Comparison of Various 20F Underquilts
Hi All!
I am currently shopping for a 20F underquilt to supplement my 40F underquilt. In order to provide a rational for choosing among the great cottage manufacturers that we have (aren’t we lucky!), I decided to run a little analysis and provide a visualization of the different offerings that I am considering. For these analyses, I used the number provided by the different vendors on their websites, for different UQ between 50 in long to 83 in long, and for down FP ranging from 800 to 950.
The following graphs do not try to account for any design peculiarity (karo baffles, longitudinal baffles, taper, draft collars...) but merely look at rudimentary metrics such as the total weight, the fill weight, the structural weight, and the fill volume to learn a bit more about the efficiency of the different designs. The goal is not to estimate which underquilt is the best value and this is the reason why I have left the cost element out of the picture.
** Image Here**
UQ_COMPARED.png
Graph 1) 2) The first and second graphs indicate the total weight and the fill weight of seven 20F rated underquilts from Hammock Gear - Incubators in green, Enlightened Equipment - Revolt in purples, Loco Libre Gear - Habanero in yellows, and Under Ground Quilt Zeppelin in blues. The first graph combines both metrics. What is interesting here is not the magnitude of the numbers, since we have different down fill powers (800 to 950), but rather the slope of these curves. In particular, you would expect that the lower FP have a steeper slope as the quilt weight will increase more for each additional inch of length. It is true that the 950FP down underquilts seem to have, on average, a slightly shallower slope but this is not the case with the LLG quilt, with the two lines for the 800 and 900FP Habaneros being perfectly parallel, is LLG taking advantage of the higher FP of 900 down? What is also apparent in this graph is how, for a given length, the EE Revolt underquilts puts a lot of effort in being ultralight. The Revolts really stand out as being lightest while not having significantly less down than its competitors.
Graph 3) The third graph tries to account for the different fill power of these underquilts by plotting the unconstrained fill volume when new computed by multiplying the fill power by the weight of down insulation. Here we can see that the LLG underquilts carry the largest amount of insulation volume among the different underquilts for a given length. This raises the following question: are the LLG baffle thicker and with more loft or are the baffles really overfilled to ensure better loft and heat retention as the down degrades over time? Also the slopes for the LLG and HG underquilts seem to be slightly steeper than for the other two manufacturers. Could that be related to the amount of taper at the extremities of the underquilts as the size increases i.e. as length increases, relatively more of the underquilt is of full 44 inch width and the taper effect gets diluted?
Looking at the third graph and disregarding the construction technique, it seems that the HG underquilt has less unconstrained down volume for a given length than the other underquilts. The EE underquilts, despite being the lightest around, are in the lower-middle of the park for unconstrained down volume meaning that their lighter weight does not stem only from a lower amount of insulation but rather from a lighter overall structure (suspension, fabric, baffle mesh...).
Graph 4) The fourth graph looks at the structural efficiency of the different underquilts. The higher this efficiency metric, the more down there is for an ounce of underquilt and thus this should give some insights about the amount of dead-weight in the underquilt i.e. weight that is present but that does not directly contribute to the insulation (could contribute indirectly through additional features such as more sturdy suspension...). Again, disregarding any fancy feature (draft collars), or down FP, the graph plots the ratio of down weight to total underquilt weight and tries to indicate how efficient the construction is. Looking at that metric in isolation, it seems that the Karo baffle technique of EE, and/or the fabrics used by EE lead to a significantly better structural efficiency. The LLG and UGQ are in the middle of the pack while the HG is towards the bottom.
Graph 5) The fifth graph is not useful in itselfas it just shows that all the manufacturer have roughly the same increase in total underquilt weight with the same increase in unconstrained fill volume. The slope however is interesting as the higher FP underquilts should exhibit shallower slopes and they do on average. The difference is marginal however, and this begs the question: is paying for a higher FP worth it? The LLG underquilts have a slightly different behavior with a steeper slope than their colleagues which is surprising as this means that the habanero underquilts become relatively heavier with increasing fill volume. Going back to the previous graph, this is also highlighted by the structural efficiency of the habaneros which does not grow as much as its competitors with overall underquilt length.
Graph 6) The sixth graph tries to normalize the attributes by dividing by the length of the underquilt. Normalizing allows to somehow take into account the length effect and highlights the contribution of the suspension weight. Here, a higher normalized fill volume (volume per inch of length or basically lateral lofting surface) should indicate the amount of lofting if the down were to be unconstrained. A higher number is better as the loft is higher if unconstrained. The normalized total weight is to be minimized, so in the end you want to be in the lower right quadrant. The LLG underquilts have a lot of down but are also quite heavy for their length and this is why they end up in the top right corner. The EE ones have less unconstrained down volume but are also very light which is why they end up in the lower left corner. The HG Incubator looks a bit like an outlier and seems to be Pareto dominated by all the other underquilts. The UGQ Zeppelins are in the middle of the pack.
Graph 7) The seventh graph shows the normalized fill weight as a function of the normalized total weight. Here you want to be in the upper left quadrant, meaning more fill weight per inch of length and less overall weight for inch of underquilt length. Again, a Pareto front seem to be highlighted with the purple EE Revolts, the blue UGQ Zeppelins, and the yellow LLG Habaneros on the Pareto front. The erratic behavior of the 900FP Habanero quilt is puzzling as it is different from all other underquilts. This is again due to the fact that LLG seems to be putting more down in their 900FP quilt as the length increases.... or there are rounding issues with the weight numbers provided
Graph 8) The eighth graph shows the normalized structural weight as a function of the normalized filled volume. Typically, one would want to be in the lower right corner to have more loft and less structural dead weight. The HG quilt looks a bit like an outlier here and I am not sure why.
Graph 9) The ninth graph shows the structural efficiency of the quilt as a function of the fill volume. It shows that as the total amount of down volume increases, the structural efficiency of the underquilt increases which is to be expected as the weight of the suspension gets diluted in the overall weight of the underquilt. Underquilts with lower FP down (800 and 850) should theoretically migrate to the top of the graph for a given 20F temperature rating. This is because more down weight is required to achieve the same loft or volume. However, this is not necessarily the case here. A given underquilt design cross-section exhibits this behavior (meaning for a given underquilt design) as is shown with the two EE Revolts, the two LLG Habaneros and the two UGQ Zeppelins. However, for a manufacturer cross section, this is no longer the case: for instance, the 950FP Revolt appears close to the top, above the Habanero 800, the Zeppelin 850, and the Incubator 850. On the other hand, the 850FP Incubator appears at the lower end of the spectrum below all others.
Graph 10) The tenth graph, I find interesting. It shows a synthetic score dividing the normalized (or not) fill volume by the normalized (or not) structural weight as a function of underquilt length. For a given target length, you would want the highest score possible as this indicates that you get more down volume (if volume unconstrained) or more down loft retention over time per amount of structural dead weight. Here, the EE Revolt underquilts appear towards the top of the graph owing to their very light construction. The LLG Habaneros and the UGQ Zeppelin are in the middle of the pack.
Unsurprisingly, the 850 and 950 FP Revolt end up on the same line which means the manufacturer is consistent in its temperature rating (i.e. same down volume, same structure weight) but this also indicates that the manufacturer does not account for the faster degradation of higher quality down (900 and more) over time.
The 850 and 950 Zeppelin ends up on roughly the same line. I am not sure why there is a discrepancy here but it seems you get slightly more down volume if you choose the higher fill power. The 800 and 900 Habaneros are close but surprisingly not confounded. Again, you get more down per structural weight if you go with the higher fill power down. It seems these two manufacturers are a tad more conservative when it comes to down with higher fill power, and probably expect the 900FP and 950FP down to lose some of its expandability / loft over time and therefore overfill a bit more these underquilts.
Take away? I am not sure there are any
Overall, it seems that the EE Revolt underquilts exhibit relatively higher scores. This is mostly due to the Revolt lighter structure I believe. It is possible that the Karo baffles require less material. It is also possible that the fabrics used are lighter. I am not sure how the ounce per square yard of the 10 denier fabric used by EE compares with Argon 67 but this could explain a lot of the trends seen in these graphs. Finally, it is also possible that the lack of true draft collars simplifies the construction and makes the underquilt lighter (the draft collar has some down in it though). The EE Revolt seems to best fit the ultralight backpacker looking for the absolute lightest and most packable option around.
Besides, it appears that the HG Incubator underquilts end-up being dominated for many of these metrics. I am wondering if this is due to the leg shelf design that may end up requiring more fabric for the same amount of down, or said differently, less down for the same amount of fabric. It could also be that Argon 67 is a tad heavier than some of the materials used by some other manufacturers.
The LLG Habaneros and UGQ Zeppelin seem to be in the middle of the pack for most metrics.
The LLG Habaneros are the heaviest but also sport a lot more down than their competitors for a given length. They ends up having a composite Fill Volume over Structural Weight score above average meaning that you get a lot of down goodness for the extra underquilt structural weight. It is possible that the LLG folks are overstuffing significantly their underquilts to retain loft over time. Maybe they are a little too conservative? The LLG underquilts seems to best fit the persons sleeping cold and/or wanting to be absolutely sure that the down is evenly distributed within the baffles and/or that the underquilt will retain its loft for a long time.
Now, all this is very much theoretical and we all know that Practice != Theory.
However, most people say you cannot go wrong with any of the cottage manufacturer. So assuming that all of them can produce an underquilt of the exact length you want, you still have to provide a rational for your purchase decision. Or you can purchase all of them.... If you do not want to simply flip a coin, hopefully these graphs can help you!
Finally, it is also true that the actual design of the underquilt and its suspension carry as much importance for your overall comfort.
Have fun!
Minos
UQ_COMPARED.png
Reply With Quote
Originally Posted by cmoulder
Bookmarks