Why bother?
Flyers in our phone bills started inviting us to sign up for high speed Internet in 1999. We called each time only to learn that the service was not quite ready in our area. We tried a satellite connection but it was sometimes flakey and still much less than high speed for a lot more money than dial-up. Since the big companies did not really seem interested, it was time to do it ourselves. This is the saga.
The setting
We’re only 50kms from Ottawa in rural West Quebec. In this picture taken from the tower site, the Parliament Buildings are on the horizon in the center. We can actually see the fireworks on July 1st from our front deck (albeit just peeking over the trees).
The rural part is the key though: copper telephone lines are old and the telcos don’t want to invest in new lines for small and spread out customers. Luckily, our 100 acre hobby farm is tucked into Gatineau Park, and the back half includes part of the prominent Eardley Escarpment—and along our eastern boundary with the park, we have a 60m elevation rise—with the final 30m almost a cliff—with bare Canadian Shield granite at the top. We wouldn’t be alone either: there was lots of evidence of bears enjoying blueberries and the acorns from the many scrub oaks nearby.
The selected site faces SW but by climbing a nearby oak tree, I could see that we would have good views S and E if we could get above the trees. As well, the western exposure would be ideal for either solar or a wind generator to power the batteries. Old strings of barb wire fence run along the ground just east of the cliff top delineating the boundary between our land and Gatineau Park, but still leaving us plenty of room to erect the tower and secure guy wires.
The plan was to haul the equipment up the steep slope, so I cleared a path to a trail near the bottom where we could get close with a truck. It is definitely a cardio workout to get to the site—but adds incentive to get generating capability to avoid regular climbs with heavy batteries!
The players
This was a local affair. Aptly-named Cliff is the local wireless provider, and provides the antennas and electronics. Eric owns the land and looked after tower erection logistics, base preparation and power generation. Erwin contributed some old TV antennas that made up the tower. Chuck helped get the tower vertical. Hugh welded a solid base to support the tower. Paul helped Cliff put up and tune the antennas.
Erecting the tower
The old television antenna came as four 10-foot sections, with the top one tapering it to hold a 1.5 inch pipe. They were surprisingly light, and quite easy to haul up to the site. However, we needed a sturdy base and strong anchors for guy wires to hold everything solid—and that meant drilling into the granite.
A gas-powered drill and diamond coring bit from an Ottawa rental location seemed like the best way to go. The drill wasn’t too heavy, but the 6-gallon tank of water we needed for every two holes was a challenge! As well, one of the guy wire anchors needed to be located on the cliff edge, so Eric had to tie on to manage the drill on an angle while leaning back over the edge.
After a couple of hours of sweating and wrestling with the machine, we had some nice core samples and six holes. Three were for bolts to hold the base plate, and we installed nuts with washers underneath to enable us to adjust the level. Special concrete epoxy secured the eyebolts for the guy wires.
We set the bolts in epoxy and left it to set for the night.
By the next morning, the epoxy was solid and we prepared to erect the tower. Although the assembled tower could easily be lifted by one person, it proved to be much more difficult to tilt it up! We ended up tossing a line over a branch of a nearby oak tree, and using a block and tackle and a forked branch to help lift it to lean against the tree. Three of us were then able to lift it in its vertical position to drop the base over the 3 anchor bolts. We all breathed a sigh of relief as the nuts were tightened.
We had already attached the guy wires, so it was a quick task to attach them and adjust the tighteners to get the tower steady. Cliff hooked up his safety belt and’very gingerly for the first few metres’made his way up the tower. “Wow! What a great view!” As he climbed past the height of nearby trees, Cliff could see the full panorama of the Ottawa Valley below. I soon followed, and was surprised at how steady the guyed tower really was.
The transmitting antenna needed to be installed and pointed to the new tower, so we cleaned up and tightened the bolts before leaving. I went back up later to add a flag so we could more easily spot the tower from the transmitting site ~10kms away.
That night, the satellite connection seemed even slower than before... just as the dial-up connection had seemed impossibly slow while we had been waiting for the dish to be aligned.
Speed at last
With a new 5GHz antenna pointed at the new tower, Cliff and Paul went up to install the 2.4GHz access point antennas. We had decided for now to just connect them to a 12V battery, and monitor the power drain so we would have a clear idea of how much power would really be needed. Then, later, I would add a wind generator or solar panel array to charge the batteries—and eliminate the need to haul heavy batteries up and down the cliff.
Cliff attached the antennas and Paul wired everything together. A quick test on Cliff’s laptop showed a good signal. Back down at the house, we were soon able to get an antenna mounted on the wall and get a fast connection to my router. Speed at last!
Over the next few days, Cliff monitored the signal strength as he went around the area. It turned out that the new access point had a much wider reach than expected, and would be able to provide service to even more neighbors. The decision to use batteries for the short term proved even more prescient: a third 900MHz antenna was installed to get a signal through to people partially hidden by trees, so now the power draw was already higher than the initial plan.
However, the batteries on the rock beside the tower base proved to be of interest to a bear one night: one battery was tipped over and the leads were torn off the other one! Bears are curious, and like to investigate anything new in their territory, so after that incident, the batteries and wiring box were lifted up into the tower.
Power at last
I purchased a deep discharge sealed 12V battery made for renewable energy use, as well as a 30A charge controller and an 80W solar panel. Initial tests were encouraging: the panels charged the battery very quickly in even fairly cloudy conditions.
However, bear and racoon tracks on the temporary solar panels we had laying on the rocks showed that an enclosure was essential. I decided to make a well-insulated box on legs so the battery compartment would be at a comfortable level. It is clad with metal to discourage bears and other inquisitive creatures, and securely attached to the NE side of the tower.
Getting the partially-assembled parts up was a chore, but Cliff and I were able to get it assembled and the solar panel firmly attached to the tower. We ran out of light before we could finalize the panel orientation, but were able to do that the next morning. I insulated the box and tidied up the interior by attaching the various power parts to a panel on the back wall. There is room for 2 additional batteries, but for now one should be sufficient.
The final item added was a trail camera to record any activity at the tower. We may replace this later with an IP monitoring camera, but for now, it can get a picture of anything at the tower day or night.
The trail camera images are remarkably clear, and even the night shots show lots of detail. This picture was taken as I finalized the inside of the enclosure: I'm on the other side of the tower in this one; most of the others just showed my back as I tidied up the inside. A secure door slots into the opening shown here.
Inspection time: And here is why the enclosure is so essential: shortly after I completed it, the trail camera captured this little guy checking it out! From the size, I assume it is a black bear cub from this year—and probably one of the bears whose signs we’d been seeing earlier. The trail camera took this photo at 17:00 on October 10—less than half an hour after I’d been up there to check the panels. I wouldn’t be very concerned about meeting one this size, but the mother would likely still be around, and she would be more of a problem.
Next steps?
I left room in the enclosure for additional components. We are considering adding a remote monitoring system that could monitor temperature and voltages, but also reset the radios remotely if necessary. With the IP camera connected to it, we could view things live from anywhere. We may need to have some heat in the enclosure for the most severe winter cold, but of course that could add significantly to the power requirement. I’ll continue to monitor the power situation to assess whether we need an additional battery or more generating capacity.
Winter update: After 12 days of no sun in November, the batteries ran out. I lugged them down and recharged them, but at 30kg each, I won’t be doing that again! They lasted for a few weeks more, but the lack of sun and the increasingly cold weather took its toll and the batteries were drained again by mid-December. It looks like we will have to add an alternative generating source—probably wind—but that will have to wait until it is easier to get back up there.
The image above shows the view looking east and southeast from the top of the tower. The edge of the Eardley Escarpment is visible on the left all the way to Gatineau, and the Ottawa River flows along through the middle of the view. The flat loam fields in the foreground were once part of the Champlain Sea.
Stability at last! During 2009 we added a wind generator to try to get more reliable power. It looked like it would work, but the existing charge controller couldn't cope with the power it put out. Moreover, the WG controller had a fault that made it unreliable in strong gusts. While we were away during 2010, Cliff removed the WG and added a new larger solar panel array with a matched controller. This seems to have solved the problem, and except for a minor outage caused by a leak in one of the wire seals feeding an antenna, everything has been running without fault for more than a year now (summer 2011).
Now maybe I can move on adding more power-draining devices up there!