camp analysis

American West: I have been fortunate to travel during the course of my work and other adventures, and am most fond of our National Forests.

The following analysis provides a bit of perspective on these trips, starting with the distribution by state:


Next, we add specific geographic location and a gradient indicating elevation in feet. Camp names are visible with a mouse-over. Zoom in with a mouse scroll to get a better sense for the terrain at each site.


Adding time to the location data produces a 3D map. Latitude is highlighted by the color gradient, with southern latitudes in red. While this graph is disconnected from a geographic background, you can still make out the distribution. By clicking, holding and moving the mouse, you can rotate the view. With latitude along the left and longitude along the bottom, it is quite similar to the previous map. Now, rotating away from that view, the time distribution become apparent. Discrete trips cluster in groups, and gaps from seasonal variability are observable. 

Similarly, comparing month of visit to latitude provides some insight into travel patterns. The color gradient indicates elevation, as in the first map. A curve, fit to the data points, demonstrates a pattern of visiting southern, low elevation latitudes in the winter, and high latitude/elevation sites mid-summer.


And here, we have some additional analysis of trip distribution. Two conclusions I draw: My enthusiasm (and mileage) is growing, and I am excited to get out and explore again each February!


Thanks to the people at Ultimate Campgrounds for helping me find many of these sites. And thanks to for visualization support.


auto tent



The concept of the auto tent began for me in the summer of 2014. I had been thinking about self-contained living while on the road. Teardops, Airstream, Westfalia, Sprinters. All terrific solutions for certain types of travel and budget. But none provided the flexibility I wanted. For my specific use, I was seeking compact, and compatible with a standard vehicle and roof rack. No towing or vehicle customization. Yet still providing comforts similar to existing solutions.

I discovered the well established industry of roof-mounted tents. Many options, from simple nylon tents on a platform, to expanding fiberglass structures. The Maggiolina, a staple of the luxury safari, and Cascadia Vehicle Tents being two notable types. Become familiar with the CVT soft tent bag, and you'll start seeing them all over the place. At least here in the Northwest. Yet even the simplest of these cost more than $1,000, with more robust fiberglass models breaking $3,000.



Design Process

And so I began to refine my requirements:

  • In basic terms, a sleeping shelter for two.
  • Constructed from available materials. Steel, wood, standard hardware. 
  • Using techniques within my capabilities.
  • Mounted to a roof rack, with flexibility for alternative rack types.
  • Weighing less than the legal limit of 165 lbs while in motion.
  • Possessing a reasonably aerodynamic form.
  • Providing comfort beyond that of a tent.

An argument could be made for simply fabricating a large platform, on which you could pitch a standard tent. This meets many of the requirements. But that's hardly the design challenge to monopolize my thoughts and free time. An elaborate build, it is.

To summarize a long process of crude sketching and research, I arrived at the clam shell structure. The base would consist of an enclosed box, open at the top, and mount to the roof rack. The top would be generally aerodynamic, and attach to the base via hinge at the leading edge. The two halves would form a sealed vessel when closed, and open to reveal a living space with canopied walls and a hard roof. The tent would be entered from the rear of the vehicle.

The hard shell was an early focus. Aside from the attractive design challenge, a rigid structure provided advantages such as improved foul/cold weather performance and opportunity for advanced features. IoT tent!

The auto tent's planar design is in stark contrast to the smooth fiberglass forms of existing hard shell tents. This primarily reflects an interest in a simplified prototype. Maximize interior space and simplify fabrication while maintaining a generally aerodynamic form. Concepts in development for future auto tents incorporate more fluid forms. This planar structure can be seen clearly in the metal skeleton:

A rigid metal structure, built from square tubing and bar stock, establishes the form and provides support for panels cut from plywood. Panels would then be set into the frame and sealed along all seams. This seal would be critical for weatherproofing. Note the simple shapes, again designed to aid prototype fabrication.

The roof rack mounting bracket was perhaps the most difficult design challenge. The mounting system would have to be confidently secure, safely carrying the weight of the tent and occupants. Several hundred not insignificant pounds. After assurances from my trusted mechanic, that yes, the roof and rack can carry the weight, I moved forward. Staying consistent with my design philosophy, I ignored commercial solutions and instead would attempt to fabricate a custom solution. A strip of perforated bar would extend along the sides of the tent, providing a base to which the brackets would attach. The length of the strip is intended to provide flexibility for mounting to varied rack arrangements. In my case, the bracket is bolted around the factory roof rack and tent. Four brackets are used.

The concept still seemed plausible. Total tent weight was calculated to be under the legal limit of 165 lbs, the material requirements were reasonable and on hand, and crucially, it looked like a lot of fun to build and use.

With a design in hand, I began fabrication.




Facilities for fabrication were provided by the wonderful Portland maker-space, ADX, and included well-equipped wood and metal shops. For my purposes, the primary tools were:

  • a track saw for cutting panels,
  • a metal saw for cutting structural members,
  • a TIG welder,
  • various sanding and finishing tools.

Many of the design decisions were driven by the available resources and my skill. A professional amateur craftsperson, I have developed reasonable proficiency with wood and metal working. My welding expertise, gained through brief instruction and significant brute-force learning, is limited. The frame really should have been made out of aluminum, but I'm better with steel. The next version will be aluminum.

Wood panels were cut from 1/8" bamboo plywood, chosen for its rigidity and high strength/weight ratio. It is at this stage that the simplified panel design shows its value. Straight lines, cut with a panel saw, make for easier fitting in the frame.

With all components cut, the process of finishing and assembly began. This is where the boat aspects of the design became most apparent. The auto tent would have to be well sealed from the elements and stable in the punishing environment of long distance driving. Water, UV radiation, vibration, and temperature extremes would all be working to break apart the structure. For this reason, the wood panels were well sealed, using several layers of marine grade epoxy, followed by several more layers of a UV-blocking urethane. This process is similar to that used for wooden boats, and should be effective at minimizing damage from moisture or sun.

The steel frame was given a powder coat finish in order to provide environmental protection. Matte textured white seemed the natural choice when matched with the bamboo. This step would likely have been skipped if aluminum were used instead of steel.

Assembly began with securing the panels into the frame. Note the tabs spaced around each section. These were welded to the frame in order to provide secure attachment points for the panels. Mechanical fasteners would have been preferred, and provide more confidence that the panels wouldn't pop loose on the highway. But in the interest of clean design, I opted for simply a marine grade adhesive. Hence the excessive clamping.

Once the panels were securely attached to the frames, all seams were sealed using first, thickened marine epoxy, and then a UV-resistant silicone sealant. This step was tedious and problematic. The design left so much exposed seam, that even the most exhaustive process might be vulnerable to future leaking. The extreme environment of vibration, heat, wind and water is just given too much opportunity. Future designs address this issue, but we'll hope that it holds up.

At this stage, I had two completed halves of the shell and was ready for installation of hardware. Most important was the hinge. At the leading edge of the tent, it would experience significant wear and need to provide stable support when opening the tent. The piano hinge would distribute the weight of the top, open smoothly, tolerate vibration, and flexibly resist misalignment.

The second type of hardware were the draw latches. These would, crucially, hold the tent closed while driving and maintain a seal against environmental intrusion. Though difficult to see in these pictures, a length of weather stripping was attached to the upper lid to complete this seal.

The final primary hardware were two gas spring. They would provide assistance when opening the tent and hold the top open when in use. Gas springs are rated by extension force, or in this case, how much weight they can lift. The selection of proper springs took a bit of research and a return to my high school physics knowledge. While the weight of the top was known, the actual force required was a factor of some interesting geometry. The distance of the spring from the center of gravity and the angle of the mounting significantly influenced the opening force. It was important to carefully balance this force so that the tent would open in a controlled manner and remain open.

The rigid tent frame was complete. The next stage was only possible to the generous support of my friend, an expert upholsterer. 




The auto tent was designed to be enclosed by a fabric canopy. Thus the hard shell frame could expand into a larger living space. While not exposed to the same punishment as the frame, it should still be resistant to weather. An outdoor Sunbrella material was chosen for an appropriate combination of environmental protection, rigidity and weight. The screen material provided air flow and visibility, without completely compromising protection.

The canopy is permanently attached to the inside of the upper frame. It is attached with snaps to the outside of the lower frame. This prevents water from flowing into the tent. When closing the tent for travel, the snaps are released and the canopy is pushed into the tent. In this way, it is safely stored and ready for reassembly.




At this point, the auto tent met my needs handily. It provided a comfortable shelter. But the opportunity to embellish was too great. Interior lighting would be nice, and ventilation would be helpful on hot days. A micro-controller would coordinate functions nicely. Feature creep began:

  • 12v power provided by auxiliary car battery
  • RGB+W LED strip lighting
  • Ventilation fans
  • Temperature sensors, inside & out
  • Ambient light sensor
  • GPS

Several actions were then possible:

  • Opening the tent brings the controller out of sleep mode, flushes stale air with vent fans and turns on bright illumination to assist in setting camp.
  • A small OLED display on the control unit provides status, temperature and location.
  • A button near the entrance sets the interior illumination to match ambient light. A dim red light is provided at night.
  • Vent fans trigger when the tent interior becomes overheated and cooler air is present outside.
  • A wake up alarm gently brings up interior illumination to match the sunrise.
  • Blue-white light alerts of moon-rise. Investigating the possibility of alerting on other astronomical events. Tricky without a large db.
  • Upon closing the tent, the controller shuts off all functions and enters sleep mode.



Field testing

The auto tent was completed shortly before I was scheduled to begin a month-long road trip. Beginning in Portland, I would make my way across Oregon, Idaho, Montana, Wyoming, Utah and Colorado.