Roger, thanks for the insight into your testing apparatus. Kudos on encouraging independent testing!

Question - you mention that you obtained a calibrated piece of insulation. What was it? Concerns - 1) Consistent ambient temperature. 2) Consistent coolant temperature. 3) Eddie currents and convective heat dissipation (as mentioned) in air-filled pads.

For pad to pad comparison, would it not do to put sand bags wrapped in an electric blanket evenly distributed inside of a sleeping bag on top of each pad and place an array of thermocouples above and below each pad? You could mount a set of six at 30 cm spacing along a 3 cm x 185 cm strip of thin lexan, for example, and align the strips vertically at their ends, effectively sandwiching the pad. Again, consistent ambient temperature would be the biggest variable, but this would allow a precise measurement of the temperature differential between the sensors above and below the pad. I haven't thought through how you would derive R value based on this differential, but this seemed closer to a real-world test to me.

Edited by nmanhipot on 04/28/2011 15:48:58 MDT.

I've been looking forward to this article, thanks

I will be interested in seeing R values of various pads and other stuff

> calibrated piece of insulation. What was it?
EVA30 CCF foam

> 1) Consistent ambient temperature.
Controlled and measured hot and cold plates

> 2) Consistent coolant temperature
Huge water tank on house, plus continuous measurement.

> 3) Eddie currents and convective heat dissipation (as mentioned) in air-filled pads.
Limited compensation by wrapping the towel around the system as shown. But the stability is greater than you would get in the field anyhow!

> an array of thermocouples above and below each pad?
This is possible but messy. Using solid aluminium plates to even the temperature out is an accepted industry method, and works fine for this case. Then you only need to monitor the temperature of the Al plates in a few places. I found *from experiment* that monitoring at the centre was adequate.

Btw - thermocouples require expensive electronics and need calibration. They are fine for very high temperatures. For ambient conditions they have been largely replaced these days by either thermistors or semiconductor sensors. I use the latter.

Cheers

The benefit of your test is not the accuracy of R, but in the consistency of the measurement across all products. We await your unbiased results.

Great article - kind of stuff you don't read anywhere else...

I would like to see tested the oft-discussed "double up" practice of 3-season air mattress coupled with a thin pad for extra warmth. I wonder how that compares to other solutions, particularly those 4-season pad with a purportedly higher R-value but at a cost of greater weight.

Finally, I do agree Peter, a pad that allows you to sleep well is the best pad out there....

Dirk

Hi Dirk

> I would like to see tested the oft-discussed "double up" practice of 3-season air
> mattress coupled with a thin pad for extra warmth.

That WILL be included. We have some foam mats for that specific purpose.

Cheers

"How does down work in practice? It's very clever actually. The down is made up of lots of very tiny fibres, and around every tiny fibre (and especially the tips of those fibres) there is some still air: a boundary layer. That still air is what makes down such a good insulator. Synthetic fill tries to copy that, but the long fibres in synthetic fill do not trap as much air as the ends of the down fibres. Synthetic fill still has a long way to go to match down."

Doesnt this imply that using synthetic fibres as wadding limits their efficiency, and that using them as short loose fibres contained by baffles (just like down) would be more efficient. I have some cheap down-style synthetic things that seem warmer than would be expected (but have heavy shells).

If I understand your article correctly, you will be the first person to test r-values for sleeping pads with the heat/cold source on TOP of the mat. Unless I was given bad info by an industry rep, pads are currently tested by placing an object on top of the pad and gradually lowering the temperature under the pad. They then measure the object's temperature to determine heat transfer. Since the object on top of the pad doesn't give off its own heat, the r-value for some pads (e.g. neoair) is not actually as high as a regular heat-emitting human user would experience.

I heartily applaud this effort.

Roger -

Thank you! I am very glad to read that you are doing the test. I very much look forward to this series.

Dirk

Great project. But crikey, Roger - how do you ever find time to go hiking?! :)

Looking fwd to the ensuing report.

Hi Alan

> that using them as short loose fibres contained by baffles (just like down) would be more efficient

Well, that would be the right idea, but in practice it doesn't work out because the cheap loose fibre fill is usually much coarser than the tips found in down. So you end up with probably orders of magnitude less tips per volume.
Btw - that's also how the very old kapok insulation worked, similar to down but much heavier.

Cheers

Hi Mitchell

> test r-values for sleeping pads with the heat/cold source on TOP of the mat.
Some thought actually went into this. If you put the heat source at the bottom you are driving convection inside the mat, which would make all the air-filled mats perform really badly compared to the foam-filled ones. It would not affect the foam-filled mats all that much.

Anyhow, having the heat on top is a closer approximation to how we use them in the field. And that was important.

Cheers

Roger

You said you're using metric R values not Imperial

I thought air pads are speced in Imperial R values

Like the Prolite is speced at R value of 2.2 and is 1 inch thick

Styrofoam from the building store is speced at R 3.9 for 1 inch, which is about 22 in SI R. I would expect styrofoam to have a little more R per inch. Prolite R value must be Imperial.

Very confusing because the symbol "R" is used for both

Hi Jerry

As I said in the article, 'The numbers for imperial R-values are about six times those for metric R-values, but are mainly used for building products in America.'

So if the imperial R-value for Styrofoam is 3.9 the metric value will be 3.9/6 = 0.65, NOT 22.

Yes, airmats are specified in metric.

Cheers

So, Prolite is 2.2 R metric? That doesn't make any sense.

It's very confusing, because like it says in wikipedia, the term "R" is used for both Imperial and metric so you have to infer from context which one it is.

Sometimes they use the term "RSI" for metric R.

But if my R-19 fiberglass insulation is 3.5 inches, that must be Imperial.

And they said at the building supply store my 3/4 inch styrofoam is R 2.9, that must be Imperial.

Those are closer to the Prolite being R 2.2 for 3/4 inch is Imperial. It makes sense the Prolite would be a little less per inch of loft than styrofoam

Imperial R 3.9 = RSI 0.65 - oops, thanks, we don't do metric here in U.S. : )

Hi Jerry

> oops, thanks, we don't do metric here in U.S. : )
Get used to it. :-)
Almost everything apart from building materials have metric R-values.
We usually specify PET bottles and Nalge bottle in Litres.
Anything in the science community is metric.

Cheers

Edited by rcaffin on 05/02/2011 17:25:34 MDT.

Nice work. Justifies my subscription.

Hi Roger,
Not to beat this particular horse much more, but is it possible the assumed temperature differential is higher when rating sleeping mat r-values than in building insulation? And perhaps assumed watts (or BTU/hr) is lower?

Because, at least according the Wikipedia table here: http://en.wikipedia.org/wiki/R-value_(insulation)
even aerogel only manages 1.76 R per inch in SI units (R-10 imperial). Hard to imagine a Prolite 3 beats aerogel.

David,
You state "...even aerogel only manages 1.76 R per inch in SI units (R-10 imperial). Hard to imagine a Prolite 3 beats aerogel..."

uhhh...as you state:

1.76 R in SI units is R-10, Imperial
The Prolite is R-3.8, also Imperial, as it is a USA item and designation.

In the positive number set, 10 is usually greater than 3.8 ... at least in this universe.

Edited by greg23 on 05/09/2011 14:39:28 MDT.

Greg,
Sorry for the confusion. Unless I misread, I think the contention in the article, and in Roger's posted response to questions above, is that sleeping pads *are* rated in SI (including Prolite, speced R-2.2 or 3.8 for Prolite Plus). If the rating *was* imperial, that would seem more in line with building insulation materials.

Assuming pads are SI rated, I wondered if the R rating for pads assumed a greater temperature differential than the R rating for building insulation (converting units, of course). By R= m^s*C/W, that would give a bigger number for R, if cross-sectional area and watts are held constant. I assume there is some standard value for temp. differential and watts so that the R for different insulations can be compared.

My training is in architecture and the building trades--R values are something I've always looked up, and used to calculate heat loss. So my assumptions may be way off base.