Opening Statement

I went to the woods because I wished to live deliberately, to front only the essential facts of life, and see if I could not learn what it had to teach
~Henry David Thoreau, Walden


Thursday, December 29, 2022

Van Insulation - the Math Behind Staying Warm (or Cool)

A lot of factors go into staying warm.  My purpose is just to explain the math behind insulating a van so that the inside air temperature is comfortably warm on cold day.  Most of the math also applies to staying cool. We want to understand:

  1. Keys to construct a well-insulated space, and what is achievable
  2. How much heat is required to heat an insulated van
  3. How to provide that heat

Keys to construct a well-insulated space, and what is achievable

In conceptual terms the answer is quite simple.  Have as many well-insulated surfaces as possible, and as few poorly insulated surfaces as possible.  So what do we mean by a well-insulated surface?  Let's pretend that a van is a rectangular box with a heated space with dimensions of 6' x 6' x 12'.  This would be an enclosed space of 432 cubic feet, with a surface area of 360 square feet.  The 360 square feet is the critical number because heat crosses a boundary measured in square feet.  So the idea is to have as much as possible of that 360 square feet well insulated.

The most important formula for us to keep in mind is the relationship between insulation value and energy needed to keep the inside warmer than the outside.  Here is the engineering formula in concept:

[Heating load] = [surface area] x [temp. difference] / [insulation value]

You can easily see that increasing the insulation value will reduce heating load.  Double the insulation value and you cut the heating load in half.  In Imperial units, heating load is measured in BTU/hr, and insulation value in R-value.  If we add the all measurement units this looks a little more complex, but it's still the same formula.

[BTU/hr] = [Surface Area (sq. ft)] x [Temp Diff. (deg F)] / [R-value (ft2·°F·hr/BTU)]

So, if we want to reduce the heating load, we want to cover as much of the square footage as possible with good insulators.  In quantitative terms, this means materials with a high R-value.  R-value has a convenient characteristic.  When we make a composite material out of two materials we can just add their R-values together to get a composite R-value.  So a material with an R-value of 1 combined with a material with an R-value of 2 has an R-value of 3.

Here are the R-values per inch of  some materials typically found in a van build:

R-value per inch by MaterialR-value per inch
Rigid foam - Polyiso6.00
Industrial felt (used for a thermal break)4.00
Closed cell foam - Ensolite4.00
Wool - Havelock Wool3.60
Synthetic Fiber - Thinsulate3.30
Double-paned glass1.60
Plywood1.25
Single-paned glass0.91
Steel0.03

These are all quoted "per inch."  Thinsulate is typically installed at a thickness of about 1.5", so you can get possibly get about R-5 from it.  Don't always believe the optimistic ratings from manufacturers though--that's what they achieved in the lab, not in a real build where some of the insulation is compressed.  Compressed insulation will deliver less insulation value, but a van is so oddly shaped that you will end up cramming insulation into oddly shaped cavities as often as you are installing it in beautiful flat sheets.  Plywood paneling is typically installed at 1/4 inch, or a bit less, so it only typically contributes about R-0.3.

A well-insulated wall or ceiling could be as high as R-10 in places, but will typically be all the way down to just better than R-1 where plywood has been attached directly a steel framing member.  Windows are difficult to insulate, as are the doors.  Putting up close fitting thermal curtains and using thermal window coverings can cut heat loss at least in half for windows and doors, so that's a common approach to staying warmer at night.

So, in addition to using good insulation wherever possible, three practices can also make a big difference: use a thermal break of at least R-1 when attaching anything to the van's metal framework; don't leave any steel elements fully exposed to the heated space; and, put coverings over any poorly-insulated areas at night.

I'm not going to get into the physics of this, but there is a bonus R-value caused by trapping air.  If you can trap 1/2" to 4" of air in part of the build, you get about R-1 from the trapped air if it's static, so some of the van structure that is hard to insulate at least has some inherent insulation value.  Also there is a boundary effect of still air against the interior of the van.  It gives the effect of about R = 0.7 on the inside of the van if you aren't running a circulating fan.  The outside also gets some value from this boundary layer if the air is still, but you can't really count on that.

Based on the materials involved and real-world experience, even with your best efforts, it's very unusual to achieve an overall performance of better than R-5 for the heated space, and R-3 is probably much more typical.  Although we are a bit light on empirical measurements, it's probably fair to describe van insulation performance in five categories:
R-1:  just factory headline insulation and air pockets in a cargo van - aka uninsulated
R-2:  lightly insulated
R-3:  well insulated, but exposed windows and doors
R-5:  very-well insulated with treatments for all surfaces
R-7:  super-insulated van

So, with these insulation values in mind ...

How much heat is required to heat an insulated van?

Given our assumption of 360 square feet for the surface area of the heated space, we can calculate the heating loads for various outside to inside temperatures.  To make the table easier to read, let's assume a target of 65 deg F. for the inside temperature, and then we can use the outside temperature as part of the table.

BTU/hr by Outside Temp. to maintain 65 deg. F. internal temp., by insulation level
Outside Temperature (deg. F.)
BTU/hr by Insulation Level55453525155-5
R1: uninsulated3,6007,20010,80014,40018,00021,60025,200
R2: lightly insulated1,8003,6005,4007,2009,00010,80012,600
R3: well insulated1,2002,4003,6004,8006,0007,2008,400
R5: very-well insulated7201,4402,1602,8803,6004,3205,040
R7: super insulated5141,0291,5432,0572,5713,0863,600

So, to be comfortable at an outside temperature of 35 deg. F., with a van insulated to R3, you will need about 3,600 BTU/hr of heat.  And at an outside temperature of 15 deg. F. you would need 6,000 BTU per hour.  

How to provide that heat

There are a lot of ways to bring heat into a van, so I'll just mention three popular options with their capacities.  If you are somewhere you can plug-in to an electrical outlet you can use a typical 1,500 watt electric heater to provide heat.  The conversion to BTU's is Watts x 3.41 = BTU's, so 1,500 watts will heat at  5,115 BTU/hr, which would keep you warm down to below 25 deg. F. in an R-3 insulated van.  That might be the right answer when parked at home or if you happen to be staying at a campground with hookups.  Espar and Webasto fuel heaters are popular; they both have 7,000 BTU/hr models and some larger models also.  Propex offers a propane heater that puts out 6,350 BTU/hr and they also have a larger model.  This is far from an exhaustive list.

This post hasn't covered the whole topic, but it's a start at understanding the capacity of the heating system required, and some of the issues related to insulating a van.  It's also important to not get too starry-eyed by the R-value insulation rating of some insulating material, and keep in mind that a van is very hard to insulate to a high R-value due to the windows, doors, cab area, and the uneven shapes.

Conclusion

Take insulating your van seriously and you can get to R-3 without extraordinary effort.  Beyond that level you will be working harder at it than most DIY builders.  You can check the table above to see what size heater you will need at a minimum.  I think the models that put out 6,000 to 7,000 BTU/hr will satisfy most people, but if you live in a really cold climate it makes sense to get the larger model.  

No comments:

Post a Comment