By the Numbers: Is Synthetic Insulation Warm When Wet?

[Stephen Seeber]

Companion forum thread to: By the Numbers: Is Synthetic Insulation Warm When Wet?

We throw some quantitative testing at an oft-repeated synthetic insulation claim to measure the thermal performance of wet insulation.

[Jerry Adams]

Good article, difficult subject.

You said:

“If we try to measure the thermal resistance of a piece of wet insulation, we need to apply heat. Once the heat is applied, the water in the test sample will start to evaporate. As evaporation occurs, some of the heat that is being added will be consumed when turning liquid water into vapor as evaporation occurs. This extra amount of heat has nothing to do with the R-value measurement but cannot be removed from the measured heat input. As the test continues, and evaporation continues, the water contained in the test sample will diminish, the effective thermal resistance of the sample will steadily increase. The amount of heat delivered for the test to maintain constant top and bottom temperatures will then decrease. Throughout this process, the amount of heat energy input to the wet test fabric will constantly change.”

I think you’ve identified what’s happening

When I’ve got my insulation wet, measured the weight, calculated amount of water, used the thermal conductivity of air and water, there wasn’t enough water to significantly increase the thermal conductivity of the garment.

The problem is that, initially, about half the heat released by the body is consumed evaporating the water in the garment.  If you apply constant amount of heat, similar to a human body, the temperature difference across the garment will be about half initially.  As the water in the garment is gradually evaporated, the temperature difference across the garment will increase.

So, the experiment similar to reality would be to get the garment wet and then let the excess drip off.  Then apply the same amount of Watts per square meter as a human body.  Measure temperature difference across garment.

Like I said, in my testing, the temperature difference is initially about half.  After a few hours it dries and the temperature difference doubles back to normal.

This would be the same as if the R value was initially halved, but gradually increased to normal.

 

[Rebecca S]

Hello, Stephen,

What a fascinating article. So useful!

I also learned a lot from your previous article about fleece, partly because it confirmed my experience, which is that my 100-weight fleece jacket is as warm (under a wind shell) as my 200-weight fleece jacket.

Regarding your latest findings, do you have any information regarding the performance of merino wool, when wet, relative to synthetic fleece or synthetic insulations? I’m assuming the sheep have it “dialed in” and that the warmth is superior. But I’d like to know for sure.

Could you tell us what the research shows about the relative warmth of garment fabrics that are made of merino wool, Polartec fleece and some of these synthetic insulations? Dry and wet?

It would help a lot of us figure out whether it’s worth carrying a wool garment that’s heavier than the synthetics but may perform better when damp or wet.

Thanks, again, and kudos.

Regards,

Rebecca

[Stephen Seeber]

Hi Jerry:  You are right.  The latent heat of vaporization is the energy required to turn liquid water into water vapor.  It is  a big energy suck, especially when combined with increased conductivity of your wet insulation.   There is a lot more to explore on this topic, and I hope to be better able to control moisture levels in the wet insulation for the next look at this topic.

Hi Rebecca:  I should have some wool tests coming up soon and intend to compare the performance of at least two types of wool and fleece.  I think a lot of us have wondered whether wool or polyester makes a better base layer for moisture handling.  I hope to shed some light on that issue.

[Michael B]

Maybe I am a bit lazy in thinking about this, also, since I am not a premium member, I don’t know if you addressed this, but wouldn’t a simpler experiment work just as well at understanding how wet an insulation is can affect its insulating properties? Hear me out and tell me if I am off base. You have these nice kettles, which can heat up a given quantity of water to a very precise temp. You have these nice insulation covers you used to cover the pots. Rather than heating up the kettles and keeping them at the same temp throughout the experiment, could you have made sure each test run had the same qty of water and started at the same temps (both the water and the ambient), and then varied the saturation level of the insulated pillow, and then just measured the rate of temperature drop of the water to see how the different levels of insulation dampness affected the temp drop rate? That seems like it would give you good data as well, and would not require infrared spectroscopy (although less fun). I’d enjoy hearing your take on that type of measurement, since I am not well acquainted with setting up these types of lab experiments.

[Stephen Seeber]

Hi Michael:  I did discuss some of the issues involved in measuring thermal resistance of wet insulation.  Yours is an interesting thought.  I suppose, mathematically, it might be possible.  As a practical matter, each kettle is filled with 9 gallons of water.  The kettles are jacketed in closed cell insulation.  They sit on 1″ extruded polystyrene foam.  As a result of all this, they cool very, very slowly.  The rate of cooling is function of ambient temperature and the kettle insulation, as well as the large mass of all the water.  The exposed surface of the insulation samples on top of the kettle are a small portion of the total energy losing surface area.    So, there are number of ways heat is lost from the kettles, so accurately measuring heat loss from a specific portion would not be simple.  Whenever I have thought about measuring input watts into the kettles to calculate “Q” or the energy loss through the test sample, I have always been stymied by solving how to ascribe heat loss to each type of surface on the kettles.   So, your method might be feasible, but I fear it would add complexity and uncertainty to a  process that is already complex enough but at least the analytical issues  and their solutions are well understood.   If you want a better understanding of how I make measurements on the kettles, you can read this article: https://www.dropbox.com/s/h82g9szn5apnyml/How%20I%20Measure%20Thermal%20Resistance%20%20of%20Clothing%20Samples.pdf

 

 

[Chris]

Nice article Stephen! I’m enjoying this series.

Two quick questions:

1. In the primaloft example, the wet/right pillow has batting seams but it appears the left does not. Is that correct? If so, the seams appear to have a significant impact on performance. E.g., in Figure 3, the center of each square between the batting seams looks comparable to the dry pillow, while the areas around the seams show significant heat leakage. Do you have any further details on that?

2. I always took the “warm when wet” slogan as more of a generalization compared to down as opposed to a literal claim.  I also thought that it related to the ability of the insulation to dry out when under active use and restore loft/insulation, as opposed to just the impact on static insulation value alone. The results in table 3 seem consistent with that. I see you are planning additional tests on down but am wondering if you are planning/able to look into the drying-out aspect of their performance relative to each other?

[Michael B]

Stephen, could you, as a base number, use a highly insulated top and measure heat loss from the system as a reference? I am sure if you made a lid from several layers of home insulation foam (R5 x however many layers you thought necessary), you could get a reasonably accurate assessment of the base loss from the rest of your system.

But that being said, was the point of the experiment in the article to understand the absolute insulation value of wet insulation, or was it only to examine the difference? In the former case, I agree with you that an accurate baseline heat loss figure is required for an accurate assessment, but in the latter case, only the delta is important, and a baseline value is not necessary. Do you agree?

As an anecdotal point, I hiked this weekend in a polar-tec powerdry hoodie, and can tell you that it neither dried quickly, nor did it keep me warm when wet :).

[Stephen Seeber]

Hi Chris:

Very astute observation!  There was a miscommunication between the person who volunteered to make the pillows and myself due to the use of the vapor barrier on the left pillow, which cannot be stitched through.  Unfortunate but, since he was donating his time and materials, I felt I would make the best of what I had available.  When I made the R value measurements, I took the readings on the right pillow used for R value calculation on the surfaces within the quilt stitching, so the high temp areas are excluded from the R value calculation.   So, for the present article, the measured values for both pillows were made with areas of similar loft.  Your statement about the qualitative comparison in Figure 3 is correct.  Within the quilted areas, the surface temperatures of the two pillows are very close, so the heat loss is going to be similar.  Without a doubt, areas containing quilt stitching will increase heat loss in those areas.  This issue was discussed in the first article in this series.  Of course, this issue does not exist for the Climashield pillows, which do not require quilting.

“Warm when wet” should not be a slogan.  It should be proved with studies by manufacturers wishing to sell their products.  Has anyone seen a manufacturer’s published study?  It is easier to make this claim than to prove it.  Consumers should not have to decide what it means and whether they believe it. Using the thermal imager, a comparison of drying time can be done.  Here is one way to do this: Take a down jacket and synthetic jacket with similar R-value performance.  Measure jacket MVTR to ensure they are similar and one is not doomed to failure as a result of face or liner fabric selection.   Weigh the jackets.  Run each through a rinse and spin cycle (both jackets together).  Weigh the jackets to determine water retention. Place each jacket on a kettle at 120F.  Start the thermal imager and produce a time lapse video.  The first jacket to reach ambient temperature is the winner.  It would make for an interesting article.    Of course, there may be pitfalls for such an experiment that would be revealed during the testing process.  Fundamentally, for such an experiment to be useful, we need to understand the characteristics of the materials that impact how easily entrained moisture can be vaporized and how readily that moisture can travel out of the garment.

[Stephen Seeber]

Hi Michael:

I think your first point references your prior question.  If you read the article for which I provided a link, you can read what I have done to demonstrate the accuracy of the infrared/kettle method as well as the guarded hot plate method.  The way the infrared/kettle test works, the losses through the kettle surfaces, other than the test sample, are not relevant.    While what you suggest may be possible, it is unclear to me what advantage it would provide.  Now, for someone trying to devise a test method from scratch, perhaps it is worth pursuing.  Once again, I invite you to read the article on how I do the testing and I will try to answer questions you may have.

In this article, I attempted to provide both absolute and relative measurements as well as qualitative data.  All are useful and the reader can use whatever set of data is most intuitive.

Drying your insulation while wearing it can be a dubious proposition.  Turning a pound of water into vapor requires approximately 1000 BTUs.  That is 1000 BTUs that may be taken exclusively from your body.  Perhaps feasible when you are generating lots of heat.  Less so once you stop moving.  I appreciate the anecdote of warmth when wet in action.

[Chris]

“Warm when wet” should not be a slogan.  It should be proved with studies by manufacturers wishing to sell their products.  Has anyone seen a manufacturer’s published study?  It is easier to make this claim than to prove it.  Consumers should not have to decide what it means and whether they believe it.

I agree! I’m glad you’re digging into these types of claims. I don’t buy down for ethical reasons, so the potential advantages/disadvantages of down vs. synthetic are rather collateral to me. But I enjoy seeing these types of generalizations put to the test.

I wonder if the drying time test could also be done with custom pillows similar to what you did here with synthetics – i.e., using the same face fabrics for each and only the fill differs? It would still be a little unrealistic since in practice the fabric and construction requirements for down are generally a bit different than synthetics. But it would at least isolate that variable.

[Josh J]

so the real question in my mind is if synthetic looses it’s insulating properties just like down when it’s wet which one dries out faster and maintains it’s loft to regain it’s insulating properties.

[Craig B]

Hi Stephen,

Nice experiment.  I read your first installment, and didn’t notice you mention the type of thermal camera you are using.  Is it by chance a Flir A65?  Those seem to be pretty popular, and I have used them myself at my job.  If you (or the audience) are not already aware, a couple things are worth a footnote about using thermal cameras for thermography (actual temperature measurement).

First, if the camera is not a ‘cooled’ camera (the focal plane array cooled to cryogenic temperatures, generally with a sterling cooler), it’s likely the digital values that it reads out will drift a fair amount, resulting in drifty temperature measurements.  In the case of a Flir A65, it will drift as much as +/- 5°C over the course of a half hour or so.  This is true regardless of how often it performs an internal non-uniformity correction with its shutter.

Second, the emissivity the camera is expecting the subject to be can have a big effect on the reading as well.  I think most thermal cameras are set to a default of 0.98 emissivity, but I think plastics are somewhat lower.  Metals are generally pretty low, depending on the surface roughness, and you can usually see neat reflections in their surface in thermal imagery.  Glass too.

That being said, this class of thermal camera is still very good at detecting differences in temperature, so your deltas will be accurate.  The smallest difference they can detect is spec’d at 0.05°C I think.  If you care about getting the absolute readings correct, you could place a makeshift black body at known temperature into the scene as a reference.  Maybe use one of your thermocouples on a piece of dull ceramic.  Oh, and of course the temperature of the camera itself is important to try and keep constant, which is hard if you change the temp of the room environment!  Anyway, just some stuff to keep in mind when analyzing the results.  Keep up the good work!

[Stephen Seeber]

Hi Craig:

I have over 40 years of experience using a variety of IR cameras for just about every application including lots of R&D studies.  The camera I use now is an A655sc.  Quite a step up from an A65.  Deeper bit rate and higher data streaming rate.   Interestingly, Flir has improved the A65 resolution over time so its pixel count is now a bit higher than the A655sc.  However, the A65 has higher noise probably because it uses a smaller physical detector size.    The drift you refer to does occur.  When I make a measurement I do a NUC and then freeze the values in my ResearchIR software.  I measure emissivity values.  The camera’s sensitivity is  better than 30 mk.  The R value measurements are quite sensitive to changes in ambient and lack of steady state in the sample being measured, so I go to considerable lengths to ensure that does not happen.  Be aware that in the first article in the synthetic insulation issue, all measurements were made by guarded hot plate.  You cannot use a thermal imager to measure  batt insulation surface temperatures.  That issue is explained in detail in the 1st deep dive article.  Thanks for reading and posting.

[Craig B]

Oh, nice!  Ok, I’m preaching to the choir here;  you obviously have it covered.  Keep up the good work!

[Ian H]

Stephen, thanks for another great article!

When I hang the washing on the line, a lot of the water drips off rather than evaporating. My wool socks dry quicker than the same brand/thickness of cotton sock, partly because the wool seems to drip more and partly I suspect because the surface has a higher nap, and it gets beads of water on the tiny surface fibres.

I’ve seen a technical fabric T shirt demo by one of the Australian Olympic team doctors, where he dipped it in a bucket, then the water ran off it like it was a pane of glass.

Their aim was a non-wetting fabric for hot sports like field hockey or soccer, so sweat would evaporate rapidly and not wet the jersey. It was a few years ago and I can’t remember any names of this stuff – it felt like a slippery Capilene shirt.

Is there any way to measure ‘dripability’, not quite the same as DWR, but the real-world ability to shed either excess sweat, or after wading through a creek/ getting caught in prolonged heavy rain? It would also be interesting to know whether claims about hydrophobic treated down have merit (I’ve seen Western Mountaineering’s website where they reckon it’s not really any better).

Obviously, a fabric that would drip dry at ambient temperature would be much better than something which you have sit in shivering in your tent, to evaporate the water. In the West of Tasmania and New Zealand, where rainfall is 4-14 metres per year, I’ve had a few trips with a good night’s sleep in dry thermals, but then having to put soggy clothes on every morning because they don’t evaporate in cool air with 100% ambient humidity. On a few occasions I’ve sworn I was going to walk next time in a 3 or 5mm neoprene wetsuit!

[Christopher S]

This really leads me toward down more and more – with a decently water resistant outer shell and down treated with DWR it just seems like a much superior choice. One can always use a down jacket warmer than needed (more down fill) to compensate for a very wet or humid environment – even if it loses more warmth than an equivalent synthetic the fact that down is so much more efficient leads the jackets to weigh about the same (and substantial performance improvement when totally dry for the down of course).

[Daryl and Daryl]

My warmest insulation, when soaking wet, is closed cell foam.

I’ve used “float coats” and home made jackets made of sleeping pad foam (e.g. bjue foam).

Float coats are heavy but they save weight in other areas so the net effect on pack weight works for me.  Here’s how:

(1) No need for a windbreaker.  It’s built in to the float coat.

(2) No need for a rain coat.  It’s built in to the float coat.

(3) I sleep on the float coat at night so no need for a sleeping pad (3/4″-1″ of padding).

(4) No need for a dry replacement garment.  The float coat is still warm when wet so I don’t need to put on a dry garment to regain warmth.

 

 

[Christopher S]

Very interesting Daryl – I thought about doing the same specifically for sleeping ( I wanted to line the upper part of my bivy with CCF

I also considered wearing a very thin wetsuit as a VBL when its super cold – sort of similar to what you are doing as neoprene is a type of CCF.

[Stephen Seeber]

I use neoprene socks and neoprene gloves in the winter.   Decent insulation value in 3mm and a good vapor barrier to protect my glove insulation from sweat.   Don’t think I want to wear a closed cell jacket or pants when hiking.  I think they would fail the MVTR test.

[Christopher S]

Definitely they would have zero MTVR but that is generally the whole point of a vapor barrier. Although I do think most people have physical ways to vent them without going through their insulation. I considered adding pitzips to a true 0.5mm neoprene top (the NRS hydroskin doesnt count – the neoprene is 0.5mm but then they have a thick fuzzy fleece lining which makes the whole thing very thick). With pitzips and a deep chest zip it might be decently comfortable when its very cold.

An integrated standard would be great for some of this stuff – different layers with zippered vents at the same locations designed to directly work with each other in a holistic way. Mix and match different layers as needed for the conditions.

[Marcus]

Lol. As a regular wetsuit wearer when I surfski (kayak) in the pacific, wearing one while hiking or sleeping sounds rather unpleasant even if it has high functional utility.

As surfski is a very wet form of kayaking and I like to paddle in the evenings, Im typically soaked near dusk as I load my boat on the car. When its windy I am quickly chilled in a 3/2 Oneill Hyperfreak (their top of the line suit). When its calm the suit retains heat much better. Based on this experience, I would not consider 3mm neoprene as a stand alone layer. At minimum I would require a wind breaker to be comfortable.

That said, Neoprene would be at the end of my personal list of things to try to stay warm due to it being clammy and never drying out. Cold wet neoprene is hell to put on, though an interesting idea.

[Josh J]

Neoprene is a good insulator but where it shines is by trapping a thin layer of water again your body and keeping it warm thus keeping you warm,  why scuba divers wear it. Also for socks where you’re feet are continually wet.

[Daryl and Daryl]

I have a half doaen float coats.  They are old and used and typically use a foam called ensolite.

I also have a neoprene coat that is more like a wet suit.

The ensolite float coats typically weigh about 2+ lbs.

The neoprene coat weighs nearly 3 lbs and is thinner than the float coats.

[Ian H]

Hi Daryl, never heard the term ‘float coat’, but it sounds interesting. Do you have a photo? And are they specifically for kayaking etc, or walking?

[Author]

Posts