Companion forum thread to:

Insulation Measurement

Jerry, thank you so much for this effort.

I really like this "quick and dirty" (relative to the standards :)) set of tests.

I recommend you measure the temp differencial across the only the foam, use this measurement to estimate the foam's insulation value and compare that to the value other value you calculated -- I bet they are close.

If you are feeling up to it I would like to see what your test tells us for incremental down overfill percentages (ie 10%,20%,30%...overfill)...that would have great value for myog.

For your theory of insulation: Air is the medium thru which heat is conducted AND convected away from your heat source. The reason you never reach 6.4clo/in is b/c of natural convection - for your test the air closest to the heat source warms locally, this localize temp increase resulting in the air's density decreasing locally. This less dense air rises and sucks cold air down to replace.

The general goal of our insulation is to slow this natural convection down. Even with good insulation (say perfect) we will still see natural convection stealing heat from the outer shell via the same mechanism as above.

I guess the graphs showing increased insulation from a supplex/event shell are a direct result of slowing natural convection currents by adding another boundary layer (ie little direct impact to CLO inside the the insulation).

The foil experiment was nice as well. I am interested in knowing the temperature of the outermost insulatation layer (event/foil) and the general air temperature for this data set.

Also, I think you are measuring ICLO (clo/unit area).
thanks again,
James

Thank you for the efforts you have put into this.
It seems that in the mid Atlantic, having an understanding of relative humidity and it's relationship to overall comfort is a valuable asset, at least in my experience. On my weekend excursions, I now look at humidity with equal scrutiny as I do temperature. As I'm sure many can attest to, hot/humid feels "hotter" and cold/humid feels "colder". How one prepares for this type of weather should affect what equipment is more appropriate.

After reading Will Reitveld's analysis of vapor permeable garments, perhaps the next study can be done with permeability in mind as well. In the building science world (which I have some familiarity with) the general analogy has always been that people throw out heat "similar to a 75watt incandescent light bulb". The reality is that we throw out a hellofalot of vapor as well, even while asleep. At home, my wife and I have found over the years that we now prefer sleeping under down in the summer as much as the winter, primarily due to feeling "less clammy" then under synthetics that we've tried. This is clearly due to the overall permeability of the comforters that we have.

Without being too long winded, I'd love to see how insulations perform with humidity in mind, since I find that this particular metric seems to not be easily measured, and can have a large affect on overall comfort. (Perhaps use some moist sponges as test subjects, or something of that nature ;>D )

Thanks again for what you have done. You've got me thinking...


Matt

Thanks for a very interesting article. I hope the editorial disclaimers don't discourage others from doing quick and dirty experiments. It confirms what we know about down vs synthetics and add insight into overfilled down, various fabrics, etc. Thanks again.

Jerry, great article and research.
I think it is interesting that you show down has a lower clo/in than the synthetics. I wouldn't have thought that was the case.


I'm still trying to figure out why I like my down sleeping quilts and bags more-

Maybe we need you to figure out personal comfort levels of insulation and fabrics combined as a whole (good luck with that one).

> down has a lower clo/in than the synthetics.
Ah, but you also need to look at the weights. Down is lighter.

Cheers

Great article. I like the effort and the data.
However I am not sure about the following conclusion.

"For example, rather than having two insulated jackets that you can layer, have just one that’s warm enough for your target minimum temperature and unzip it if it’s too warm. Rather than having a wind shirt and a shell, have just one of them. Rather than carrying a sleeping bag with two layers of fabric and squished insulation underneath, simply use a quilt."

This doesn't match my own experience. In regards to clothing layering is much more flexible in matching comfort when active.At rest is another story. There is the matter of sweat. I think a windshirt over fleece wins over a nylon fabric covered synthetic, especially in a wet cold environment.There are a host of factors in real life situations which are dificult to replicate in laboratory research.

As for quilts they have their place but if I was at the South Pole I would be using a bag even if I was squishing part of it.

A couple comments

I don't disagree with the editorial disclaimers, but since I'm just doing some relative measurements I believe my conclusions, although there is some uncertainty.

One thing I don't like is that I measured in my office and the temperature inside the insulation was not typical of skin temperature (90 F). I may repeat this outside under the night sky with 90 F "skin" temperature and outside temperatures around freezing. This especially affects the IR heat loss - my data is just a hint of what might be happening.

In my opinion, it's too difficult to include humidity, but I would like to measure insulation that's wet - have to have skin temperature that's 90 F and outside btemperature about freezing.

You like your down better because, like Roger said, the clo/oz/yd2 is smaller - less weight for the same warmth.

And as far as having the fewest number of layers, I know this goes against the conventional wisdom, but it's just that extra layers have extra facing fabric which weighs a few ounces but contributes little to warmth. Yeah, more layers is more flexible. I have used just three layers - base, insulation, and rain jacket - for years in the Pacific Northwest in the winter where it rains a lot and down to 20 F. If you add an extra layer, it only adds a few ounces, so that can work good too. I give a few numbers so we can talk about this quantitatively.

One thing that really bothers me is that I disagree with Roger and Richard Nisely about overstuffing down, because they are more expert on this subject than I am. They say adding down increases warmth even if it doesn't increase loft. I have repeated this measurement multiple times and come to approximately the same result. Someone recently posted the link http://www.backpackinglight.com/cgi-bin/backpackinglight/forums/thread_display.html?forum_thread_id=12505&startat=20. On page 2, there's a plot of "The effect of down density in a fixed baffle size on the insulation value of sleeping bags". On the left side, there's a clo of 4.4 for 16 ounces of down = 0.275 clo/oz. On the right - 5.73 clo for 28 ounces = 0.2075 clo/oz. So, for a 75% overfill, the clo/oz decreased by about 25%. This is pretty close to my data - for a 82% overfill, the clo/oz decreases by the same amount, 25%.

What difference does this make? If I'm making a down garment, overstuffing by a little, like 10%, will make sure all the baffles are completely filled, but if I overstuff by a large amount, like 100%, then I've lost the advantage of down over synthetic, but I think the down will be much worse if it gets wet.

> I may repeat this outside under the night sky with 90 F "skin" temperature and
> outside temperatures around freezing. This especially affects the IR heat loss - my
> data is just a hint of what might be happening.
Now that would be very interesting!

Note that having a single layer of tent fly between you and the night sky can cut the IR losses by a huge factor. A clear night sky can look like -70 C, while the fly might look like 0 C. That is very significant.


> What difference does this make? If I'm making a down garment, overstuffing by a
> little, like 10%, will make sure all the baffles are completely filled, but if I
> overstuff by a large amount, like 100%, then I've lost the advantage of down over
> synthetic, but I think the down will be much worse if it gets wet.
There are so many variables it is hard to make a simple rule out of this.

But where did this 'overstuffing' concept come from anyhow? I have no idea. If I manufacture a quilt then I decide how much down to put in and how big the baffles will be; the next guy making a quilt might decide to put 20% more down in for the same baffle height. Am I under-stuffing, or is the other guy over-stuffing? There are NO rules. So the concept is very artificial.

Mind you, putting enough down in that the cells are filled, with no air channels left, cannot help but be a good idea! That's not over-fill; that's sound design.

As for the down getting wet - there's an article somewhere here on BPL which actually looked at that. It was found that getting water right into the down is extremely difficult to do - much harder than with synthetics. Anyone who has washed a quilt or bag will have met this problem too. So I think the 'wet down' problem is rather over-hyped by the synthetic vendors.

Cheers

Jerry, I can't find the post/thread but I am pretty sure the "2.5X with no penalty" argument hinges on the idea that an fully lofted down garment will have more shell weight than a 250% overstuffed garment (it will have taller baffles and a larger outer surface area). So while overstuffing results in reduced thermal effiency in the down insulation(per unit weight) it also has a better down/shell weight ratio. It seems for the specific case(s) the tradeoff cancelled out until ~2.5X.


*The larger surface area of a fully lofted vs compressed garment also means you will have a larger outer surface area losing heat via convection to the outside air and therfore more heat loss - though your test doesn't see that as you are using a flat surface of heat transfer vs a curved one.

"But where did this 'overstuffing' concept come from anyhow? I have no idea. If I manufacture a quilt then I decide how much down to put in and how big the baffles will be; the next guy making a quilt might decide to put 20% more down in for the same baffle height. Am I under-stuffing, or is the other guy over-stuffing? There are NO rules. So the concept is very artificial.

Mind you, putting enough down in that the cells are filled, with no air channels left, cannot help but be a good idea! That's not over-fill; that's sound design."

If the down is rated 800 fill power, then one ounce of down should fill 800 cubic inches. Theoretically, you calculate the area of each baffle, multiply by loft, divide by fill rating - that's how much of an ounce of down for that baffle.

My sample was rectangular and I measured it pretty carefully and the regular fill came to 734 cubic inches per ounce. It was rated 800 using American standard. 734 in3 is 8% less. So, you need to "overfill" by 8% just to account for the over-rating of American system. If you have 800 fill power rated by European system then I don't think you have to do this.

In my limited experience of making down garments, I think you want to do it be weight for each baffle. If you put some in, sewed the baffle shut, fluffed it up, and then added more if the baffle wasn't full, it would be too time consuming and it would be unlikely to get evenly filled baffles with the least amount (weight/cost) of down.

Then maybe you'd want to overfill a little more, just to account for the fact you won't perfectly get the right amount of down in each baffle. So, maybe 15% if you're using American standard.

There have been threads where people complain about down garments having baffles that weren't full - there were places where there is just facing fabric from each side against each other without down between - you want to avoid that.

> If the down is rated 800 fill power, then one ounce of down should fill 800 cubic inches.
ROTFLMAO!
That applies to fresh clean dry down under negligeable pressure. It's a measurement spec for selling down, not a field guide.
Now, after a week in the field ...

Cheers

Exactly Roger : )

That's why you have to overfill a little

A thought provoking article; well done Jerry.

This compliments a piece I'm still working on nicely. I'm looking at the application of down v. synthetic fill v. fleece as insulating materials, trying to figure out how much substance lies beneath the conventional wisdom.

One big conclusion is exactly as Roger said, that getting down wet from simple outside precip is quite difficult. The area is which Primaloft seems superior is resisting humidity, especially that from the inside (sweat, wet layers).

Jerry,

Nice article. Pioneering seems apt. Glad you took it on your non-scientific self to try and duplicate your results and come to some general conclusions. It sets the bar and moves understanding along. This is what I like about BPL.

I read ROTFLMAO, then googled it, OMG!

"I'm looking at the application of down v. synthetic fill v. fleece as insulating materials, trying to figure out how much substance lies beneath the conventional wisdom."

I'm looking forward to that Dave

I wear a fleece vest a lot when I'm not backpacking and I see what people like about it even if it's not the lightest weight

My experience has been that it is day after day of damp conditions where I have had loss of loft with a down sleeping bag. Not getting it wet in any one-time event, but just picking up moisture continually for several days. I'm thinking specifically of snow camping trips in weather that hovers in the high 30's - cold enough for condensation to form inside the insulation, but warm enough so that things are melting and the humidity is high. I've also experienced getting a down bag really wet in use (waking up in a puddle, so fun), and I can't imagine how long it would have taken to get it even partly dry in the backcountry, fortunately I was only a day from the trailhead and could bail.

I would love to see the two insulation measurements come together, this one and the the R-value of bottom insulation.

I would really like to see if a R-value of 1.5 bottom insulation and 3.5 top would be as warm as a 2 bottom and 3 top. A nice difference of compared top vs bottom that would give you the best ratio.
Then go into testing the weights of of both the bottom and top items to find the best weight ratio.

I always go with more top loft and much lower bottom loft due to the lower total weight. In fact I will even use a 1/8" closed cell pad until the temp gets down to about 35*. Yes, I have to use a 15* bag to stay warm, but a 15" bag will weigh 6 ounces more where the pad will come in 9 ounces more. (Note: I almost always sleep far away from camping locations and have no problems sleeping comfortable on an 1/8" pad on soft ground).

It seems when I use a 30* bag in these conditions, 35* and a light higher R-value pad, then I will not sleep nearly as warm as I would with the other combination.

I also think that with testing, I would be proven wrong in some point that would allow me to have a warmer combination for less less weight than what I currently use.

This type of test would greatly assist new backpackers in buying pieces of gear and even the experienced when looking for a replacement piece of gear.

Edited by awsorensen on 06/18/2012 11:29:49 MDT.

After repeating my disclaimer of not being the expert on this : )

I think that R value on the bottom is heavier than R value on top. On the bottom is has to withstand the weight of your body. So you'de want more R value on top.

The ground temperature is usually warmer than the air temperature, especially in the situation that you are pushing the limits of your system. So you'de want to have more R value on top. If you were on snow it might be the opposite. After sleeping for hours, the ground directly below will actually warm up a little.

There is more conduction of heat into the solid material of the ground than the air above. i.e. the ground will suck more heat out of you. i.e. the temperature at the boundary between you and the ground will be less than above you, where there is sometimes a skin of air that adds insulation. So you'de want more R value below.

Maybe it's just too complicated to analyze quantitatively.

Maybe you could instrument yourself when you were sleeping in the field.