Insulation efficiency of different down densities?

I can imagine 800 power could be more than twice as insulating as 550. As you say lower density will reduce conduction throughout the down as well as forming a thicker layer of insulating still air in the case of the 800.
Compressing the down slightly may increase the insulation per inch. I do not know.
What I thought Richard was saying was that compressing 800 fill down slightly from maximum loft increased the absolute insulative value of a given weight of down. This I do not believe. I tried to question Richard about one of his graphs on this point and we could not understand one another so I gave up.
If you provide 8 oz of down in a sleeping bag you mostly want it arranged for maximum insulation. not maximum insulation per inch

I am also interested in this question, as well as Richard's theoretical comparison of a PHD minimus jacket's warmth (unbaffled) to a Skaha (baffled). Because the Minimus had more total down, he rated it warmer, indicating that baffles add no additional warmth to a bag/garment. Aside from not believing this, I know from personal experience that the Skaha is warmer than the minimus.

What I took from Richard's graphs was that most bags were not filled to their 'optimal' density. I can't rectify this with the real world where we are more interested in the optimal weight:warmth ratio, but it did sound like he was implying that adding 7 times as much down, or alternatively compressing your down 7-fold would lead to 'optimal' density. I guess this is the amount beyond which adding more down does not increase warmth???

Well, Marmot makes different bags with different fill-powers for the same temp rating. Dimensions are pretty much identical.

Their 15 degree bags are:

Helium – 19.5oz of 850+ fill
Pinnacle – 22.5 of 800+
Sawtooth – 22oz of 600+

I'm at work so don't have the moment to work out the percentages.

Unfortunately if you take test data from different sources then try to find a trend it may be incorrect. That may have been the issue if 550 down was from one test and 800 down from another.

Unfortunately most of us have to rely on test data that we find online, with little background on the test protocol (or financial motives of those publishing the data).

Lynn,

I said that the optimal linear range for increasing down density without a penalty is about 2.5x, not 7x. Down parkas designed for expedition use typically provide baffling to allow ~16 oz of down to fully loft. For extreme high altitude use manufacturers offer their parkas overstuffed with up to 60/70 oz of down in the same baffle size. This goes way beyond the linear region but they still accrue additional warmth benefits.

Last year I conducted thermal lab tests on a large number of jackets from different manufactures. There R2 correlation was .9928 between the amount of down and the Iclo value regardless of whether the jackets were fully baffled or sewn through the range of 6 oz of down. A standard baffled Skaha (M) has 4.5 oz of down. Obviously there will be a measurable improvement (>5%) in the baffled jackets over stitch through designs at some fill level. Looking at the marketplace, most manufacturers seem to feel that baffling begins to offer measurable benefits after about 6 oz of down.

That clears things up a little…

"the optimal linear range for increasing down density without a penalty is about 2.5x"

I guess I'm not clear what the 'penalty' is? Does this mean, for instance, that a Skaha would benefit more from adding extra down (up to 2.5x) without increasing baffle height, as opposed to adding extra down AND increasing baffle height? Or would the two options be warmth equivalent? Would a Shaka with baffles half as high but the same down fill be as warm as the one inch high Skaha?

On a side note, have you or are you willing to do actual thermal testing on a Minimus and a Skaha if someone supplied you with one of each? I think a lot of folks would find that kind of info of value (can't remember if you've tested a MontBell UL down inner, but that would be good to compare as well).

Richard,

Thanks, I went back and read some of your reviews. I had never imagined that packing more down into a confined space would add insulation value.

Now I understand how Feathered Friends can pack 21.4 ounces of 850 fp down into a jacket. I figured it would need 8" baffles to hold that much down if it was meant to fully loft.

I'd be willing to say that overfilling garments by maybe 10% – 15% would be a good idea as garments tend to get constantly moderately compressed in certain areas like shoulders and arms. "Overfilling" would maintain loft better under this moderate compression.

Overfilling the bottom of a down sleeping bag makes some sense but I think the few hybrid bags around with down tops and synthec filled bottoms are more than likely the warmer option.

I will say from sad experience that wearing a down jacket UNDER a backpack leaves the back of the down jacket soaked with sweat and useless as an insulating garment. For backpacking purposes down is a camp garment, when metabolism is lower and the cool night air comes in.

Jim,

I just looked at your original post in this thread and am now attempting to answer it.

If I said that 800 fill power insulates more than twice as well as 550 down, then I made a mistake. I can only recall my posts dealing with the effect of different densities of the same fill power. If you can find the URL for that thread, I will correct it.

I think the easiest way to understand the impact of different down fill powers is to explain what would happen with a typical 3 season+ mummy filled with 22 ounces of 800 fill down and EN 13537 LLimit rated at ~10F. Using durable down proof nylon fabrics (300 warp yarns per in. x 300 filling yarns per in.), the bag’s total weight would be 43.6 oz. Assume you replaced the original 22 ounces of 800 fill down in the bag with 22 ounces of different fill powers. For each 50 in3 difference in the down fill power rating used, the clo value of the bag would change by about .1 and the resultant LLimit temperature rating 1.8F. So that same bag filled with 550 down would have -.5 clo and an EN 13537 LLimit rating that is 20F.

Down clusters are the "magic stuff" that provides excellent insulation. The difference in the down cluster content versus "the other stuff" is relatively small between 650 fill and 800 fill. "The other stuff" consists of dirt, broken and whole feathers, residue, and down fibers broken from the cluster. They are commonly present along with down clusters. 650 fill has on average 85% of the "magic stuff" and 800 fill has 95% of the "magic stuff". Although your original question was relative to the difference between 550 fill and 800 fill, it is now rare to not have 650 fill being the lowest fill option available.

Richard-
To avoid embarrassing myself I sent a PM. I may have misunderstood one of your posts from 3 years ago. Let me know.

Jim

Lynn,

The Skaha would insulate not more, but approximately the same. The relationship between density and insulation is hockey stick shaped. If you stay in the upper handle range the solutions are optimized for UL backpacking.

Another way of looking at this issue is to compare a Montbell SS sleeping bag with a Western Mountaineering bag using the same approximate amount of fill. The Western Mountaineering bag will loft more and LOOK warmer but the bags will thermal test the same. That is because the MB SS bags stretch elastic compresses the down fill in the linear 2.5x range.

I am working on another project now but when I free up I will contact you about testing the Minimus and Skaha. I agree with you that it would be very interesting to see the results.

I tested all of the MB products and posted a chart to the forum. I tested many other jackets also. Since they weren't popular with the UL community I didn't include them in the chart.

Just one note about the Nunatak gear (who make the Skaha) – in my experience (2 quits, 1 vest, 1 jacket) the shipped weights have deviated from the spec, so it seems quite possible that the fill weights aren't totally consistent either.

>I know from personal experience that the Skaha is warmer than the minimus.

Well this is very very tricky to judge from perceived warmth (assuming this is what you mean). The two jackets are surely going to be at least similar in warmth, are you sure you can really feel the difference, given all the other factors?

Just a further note on Nunatak fill weights.. Now, don't get me wrong, I love Nunatak gear (I have plenty of it), and there's no doubt it's warm and ultralight – even if it is very expensive.

But, for example, look at the Skaha Vest published weights in Pertex Quantum, size M:

Fill weight: 3.75oz
Total weight: 5.5oz

From this we can deduce that the fabric weight is 1.75oz… which I don't really believe, given that the lightest wind vest (admittedly full zip) I can find is 2.2oz, which is only a single layer of fabric. And at any rate, the vest in reality is most likely heavier – my stock Skaha Plus Vest in Quantum 0.8 is 7.5oz instead of the listed 6oz weight. Is the extra weight more fill, or is the fill accurate and this weight the real fabric weight? Or is the fill in fact less than listed, who knows?

So the problem with calculating warmth based on down weight is that you have to trust the published fill weight (if there even is one).

My clusmy way of measuring warmth (assuming fabric weight+type is similar): how hard is it to stuff into a sack :) It's not totally ridiculous – I used to have a custom Minimus Ultra Vest with no pockets and a stud on hood, but replaced it with a Skaha Plus Vest (mainly for the better hood). The Skaha takes more stuffing into the same volume, therefore it's warmer (and personal experience seems to bear this out, but frankly I don't trust that).

"are you sure you can really feel the difference, given all the other factors?"

Exactly why I would like Richard (or someone) to do a comparison test.

Adrian,

The Skaha Vest published weight for a size medium of 5.5 ounces and 3.25 fill weight leaves 1.75 ounce which includes both the inside material and the outside material. Then you add a small amount for the 14" zipper and the elastic cord and cord lock around the bottom of the vest.

The material for a size medium vest would end up being at least 1.5 squard yards. Add all that together and you get the 1.75 ounce number.

With 3.75 ounces of Down even one made out of the lightest Cuben would weight about 5 ounces.