A simple mathematical model is developed to identify (and avoid) conditions leading to condensation within a vapor barrier clothing system in the absence of overheating.
The potential benefits provided by wearing vapor barrier (VB) clothing are well documented. Also well documented are the potential problems. Although my feet end up looking pickled, I’m very fond of VB socks for winter hiking to keep my boots and insulating socks dry. VB handwear can offer similar advantages and ease of use. I’ve also had good luck using VB pants as sleepwear to reduce accumulation of moisture in the down of my sleeping bag. (A half-length VB liner would work similarly, but if I have brought along insulating pants, I want the VB under them, so I use VB pants.)
Unfortunately, I have not had such great luck with VB shirts. Because my upper body usually has more layers of clothing than my lower body and stripping down to my base layer to put on a VB shirt before I go to bed is not something I enjoy, the obvious solution is to wear a VB shirt all the time. And I’ve tried it. And I stayed warm. And my outer layers stayed dry. And when things went well, the feeling of perpetually standing in a tropical greenhouse was tolerable. But when things didn’t go well, which was not infrequent, I became a walking distillation column. (But I didn’t collect and drink the condensate as described in the novel “Dune”.) Because the VB shirt kept this water from soaking my outer layers, I was never in danger of hypothermia, but waiting for the water to diffuse back into my skin was never pleasant.
Much of what I’ve read concerning VB clothing points out that if the user allows themselves to overheat, which is not difficult in VB clothing because evaporative cooling is no longer taking place, then profuse sweating is to be expected and could have been avoided through better temperature management. This is true, but it’s not my primary concern because we know how to avoid it.
In what follows I will instead address the issue of condensation within VB clothing in the absence of overheating. This work was motivated by a desire to identify conditions under which condensation is likely to occur so that I might someday arrive at a workable solution without so much error in my trial-and-error experiments. I tried to approach the problem as a lazy engineer might, using reasonable assumptions wherever possible to keep things simple without losing touch with reality.
The skin surface is at temperature Ts. Outside the skin is an insulating layer of thickness L1 and thermal conductivity K1. This first layer accounts for insulation due to clothing inside the VB (if present), stagnant (air) boundary layers, and VB lining (if present). Moisture within the air is assumed to be in equilibrium with moisture in the skin. The VB encloses and completely seals this first layer. The VB is assumed to be at temperature Tvb and to have zero thickness. Outside the VB is an insulating layer of thickness L2 and thermal conductivity K2 to account for clothing worn over the VB. The outer surface of this second insulating layer is in contact with outside air at temperature To. Boundary layer heat transfer resistance between the outer insulating layer and the outside air is ignored, which would be valid on a windy day.
Assuming the layers described above are flat (a pretty good assumption for the torso, not so good for a finger) the steady-state heat flux from the body to the environment is
Q = (Ts – Tvb) * K1 / L1 = (Tvb – To) * K2 / L2.
Setting Tvbmin equal to the VB temperature at which condensation will take place on the VB surface, the maximum rate of heat transfer that is possible without condensation is
Qmax = dTvap * K1 / L1 = (dTtot – dTvap) * K2 / L2min
where L2min is the minimum outer layer thickness necessary to maintain Tvbmin and
dTvap = Ts – Tvbmin,
dTtot = Ts – To.
If it is possible to use VB clothing without condensation, the variable dTvap must be greater than zero. This will be the case if the water vapor pressure in equilibrium with the skin at Ts equals the vapor pressure of condensing water at temperature Tvb with Ts greater than Tvb. I have not yet found a reference that gives vapor pressure above skin as a function of skin temperature, but personal experience suggests to me that the value of dTvap is in the range of 10 to 20 degrees F. (Skin permeability to moisture reportedly increases with body temperature, so actual results my vary, but this guess means I would expect to eventually get condensation inside a well-sealed raincoat on a cloudy, breezy 70-80 degree F day.)
Solving for L2min we get
L2min = (dTtot / dTvap – 1) * L1 * K2 / K1.
To see if our equations will tell us something useful, let’s look at a few examples. To keep things simple, assume K1 and K2 are equal, which is probably a pretty decent assumption for most insulating materials. For the sake of argument, set dTvap equal to 15 degrees F and assume a skin temperature Ts of 95 degrees F. We will consider two different outside temperatures To (20 and -10 degrees F) and two different inner layer thicknesses L1 (5 mm, which is probably realistic for a thin base layer under a close-fitting VB when one considers stagnant boundary layers, and 2 mm, which represents a very snug VB). The results are
L1 = 5mm, To = 20 L2min = 20mm (1 inch of total loft!)
L1 = 5mm, To = -10 L2min = 30mm
L1 = 2mm, To = 20 L2min = 8mm
L1 = 2mm, To = -10 L2min = 12 mm
Several conclusions can be drawn. Foremost, if anything more than moderate activity is required, it’s going to be tough to stay cool and dry except with a very snug VB at very low temperatures. This is not surprising to anyone who has used VB clothing. More potentially useful (to those who appreciate irony) is the realization that shedding outer layers to avoid overheating might actually make you more wet. There is a point at which condensation will occur even without sweating, and in my experience it is a very sharp transition with very dramatic results.
It would be possible at this point to also calculate Qmax for various scenarios and compare the results with Q values for a given skin surface area and activity level, but I don’t think it would be much fun to read about without colorful graphs, so I’ll skip it.
Now that we’ve shown how problematic VB clothing can be, can we conclude anything potentially useful? Possibly. Here are some thoughts:
• I don’t expect to be anywhere cold enough to use VB pants while moving.
• A very snug VB shirt with a highly heat-conductive inner layer of minimal thickness will keep me cool and dry over a greater range of conditions without venting, but…
• A VB shirt probably needs to be extensively ventable if I’m going to wear it all the time.
• A VB shirt with an appreciable amount of built-in outer insulation might reduce condensation, but avoiding overheating might become even more of a challenge.
• I should maximize heat loss elsewhere in order to minimize the need for heat loss through a VB shirt. My legs are natural candidates for heat disposal, so I should try to wear pants that are comfortable only as long as I keep moving. For rest stops, I can wrap my bivy or ground sheet around my waist as a “wind kilt” to reduce heat loss.
• A very thin hydrophilic lining on the inside surface of a VB shirt might lower the temperature necessary for condensation (a good thing) by increasing the surface area (and so average energy) of the resulting liquid water phase. Re-evaporation of liquid water might also be enhanced if a soaking event should occur. It would be nice if this lining were resistant to microbial growth.
• I should try to keep the inside of the VB as clean as possible, because water vapor will generally condense more readily into dirty (sweaty) water.
• Breathable garments are more user-friendly (at least until my sleeping bag goes flat from nightly condensation). Millions of years of evolution in Africa means that covering substantial areas of my body with a VB layer requires careful thought.
• The same analysis applies to VB sleeping bag liners, but they are less problematic because they are usually surrounded by ample insulation and we are not trying to shed excess heat.
Incidentally, most of my experience has been with using sil-nylon for VB clothing. The Stephensons catalog was quite an eye-opener when I received it many years ago, but I have no experience with their product. I’ve heard that RBH is working on a VB active shirt, but I don’t know much about it. I’m assuming that somebody somewhere has done an analysis similar to that above, but I’ve never seen it, so I thought someone might find it useful.