge structure hence lower heat loss.

The fitting to the drain valve at the bottom of the tank was an iron pipe to a metal valve. Both good conductors of heat to the basement.

The old tank used to sit right on the cold concrete basement floor.

"Verified for Energy Performance" in 1994!

The drain is now a recessed plastic valve, a much better insulator.

I've raised the tank off the floor on an insulated base, making it easier to do the annual residue draining and reducing the heat lost to the cold concrete floor.

The wooden base (about 8 inches high) is topped with a 1.5 inch foam disk.

The diameter of the base is larger than the water heater by about the thickness of a batt of rock wool insulation. It could have been a square platform to be simpler to make.

I will wrap the entire new water heater with an insulating blanket after installing seismic straps. Not required by the code here but a good idea anyway for a 100 year event. Earthquakes do occur in Canada.

Time of use inhibit I wrote about here.

I plan to install direct energy metering for the water heater as it is one of my top energy consumers. NRCan on water heaters. Note that there are no storage tank electric water heaters that are Energy Star rated. I think mine will qualify when finished!


Thank you for your interest,

George Plhak
Lions Head, Ontario, Canada.

INDEX to the series

Nameplate for the old Giant.

The date code means the tenth month of 1994.

Nameplate for the new GSW lists Fergus, ON CANADA for address of manufacture?