This article introduces a new long term initiative at Backpacking Light: to analyze existing standards of design, manufacturing, and materials of ultralight shelters and relate these standards to observed performance in response to adverse environmental conditions.
Shelters of interest to the ultralight backpacking community may be characterized by one or more of the following attributes:
- The use of lightweight fabrics to save weight;
- The use of construction techniques that are less reinforced to save weight;
- The use of construction techniques that are less sophisticated due to the lack of availability of commercial equipment, the desire to minimize labor costs, or design and/or manufacturing inexperience.
- The use of minimal structure (e.g., pole supports) to save weight and increase simplicity.
The combination of these factors results in a product market that is challenging to analyze, because of the wide variability in materials, style, design, and manufacturing quality. In addition, because most of these products are sold direct via the websites of cottage manufacturers, the consumer doesn’t have the ability to carefully inspect the products prior to purchase.
Finally, as the trend towards ultralight backpacking continues to expand, users are either trying to extend their ultralight shelters into “shoulder seasons” and even winter, or they desire to add another ultralight shelter to their inventory to handle stormy conditions that may challenge a shelter’s design. The two primary storm conditions we are interested in include snow loading and wind resistance.
Thus, we embark on a new journey to investigate the storm resistance of ultralight shelters.
One of our wind testing locations in Montana – the Yellowstone River corridor near Big Timber. Winds reliably blow in the 15-25 mph (24-40 kph) range (gusting to 34-40 mph / 55-64 kph) here, with frequent periods of steady winds at 30-40 mph (48-64 kph) (gusting to 50-60 mph / 80-97 kph). We’ve clocked wind gusts here in excess of 75 mph (121 kph).
Modes of Failure
When we talk about either snow or wind loading, we consider two modes of failure:
Non-catastrophic failure occurs when the load (from either snow or wind) alters the shape of the shelter and thus compromises various other shelter properties (warmth, condensation resistance, a change in interior volume/livability, and occupant distress).
This type of loading can be either dynamic (as in the case of a wind gust), or static (as in the case of accumulating snowfall during a blizzard). Non-catastrophic failure is characterized by the simple criteria that upon removal of the load, shelter structure and performance properties are fully restored. Some shelters, even those with poles, are designed to collapse entirely in extremely high wind gusts. An example of this is the common “wedge-type” mountaineering tent (example: Integral Designs MK1). I’ve experienced gusts up to 80 miles an hour (129 kph) that have completely flattened this tent while I was in it, and it popped right back up. It’s disconcerting to experience this, but it’s nice to know that it will pop back up when the gust is over! I do wonder how many times a tent can be subjected to non-catastrophic stresses before proceeding to catastrophic modes of failure.
Catastrophic failure occurs when the load (from either snow or wind) transfers high forces to small areas, resulting in the creation of extremely high stresses that cause the irreversible failure of seams (ripping through fabrics or thread breakage caused by tensile stress), fabric panels (ripping caused by tensile stress), or poles (breakage caused by bending stress).
Catastrophic load failure always results from one primary factor: large, unsupported fabric panels that transfer high forces to weak areas (seams, fabric areas near seams, and poles) resulting from some combination of inadequately supported fabric panels and/or insufficiently strong poles.
The ability for a shelter to resist failure in response to snow or wind loading depends primarily upon how the design meets the following three objectives:
- Minimize loading;
- Equalize load distribution;
- Create strength in high stress areas.
The purpose of #1 is to minimize non-catastrophic failure, and the purpose of #2 and #3 is to minimize catastrophic failure.
Let’s see how to apply these principles for both wind and snow loading.
- Minimize loading: An effective winter shelter will shed some (or all) of the snow as it accumulates, to minimize loading. This ability is enhanced by using (a) a combination of “slick” fabrics and steep walls that don’t promote the adsorption of snow crystals as they fall onto the fabric surface, and instead, slide off before a high load is able to accumulate, and (b) steep walls and structural resistance that prevent snow accumulating at the bottom sides of the shelter from exerting inward pressure onto shelter walls that reduce interior livability, and alter its structural shape in a way that compromises its aerodynamics (wind loading) or ability to support further snow loading.
- Equalize load distribution: Create resistance to downward force (i.e., exerted by gravity) from snow accumulation on the surface of the shelter, through tensile-facilitated structure (structural tension elements in the fabric panels, such as “ridgeline” type seams between apex poles and stake-out points) and semi-rigid pole structure.
- Create strength in high stress areas: Use structural components (poles) that are strong enough to resist high snow loads.
Even light snowfall can significantly alter the performance of an ultralight shelter. One of the more common impacts of snow loading is the accumulation of snow at the base of the shelter, forcing fabric panels inward. This reduces interior volume, increases tension forces in seams, guylines, and stake-out points, and compromises the shelter’s aerodynamics, thus reducing its wind resistance. This photo shows the results of 1 inch (2.5 cm) of snowfall during a six-hour period on a five-sided, symmetrical, single-pole pentamid.
- Minimize loading: “Spill” wind across an aerodynamic design that minimizes overall deflection in the shape of the structure via structural (pole) support and high fabric panel tension.
- Equalize the distribution of high loads: Distribute forces generated by high wind loads to multiple points of stress, through a combination of using small fabric panels (perhaps in combination with structural pole elements), uniform distribution of stress across fabric panels, and symmetrical design elements.
- Create strength in high-stress areas: Engineer resistance to high stress concentrations (i.e., guyline and stake tie-out points) using the appropriate combinations of reinforcing (or higher strength) materials and effective manufacturing (sewing and bonding) methods.
Failure Limitation Hypothesis: It’s All About the Fabric
A shelter can fail in response to storm (snow or wind) loads for a variety of reasons. I’ll discuss each in turn, in the context of how much distress they might cause the user.
When shelter stakes pull out, this is usually not caused by a shelter defect, and can generally be attributed to user error: the user has failed to match the right type or length of stake to the terrain or conditions, or has improperly inserted a stake (e.g., at an angle too far from perpendicular relative to the direction of pull). Generally, moderately stormy conditions (snowfall equivalents of several inches through a night, or wind loads induced by 30-40 mph / 48-64 kph winds) can transfer up to about 40 pounds (18 kg) of tension force to guylines and stake-out points for most solo shelters. If a stake pulls out of the ground with only 40 pounds (18 kg) of force, and you are expecting moderate storms, then you might reconsider your choice of stake. For more information, refer to Will Rietveld’s research on stake failure forces.
Dealing with stake failure in the field is thought by most to be merely “annoying,” but in reality has consequences that can be more than annoying when conditions are more serious (cold and wet), and especially, in the presence of high winds. When a stake fails, wind loads can further cause complete collapse of the shelter, and instigate a chain of events that might include the failure of subsequent stake-out points, the breakage of poles, and the ripping of fabric or seams. These obviously are more serious consequences than simply getting out of your shelter to repitch it. However, if you’ve ever had to repitch your shelter (especially a floorless one) in the middle of the night during a blizzard or thunderstorm, all while keeping your gear from blowing away and getting wet, while in your underwear and socks, then you might also know the feeling of reminding yourself to bring more robust stakes on your next trip. Stake failure may result from a design limitation (due to poor design transferring higher-than-necessary forces to a particular stake), but it should not generally be considered a manufacturing defect.
Whether you are using trekking poles or tent poles, aluminum or carbon, pole breakage is a common mode of shelter failure. It most often occurs in response to moderate to heavy snow loading, and very high wind loading. Pole breakage is catastrophic in and of itself and can further cause a chain of events that results in other modes of failure (e.g., fabric punctures from broken pole tips). While tent pole repair is possible in the field, it requires extra gear, time, and, if in the middle of a storm while you are huddled under your flapping, collapsed shelter, inconvenience and drama. Pole breakage is usually considered a manufacturing or design limitation (not necessarily a defect, e.g., in pole material, but it can be). Further, pole breakage may have resulted from any combination of inadequately strong poles or failure of the design to distribute stresses appropriately. Both of these responsibilities, of course, would fall on the manufacturer as a defect in the shelter’s design.
Seam failure occurs most commonly where high stresses are concentrated, i.e., guyline tie- and stake-out points. They occur predominantly from fatigue due to repetitive wind stress, ripping in response to high wind gusts, and tripping over guyline cords. Dealing with these types of seam failures in the field requires a sewing kit, time, and more skill than what is required when dealing with pole breakage. Seam failure should always be considered a manufacturing defect. Seams should always be stronger than the fabrics used to construct them. This includes the failure of panel-to-panel seams, guyline and stake attachment points, as well as the failure of a seam resulting from accidentally tripping over a guyline.
Fabric Panel Failure
Fabric panels fail in one of two ways. The first is that of excess tension stress being built up in the panel, thus overwhelming the strength of the fabric. This rarely occurs for two reasons: (a) fabric used in most ultralight shelters is tremendously strong and (b) shelters tend to fail in other areas (e.g., poles, seams) before failing in the fabric panels. This type of failure may result from a design limitation, but it is never a manufacturing defect (it may result from a fabric defect, but this is rare).
The second mode of failure of fabric panels occurs when nearby seams weaken neighboring fabric structure. So, the fabric tends to rip near the seam. This often occurs when needle holes are too big, inappropriate stitch types and dimensions (or thread types and sizes) have been used, and poor design concentrates too much stress adjacent to a seam. This type of failure is always a manufacturing defect. Fabric panel failures are not fun to repair in the field, especially during a storm. However, the difference between repairing them years ago and repairing them today is the ready availability of very high strength tape bonds that can easily patch a fabric. The addition of supplemental glues and perhaps, sewing, for stronger fabrics, can create a fabric tear patch area that is stronger than the base fabric itself.
Based on this discussion, what follows represents my own hypothesis regarding how failure should be engineered (accommodated) in any ultralight shelter.
It’s an extraordinarily simple hypothesis. Are you ready? Here it is.
Fabric ripping in response to excessive tensile stress accumulation should be the point of failure.
Here’s why I believe this.
First, I believe that failure that results from a (sewn or bonded) seam occurs for one reason only: the seam was not constructed properly (i.e., the seam construction technique is inappropriately weak considering the strength of the fabric). Seam failure is inexcusable for any type of soft good, and more so with soft goods made with ultralight fabrics, including shelters.
There are two ways to look at this.
- If you are using a comparatively heavy fabric to construct a shelter, and the seam is the point of failure, then consider that the shelter could be lightened up by using a lighter fabric without affecting the functionality or mode of failure of the shelter. In other words, your heavy fabric has resulted in a shelter that might be overbuilt.
- If you are using a comparatively light fabric, and the seam is the point of failure, then consider that your seam manufacturing process is inadequately matched to the fabric. Consequently, you might change the seam construction technique, or add reinforcing (or heavier) material to seams where stress accumulates excessively, for very little weight, and add significant overall strength to your shelter.
A seam sewn in Cuben Fiber at a guyline tie-out point, after being subjected to dynamic wind loading in the range of 15 to 25 mph ( 24 to 40 kph) for 12 hours. This photograph was taken when the guyline tension was approximately 15 pounds (6.8 kg).
Second, I believe that failure that results from the breakage of a structural element (e.g., a pole) occurs for two reasons: the structural element used was too weak and not matched appropriately to the shelter design and materials, and/or the shelter design inadequately provided for the even distribution of stress to the overall pole structure.
There are also two ways to look at pole vs. fabric failures.
- If you are using a comparatively heavy fabric to construct a shelter, and a pole is the point of failure, then consider that the shelter could be lightened up by using a lighter fabric without affecting the functionality or mode of failure of the shelter. In other words, your heavy fabric has resulted in a shelter that might be overbuilt.
- If you are using a comparatively light fabric, and the pole is the point of failure, then consider that the pole materials, dimensions, or configuration are inadequately matched to the fabric. Consequently, you might change the properties or configuration of the poles, which may cost very little in weight while adding a significant amount of strength to the shelter.
Third, repairing fabric tears in the field by using some combination of bonding (glue), hand sewing, and patching with other fabric, is sometimes a more feasible mechanism of in-situ repair than repairing either structure elements (poles) or rebuilding seams.
The conclusion that can be made from this hypothesis is vitally important:
You must absolutely pay attention to construction techniques and structural design. If manufactured elements (seams or poles) fail before panel fabrics, then the shelter is either heavier than it needs to be (because lighter fabric can be used) or it’s not as resistant to storm loading as it could be (with significant resistance able to be added for very little added weight).
The primary challenge for the shelter designer or manufacturer is not to overbuild the common failure points (seams, poles) just because they are using a strong fabric. Perhaps the ideal shelter is built in such a way that there is an equal probability of failure of any particular part of the shelter.
This tent, designed and built by Roger Caffin, weighs only two pounds, but the combination of three carbon fiber poles, 14 stake-out points, and a tunnel design means that once it’s staked out, even winds as high as 30 mph / 48 kph (broadside) exert less than 5 pounds (2.3 kg) of additional force to any of the side guylines that are perpendicular to the wind direction. In comparison, most tarps and tarp tents result in the transfer of as much as 40 pounds (18 kg) of force per guyline in the direction perpendicular to the wind in response to 30 mph / 48 kph winds. The primary reason for this is that large, unsupported fabric panels on tarps and tarp tents deliver large forces to guylines and concentrate high amounts of stress to guyline attachment points. Therefore, perhaps not intuitively, the guyline attachment points on tarps and tarp tents might need to be more robust with a higher standard of manufacturing than a tent like this one.
Designing for Fabric Panel Tension
When a shelter is designed in a way that its engineered elements are designed not to be the point of failure, then an extremely important design principle can be engaged:
Resistance to storm loading and fabric panel strength is maximized by building a shelter that distributes high levels of tension uniformly across the entire surface area of a its fabric panels.
The reason that fabric panel tension is so important for storm resistance is that tighter panels resist deflection in response to both snow and wind loading, are more aerodynamic, and distribute forces more evenly across seams, corners, and guyline/stake-out points.
While tension can be distributed in a variety of ways (by adding structural elements such as poles or tensile elements such as seams), the best way to induce high levels of tension in fabric panels (assuming sound design, cut, and sew has been employed to avoid sloppy panels) is the good old fashioned way: tight guylines that connect the shelter to stakes that can hold a lot of tension.
For example, tunnel tents achieve their strength by stretching their canopies tight from the head to the toe of the tent. For a typical one- to two-person two- or three-pole tunnel tent, the ability to create a stake-out tension of 30 to 60 pounds (13.6 to 27.2 kg) along its longitudinal axis is critical for creating tension in fabric panels between each hoop pole. This tension has a significant impact on the tent’s ability to shed both wind (lateral loading) and snow (vertical loading). Further, it prevents the poles from rotating out of their resting plane in response to asymmetrical wind loads (a wind load that is not parallel to the tent’s longitudinal axis) and resulting in pole breakage. I have observed this mode of failure on both single pole hoop tents (the Hilleberg Akto) and double pole hoop tents (Stephenson’s 2R) at winds in excess of 60 mph (97 kph).
It turns out that 30 to 60 pounds (13.6 to 27.2 kg) of tension in any direction is somewhat of a “magic bullet” of a number for one- to two-person shelters. Empirically, during my preliminary analysis of wind loading on ultralight shelters, being able to exert enough tension on guylines at zero wind load was a critical prerequisite to maximize the shelter’s performance in the presence of high winds, because less dynamic loading would occur in response to wind loading (since fabric panels were prestressed, and better aerodynamics were induced through tension).
It follows, then, that a significant mode of failure of ultralight shelters, especially those that employ low-stretch fabrics such as Cuben Fiber, is the shelter’s inability to be staked out tightly at high enough loads that induce a good distribution of tension evenly across all fabric panels. It further follows that sloppy design has significant negative impacts. Orienting fabric panels in the proper bias direction and accurately cutting and sewing (or bonding) them goes a long way towards evenly distributing stress across the entire surface area of the fabric panels, and (most important) along their edges and seams.
Common reasons that ultralight shelters cannot be staked out tightly include:
- Weak seams, and thus, fear by the user of inducing such high loads,
- The use of stakes that are too small or otherwise don’t hold appropriately in the ground,
- The lack of additional tie-out points on fabric panels or structural elements (seams or poles) to increase tension across fabric panels and further resist collapse of panels by either snow or wind loading,
- Incorrect fabric bias orientation relative to tension forces,
- Sloppy cut-and-sew techniques that do not evenly distribute stress across the entire surface area of fabric panels.
There is a reason that Cuben Fiber and other ultralight low-stretch fabrics are not used much in tents that are considered “stormworthy” – their low stretch demands low error tolerances in dimensioning and orienting fabric panels. Cuben Fiber is not a forgiving fabric for the casual shelter maker who is used to hiding minor design and manufacturing errors in the elasticity of silnylon.
This shelter is one of my testing standards. It has a symmetrical, five-sided design, is made with stretchy and strong silnylon fabric, is supported by a single center pole, and is manufactured with guyline attachment points that are robust enough to support more than 50 pounds (22.7 kg) of tension – per guyline. As you might expect, it handles high winds exceptionally well when pitched in this configuration.
I’m in the process of testing a variety of ultralight shelter types that employ a wide range of fabrics, manufacturing techniques, and structural design elements. My goals with this testing are:
- Expose substandard manufacturing practices that are inappropriately matched to the panel fabric, other materials, and structural design elements used in the shelter.
- Identify manufacturing techniques and specific products that are capable of significant storm resistance, for the purpose of increasing user and consumer confidence when faced with a decision to take (or buy) a particular shelter for inclement environmental conditions.
- Study how both wind and snow loading induces both non-catastrophic and catastrophic failure in ultralight shelters, and how those failures impact field experiences for ultralight backpackers.
The basic testing framework is based on the following:
- Exposure of shelters to snow and wind loading environments.
- Staking out shelters in a configuration that induces an even distribution of tension in the shelter’s fabric panels.
- Monitoring tension forces at stake-out points (using in line load cells coupled to a data logging system) generated by wind and snow loading in real time and correlating these values to measured wind speed and direction, calculated normal and distributed fabric panel forces, real time video capture, and deflection measurements.
When I was originally proving the concept of a force measurement methodology, I simply used an inline spring scale to measure tension forces in guylines, and wrote the measurements down using good old fashioned pencil and paper, while keeping my second eye on a wind meter (mounted on the tripod). Now, data from both the wind meter and the inline force sensors are transferred to data loggers, so data can be collected from multiple sensors and multiple shelters simultaneously, and monitored in real time on my laptop from the safety and comfort of my truck canopy.
The use of a spring scale doesn’t provide enough of the proper information required to fairly assess tension forces in real time. In addition, a scale such as the one above is too big and heavy – and dampens the true tension force at the stake-out point. Our revised and current experimental system uses aluminum load cells that monitor both force and acceleration (the combination of which can give us information about dynamic loading) at the stake-out point with a resolution as low as 0.1 millisecond (time) and a sensitivity of 0.01N (0.002 lbf). Both force sensors and weather meters then deliver data in real time to a laptop via Bluetooth. This will allow for very precise correlation between measured wind speed, measured deflection (determined via real time videography), and forces transferred to stake-out points. The example graph above shows tension measured in a tarp guyline that was originally staked to a force of about 5.6 pounds (2.5 kg). During this test, a light breeze was blowing (1-2 mph / 1.6-3.2 kph), and you can see the impacts of that breeze on guyline tension by the positive peaks in the graph, that show even a very light breeze can increase guyline tension by 1-3 lbf in this case. (Aside: we are adapting our sensor and data logging system to other projects in 2012 as well, including studies of backpack load stability and the impacts of trekking poles on impact forces transferred by the body during walking).
For the casual ultralight backpacker that needs occasional protection from light wind, rain, and insects, there are plenty of ultralight shelter options available. The choice of an option suitable for most summer conditions in temperate climates will depend primarily on weight, cost, ease of set-up and use, and features related to a user’s livability in the shelter.
However, many ultralight backpackers, at some point during their ultralight evolution, will begin the process of testing the limits of their ultralight gear and style. This includes the use of ultralight shelters during shoulder and winter seasons, above the tree line in mountain environments, and for other scenarios where environmental conditions demand more performance that can only be achieved at very light weights through sound engineering design and the intelligent use of the right materials and manufacturing techniques.
It is my hope that this ongoing research into the storm resistance of ultralight shelters contributes a meaningful body of data, observations, conclusions, and knowledge that will help us explore more hostile environmental conditions with less weight on our backs. In addition, I hope to spur both do-it-yourself enthusiasts and commercial manufacturers to pay close attention to their designs so as to improve their performance beyond casual summer environs. I believe that the market demand for “ultralight” shelters capable of withstanding harsh storm environments is increasing, and that the current product market for stormworthy ultralight shelters is sparse, underserved, and ripe for opportunity.