The industry uses a few different metrics for assessing wind and rain shell performance. CFM (cubic feet per minute), which is a measurement of air permeability, or what volume of air can be pushed through a fabric. MVTR (moisture vapor transmission rate), which is a measurement of how much moisture vapor gas can be pushed through a fabric. HH (hydrostatic head), which is a measurement of how tall a cylinder of liquid water a fabric will hold back before leaking. I think the industry has a pretty much right with regards to which of these metrics they use for which garments. Generally, CFM is used for wind shells that are not waterproof. MVTR is generally used for waterproof breathable membrane fabrics because they usually have little to no actual CFM. They can only transmit moisture vapor gas through micro pores in a membrane. HH is generally used for anything meant to be waterproof. I think HH is fairly straightforward and this article will focus on the nuances between CFM and MVTR.
There is some push these days to re-think the models of which metrics we use for what fabrics, but the line of thinking seems to head in different directions from there. Some would argue that we should see all of this data for all different types of fabrics. This I can agree with. It can’t hurt to have more data as long as we are able to use it proerly. The other direction is the idea that CFM isn’t relevant and MVTR is the holy grail. As usual, it just isn’t that simple. For a fabric that has limited CFM, MVTR is really the best option because CFM isn’t high enough to contribute much. Once we get above a certain threshold ( I don’t really know where this point is) CFM really becomes the most important metric. That air permeability we are measuring is also the primary means of transmitting moisture vapor, above a point. Moisture vapor molecules are smaller than the other air molecules, so it is possible to have a fabric with holes small enough for moisture vapor gas to escape, but air to not….and of course, to keep the even larger water molecules out. However, when we get up to a point where significant volumes of air can move through the fabric, it is inevitably a mixure of air and the much smaller moisture vapor. The air movement is the primary driver of moisture movement.
The REALLY big aspect that is overlooked about CFM, as a metric, is what actually happens in real life. Real life is not a static lab test. When we are wearing a high CFM wind shell out on the trail, air is not only moving out of it from our heat and moisture build up, but it is also moving in from the outside. In use, there is very little resistance against air movement through a fabric, into a garment. All the edges of a garment, like the hip, cuff, neck, etc, are not sealed. They are actually very leaky. Therefore the only resistance to air movement into a garment is the CFM of the fabric itself. So if air can move out of the garment at X, it can also move into the garment at X. Since most of the time, either the wearer is moving, or the air is moving, there is constant air transfer going on. This air transfer not only moves moisture, but also moves heat. THIS is key. If we take a waterproof breathable membrane garment with little to no CFM and compare it with a wind shell garment with CFM that happens to have the same MVTR in a lab, the wind shell with CFM will almost always come out the more comfortable and drier garment. CFM regulates temperature. A high MVTR WPB membrane can move a certain amount of moisture, but if it doesn’t move any air, then the heat will continue to build up more and more. As heat builds up, the body produces more and more moisture, which inevitably overwhelms the MVTR of the membrane. A high CFM wind shell with the same MVTR in a lab will be more comfortable and more dry because air will move in and push out the hot air and moisture. This cools the body and reduces the amount of moisture produced.
Of course, the reality of a WPB membrane fabric having a similar MVTR to any high CFM wind shell is very unlikely. There is some recent single source lab data suggesting they can match up but this generally contradicts real world performance. We all know very well that bivy sacks built from WPB membrane fabrics commonly have internal condensation issues. It is really just an accepted drawback of this type of shelter. However, the UL backpacking community has really embraced the use of Bivys built from high CFM fabrics like Argon 67 (50cfm) because they can be used in most conditions without the build up of internal condensation. This is because the high CFM moves more air and moisture than the membrane fabrics, even in stagnant air. These results really align with general common sense. Microporous WPB membranes can be transmit vapor while being waterproof simply because the pores are small enough to keep water out but large enough to allow vapor through. If the pores in the Argon 67 fabric are so much larger, as to let 50 CFM of the much bigger air molecules through, then we can easily predict that it will also be letting an increased number of way, way smaller vapor molecules through.
I think it is important to use lab data properly. Real world performance is the the end result that inevitably factors in all the variables. I think the purpose of lab testing is to help explain why we see the real world results that we do. We are getting into trouble when lab results contradict the real world results that we clearly see. One of the benefits of lab testing is that you can control variables, but this is also, in a way, one of its drawbacks. The real world performance of something is a combination of many different variables and conditions. It is very easy to isolate a testing procedure in a lab down to a point where you see results that contradict, or don’t apply at all to the real world use case at all.
It is also very important that any conclusions derived from lab testing comes from multiple sources in order to prove repeatability. Single source lab results that contradict other data should be set aside as an anomaly until there is something to back it up. Single source data that contradicts other data AND real life performance is almost certainly flawed.