Question of the day?

Richard, I always enjoy reading your responses, thank you.

I was a little surprised that the R value you measured was pretty close to the advertised amount, and yet your calculated comfort rating was only 44F. I've read many people on BPL say that the pad is only good to around 35-40F so I'm inclined to think your calculation is pretty reasonable. BA is not being terribly honest by advertising a pad rating of 15F, which leads one to believe that this is the comfort rating.

Quick question… what sort of R value would a pad for an average male need for a comfort rating of 32F (freezing point)? Is there a simple (eg. linear) formula one can use for a particular body type (eg. average 30 year old male).

Thanks!

Ashley,

The 32F answer is ~ R value 5. Not coincidently, all certified laboratories use this R Value pad to determine their EN13537 ratings.

The simplest way to determine the required R value for a pad to match a bag is to select one equal to the insulation value of your bag. You can do this by calculations based on page 30 of http://www.mammut.ch/images/Mammut_Sleep_well_pt1_E.pdf. Use the Lower Comfort EN 13537 linear regression.

The most accurate way to determine the required pad value is to also calculate what impact your preferred sleeping position will have on the rating. Approximately 18% of your total insulation is provided by the compressed bag and sleeping pad when sleeping on your side and 35% when on your back. You can make a wide range of tradeoffs between the bag insulation and the pad insulation to provide the composite insulation you require.

Rod's correct about the BA Aircore/POE Insulmat general construction, which is a relatively thin insulation batting laminated to the top face of the pad. However, it's more like 1/4 or 3/8 inch thick at best. It can also delaminate, in my experience–something to watch for.

I find it gives 15, perhaps 20 degrees F colder performance compared to a plain air mattress, but in no way provides wintertime condition warmth. For affordable winter insulation, I'll suggest a 3/4 length self-inflating pad over a full-length closed cell foam pad. If you can afford a down mat instead, go for it!

Thanks Richard. Just to confirm that I'm doing this right… reading off the graph, for 0 degrees C it looks like I need a 0.9m2K/W thermal resistance. So R=0.177*0.9 = 5.1, which is approx what you said.

I'm still unsure on one thing though. For 0C/32F comfort I should also select a sleeping bag that measures roughly 5.1R (=5.7clo) right? I assume you want the same clo value of insulation on top as you do underneath. However, I've read in a previous post of yours that 800+ down fill provides approximately 6.562 clo/inch of insulation. The western mountaineering summerlite has roughly 2 inches of single layer loft, which would give it a clo value of roughly 13… which is way warmer than the 5.7 clo I thought was needed. Obviously I'm doing something wrong somewhere….!

I would think the insulation value for your pad could be much less than your bag. I base this on the idea that the ground below your pad is a better insulator than the air above your sleeping bag.

Reckon Evan is right. When we take our tent down after a night on the snow, we find a hip-shaped depression underneath. This tells me that the snow has melted slightly under our hips – which is pretty reasonable. But, and it's a big but, that means the snow has warmed up to melting point, and that may be a lot warmer than the ground.

On the other hand, the air wiffling through our tent can be a whole lot colder than that – all the time.

Cheers

Hi Rod

> I'm not sure whether this also acted as a heat sink for the rest of the mat, pulling heat out generally.

Ooh – 3-D heat&mass flow problems … :-)
Yep, I think you are right. The air can circulate lengthwise along inside the mat, so the bit touching the cold ground could act as a heat sink for the rest. An interesting thought.

Ashley,

Western Mountaineering and other manufactures targeting UL backpackers manufacture their sleeping bags with the lowest possible density of down. Their density is 2.16 kg/m^3 which yields ~ 2.556 clo/inch of loft. In the absence of an independent EN 13537 test, UL backpackers will typically buy the bag or quilt which has the most loft for the weight rather than the warmest bag for the weight.

The optimal fill density for 800 fill down is 24 kg/m^3 which yields ~6.562 clo/inch of loft. Some manufacturers offer "over stuffing" options of 2 -3 ounces for a nominal additional fee. Most UL backpackers don't realize that with the 2 – 3 ounces of "over stuffing" they have moved the down density toward the optimal 24 kg/m^3. Yet they are still no where close to down’s optimal density for warmth.

Fascinating, Richard. I've noted with great interest your comments previously that more densely packed down is not a bad thing (in reference to Montbell bags which don't tend to loft as much as WM bags). One often gets the impression from reading articles or forum comments on BPL that it is only the inches of loft that counts.

From your figures then, a sleeping bag could theoretically be compressed and still retain the same warmth. It seems to me then, that it is not particularly critical to avoid compressing the down in a jacket worn inside a sleeping bag. Obviously you wouldn't want to crush it (as happens beneath your back) but some compression of the loft of the jacket and bag should be fine right?

It would be good to see more EN13537 ratings on US sleeping bags. Marmot seems to have them on most of theirs, and they generally agree with the advertised temperature… although often the advertised specs are a degree or two on the "optimistic" side of things. Presumably WM is usually bang-on target, and Montbell too (I'm basing that purely on user reports).

Using your revised figure of 2.556 clo/inch for lower density 800+ down, gives roughly 5.7 clo for a summerlite if you assume 2.25 inches of loft on top (it has continuous baffles so presumably this is quite achievable). So that all works out quite nicely for a 0C comfort rating.

Thanks once again for your helpful comments.

[Edit: deleted an erroneous figure after reading Richard's posts below!]

Richard,

Thank you for some info regarding clo and down. It's always bugged me to have no objective comparison of the insulation value of an ounce of down vs an ounce of synthetics. Can you steer us to a source of down's clo/inch for other densities corresponding to, say … 10% overstuff and 20% overstuff?

However, in this discussion domain (UL backpacking) would not the optimal down density be that which yields the highest clo per wt_per_surface_area rather than clo/inch_of_thickness? I believe that'd be expressed as clo*m**2/kg. Clo/inch would certainly be the right choice where volume of insulation were the overriding factor.

Fascinating stuff! Thanks, Richard.

Two questions:

1) I just want to double check your 2.16kg/m^3 and 75kg/m^3 figures for typos. If I read correctly, the optimal clo/thickness ratio is achieved at almost 35 times that typically used in LW bags!?

2) As Jim mentioned above, for LW backpackers, clo/oz is often a more valuable metric than clo/in. What density optimizes clo/oz?

And…one comment:

I want to make sure readers realize that if you took a given amount of down and compressed it from 2.16kg/m^3 to 75kg/m^3 (such as compressing the down under your body in a mummy bag), you would *not* increase its warmth. Using Richards figures, the clo/thickness would increase approximately 2.6 times. But, the thickness itself would decrease approximately 35 times. The end result would be over 13 times less insulation than the uncompressed down.

Best Regards,

-Mike

Jim and Jim,

"It's always bugged me to have no objective comparison of the insulation value of an ounce of down vs an ounce of synthetics"… things are even worse in the realm of synthetics. All of the synthetic insulation garments I have tested provide less insulation than the insulation manufacture's clo/oz published specifications. Sometime in the future, I will talk about this in more detail. In the mean time I am leaving for a one month trip to tramp the bush.

The average commercial mummy sold in the US for three season use has a US advertised temperature rating of 20F. Ironically the EN13537 would only give this bag a Comfort rating of only 40F but that is another topic. It is an 800 fill mummy shaped and has a hood, draft tube, 40 denier fabric, circular neck collar with a draw cord and 3.25 inch baffles throughout the body. The same bag manufactured with different levels of overfill yield the following values.

“1) I just want to double check your 2.16kg/m^3 and 75kg/m^3 figures for typos. If I read correctly, the optimal clo/thickness ratio is achieved at almost 35 times that typically used in LW bags!?”

2) “As Jim mentioned above, for LW backpackers, clo/oz is often a more valuable metric than clo/in. What density optimizes clo/oz?”

1) Since geese like people are quite individual, the curves for each batch of 800 fill down tested will be different but the average is found in the following table. The curve looks like a hockey stick. The clo increase verses density is approximately linear from 2.16 kg/m^3 to 18 kg/m^3. It then changes its slope to become almost flat until ~ 75 kg/m^3 where the clo starts to decrease. As UL backpackers we are only concerned with the linear segment of the curve.

2) For all practical purposes the clo/kg for different density stuffed bags are equivalent. Granted there is a~3 % benefit for the lowest density but is in the realm of computational noise.

Thanks, Richard!

That's very valuable information — as always from you. :)

So, assuming I have understood this right, in the linear region of the curve (where UL backpackers operate) if you decrease the thickness of your typical UL sleeping bag by a factor of 2, you will not affect the warmth of the bag. Why? Because the density will be doubled, and so the clo/inch will double, but the thickness will be halved… so 2*0.5=1 and the warmth of the bag will stay the same.

I think is a particularly interesting fact. Surely it means that you needn't worry so much about needing lots of room inside your sleeping bag or hard shell to allow a down jacket to loft fully. A partially compressed jacket should be just as warm, so long as you are not completely crushing it (eg. lying on it!).

Richard, please correct me if I've got this all muddled up! It wouldn't be the first time… ;-)

Ashley,

Spot on!

>> if you decrease the thickness of your typical UL sleeping bag by a factor of 2, you will not affect the warmth of the bag. Why? Because the density will be doubled, and so the clo/inch will double, but the thickness will be halved… so 2*0.5=1 and the warmth of the bag will stay the same.

>> Spot on!

Hmmm…

I must be missing something here. The numbers aren't adding up for me.

If you start with down at 2.16 kg/m^3 and 2.556 clo/inch, and compress it to 18 kg/m^3 at the end of the linear region, it would need a 21.3 clo/inch value (at 18kg/m^3) to have the same clo it started with.

21.3 clo/in far exceeds even the maximum value of 6.562 clo/in which occurs well outside of the linear range.

What am I missing?

Good question Mike! It would appear that the linear region would have to tail off rapidly at about 5kg/m3 in order to not exceed 6.56 clo/in at 18kg/m3.

That would mean you could compress your bag by roughly 2.5 times before it started to become a problem.

I note that in Richard's example figures for an overstuffed mummy bag the amount of down in the bag is effectively doubled during the experiment whilst still retaining a linear increase in clo… so equally well you can can be fairly sure that squashing a bag by less than 50% won't affect it's performance.

And if you decrease the thickness of the down by a factor of 100 you will increase the density by 100, meaning that a layer of down one thou of an inch thick will have an insulating value *per inch* exceeding anything on the planet.

Yeah, right. Beware extrapolations! Looking closely at what Richard wrote we find that the linear region only goes so far, then it fails. Yup.

The insulating value of the down will rise with density for a little while, until the ends of the down fibres are starting to touch everywhere. When that happens the Kozeny equation (I think that's the correct spelling) breaks down, because it is mainly the air around the tips of the down fibres that insulates. Continuous filament insulation (synthetic) is not as good as down precisely because it does not have the fibre ends, only continuous filaments, to trap the air.

The same thing explains why damp down is a poor insulator: the microscopic fibres start to stick together. Loss of ends => loss of trapped air => loss of insulation.

Cheers

> Yeah, right. Beware extrapolations! Looking closely at what Richard wrote we find that the linear region only goes so far, then it fails. Yup.

I don't think anyone is in disagreement on that point! The interesting thing from Richard's data however is that there *is* a significant linear region and you don't have to extrapolate beyond the data points to show that 50% down compression on most UL bags should not be a problem. Losing loft because your bag is damp is another matter altogether!

Richard, it has been too long since I was in school and I think I went skiing the week/semester they discussed thermodynamics. Maybe it’s too early in the morning for me but what is clo (I probable know this but I’m having a brian cramp). I think I’m following what you are saying but the clo thing keeps me guessing.
I really appreciate you guys- It is not enough for me to just know I was cold but the why and how to fix it. It also helps in other areas to understand how not to let it happen with other equipment.

>> The interesting thing from Richard's data however is that there *is* a significant linear region and you don't have to extrapolate…

Indeed. We're all on the same page here. My original point was that according to Richard's numbers, the linear region extended out to 18kg/m^3 and "flat-ish" region extended from there to 75kg/m^3. So my last example was actually interpolating within the values Richard gave us, not extrapolating past the data points.

I suspect that the reason for the discrepancy is that the Y intercept of the clo/in vs density graph is above zero. If true, this would reconcile the data points we have been given. But, it also would mean that doubling the density would *not* double the clo/in — even in the linear region.

EDIT: I just found this old post from RN that supports my suspicion:

http://www.backpackinglight.com/cgi-bin/backpackinglight/xdpy/forum_thread/12505/index.html?skip_to_post=94791#94791

(Look at the thermal conductivity vs density graph.)

Cheers,

-Mike

Tad,

As a builder you understand that R Value measures the resistance to heat flowing through an item. clo is the same thing as R Value. To convert R Value to clo, multiply the R Value by 1.13931529284938. To convert clo to R Value, multiply the clo by 0.8832059100018.

clo is generally easier for people to relate to when evaluating clothing or sleeping bags. 1 clo is the COMBINED insulation provided by a complete winter weight business suit ensemble. This ensemble includes the shoes, socks, underwear, shirt, tie, long sleeve shirt, winter weight pants, and winter weight jacket.

Richard, Thanks- BTW is that a "Hickey Freeman" wool suit or a "Men's warehouse" wool suit? There is a difference. Just kidding.
What does the initials "clo" stand for?

Mike, I see what you are getting at and am now rather puzzled! Reading the thread you linked to I found some comments by Richard similar to the point I was making above:

"You can increase the aerial density of 800 fill down about 2.5x before it is becomes less thermally efficient than fully lofted down. A compression sack will easily compress the down much more than this; so, it is important to let it loft up. This same phenomenon applies to an earlier thread dealing with the diamond quilting on a MB women's down jacket resulting in less loft than the rectangular quilting in the same man's jacket. Again, it doesn't make any difference."

Further down the page…

"If your Parmo raincoat doesn’t compress your Gilnet jacket such that the density increases more than 2.5 X, you should be fine. You can probably find hundreds, if not thousands, of Internet posts which make the statement that the loft of a down garment or bag determines its warmth. My contention is that this is true only if they are using the same fill power and density. You can make a down bag or clothing warmer by increasing its density even if the loft is less."

I'm not quite sure how these statements can be reconciled with Mike's point that the linear equation seems to have a significant y-intercept constant (meaning that doubling the x value won't double the y value).

I'm a writer. My eyes (and mind) glaze over when I see equations. A couple of "real world" examples on density, FWIW. We all know and respect Western Mountaineering bags because, along with being light, they're really warm. If you've had the experience of comparing not just loft, but density of loft, of similarly rated bags, you know that the Western bags have more down in each baffle. In my shop right now I have about 40 models of down sleeping bags hanging. If I compare the density of loft between models, Western bags basically always win. The down vest I recently made used much more down than some would expect; several people felt it was overfilled. But it didn't lose any warmth. In fact, it's the warmest garment I have.

To be as unprecise and speculative as possible, I'll also ask this: just how much loft does a goose have? The birds I've seen aren't several inches thick of loosely packed downy poofs. The down on the birds I've seen is pretty darn dense. Yet it seems to keep them warm…