Insulation effectiveness tests

[Zeppelinium]

Styrene 1-1/2". This was a frivolity. It was just solid pink polystyrene foam with an aluminum panel glued (at the edges) on both sides. It performed wonderfully, but you'd never achieve this kind of quality, leak-free, conformal insallation in real life. But it does give one an idea of what good insulation can do.




Styrene 1/2". The above panel was opened and the interior 10x10" area was hogged out, leaving 1/2" of styrene against the inner panel. This was a more representative test, since the styrene had been installed in 3" wide strips, so now the slight gaps were able to contribute some leakage.




and now to make some conclusions...

[Zeppelinium]

Here's all the data in one chart. The 080322DA number is the year/month/day and file ID, so I can find the original data if you want to see it. You can have as much fun as I've had, well, NOT....



If you average the data, then sort it (left colum is based on heat loss, right colum is based on inside panel temperature), you can rank order the quality of the panels as shown here.



It should be no surprise that the rib just killed the insulating quality of the panel it was installed in. The real surprise is the variability of the Prodex panels with glass mat. Either the wind really affected these tests or it's true that the Prodex doesn't like to touch the wall--it may perform poorly when conduction is a big factor. Either way, the Prodex did OK. But for all of you who are not tearing out your inner shell walls and are keeping your glass mats, you should be happy with its consistent good performance.

That's all for now. Next tests will be to see how these panels do in the hot sun. Sunlight provides about 100 watts/sq foot of radiation, so maybe I can do some hot tests with a heat lamp and not wait for summer...

Let us not forget that Kip (AEROWOOD) made this possible with his generous donation of the 12x12 aluminum panels, way back before I knew him. Thanks, Kip.

Zep

[Aerowood]

Interesting numbers Zep. More food of thought

[Wabbiteer]

Using a wet-bulb type thermosensor along with OAT unit could help correlate the wind factor, if desired, though you may have to wait until next winter

Thank you for taking the time to maintain precision across the samplings - I've been waiting for your results!!

It looks like I will be trying to incorporate your 1-1/2" foam at least in the center run of the ceiling, I have a band saw and know how to use it to achieve the barrel-stave effect so it may get continued out to the ceiling edges too. I've installed tractor-trailer loads of ceiling tile and drywall back working summers in High School to it should be an easy puzzle.

I also have purchased silicone sheet to cut strips out of to be thermal barriers for the ribs - material identical to Aircraft Spruces' silicone cowl seal ribbon. I am planning on doubling or tripling up on the rivets at a minimum in case I am unable to get a solid clamp from them....

[markdoane]

Zepp,

Thanks for all your work.

I guess the original airstream designers weren't the dummies some of me thought they were. Sounds like glass mat worked pretty good.

[Jim & Susan]

Dang, I guess that makes me the guinea pig for Reflectix and 1" glass mat.

Thanks for doing this Zep. I don't fully understand all of the data, but that last chart helps.

Jim

[vhord]

Originally Posted by Zeppelinium

Next tests will be to see how these panels do in the hot sun. Sunlight provides about 100 watts/sq foot of radiation, so maybe I can do some hot tests with a heat lamp and not wait for summer...

Since I live and play in the deep south I will be very interested in those results! We do not have to worry too much about heat loss around here.

Thanks Zep for all your work and for sharing this with us.

Vernon

[malconium]

Zep,

My premise right along for saying that I thought reflective foil was a better insluation product than most was that it seems to be better at blocking radiant heat loss than the other types of products are. The other part of my premise is that the major component of heat loss through walls and especially ceilings is radiant. My thought then is to block the radiant heat loss and most of the job is done.

One thing that I am still a bit puzzeled about with your measurement technique is how you determine what type of heat loss you are measuring? It seems to me that if you have a box that has warm air on the inside that you are really measuring heat loss by convection and conduction. Do you think that your tests accurately reflect (pardon the pun) the heat loss do to radiant effects? That is the particular type of heat loss that I was trying to measure with my setup where I had ambient air on each side of the panel and used the sun as a radiant heat source on one side.

What do you think?

Malcolm

[Wabbiteer]

I hate using 20/20 hind sight (for those finished with reinsulating) but when conduction and radiation heat flow are addressed then convection ramps up in importance - there should be baffles placed regularly to keep airflow down inbetween ribs, like fire stops placed in wall cavities...

[Jim & Susan]

Originally Posted by Wabbiteer

...... there should be baffles placed regularly to keep airflow down inbetween ribs, like fire stops placed in wall cavities...

I guess there's still a lot of stuff I don't understand about the whole insulation issue. Are you saying that movement of air inside the walls should be kept to a minimum? As in, don't allow air from one portion of the wall to move to another portion thru normal heating and cooling?

Jim

[Melody Ranch]

I still believe in spray foam insulation....the correct kind.

Its been used, its effective, its light weight, it stays put, etc, etc.

http://avion.gradeless.com/1972Avion/1972_AVION_12.jpg

[ticki2]

Originally Posted by Melody Ranch

I still believe in spray foam insulation....the correct kind.

Its been used, its effective, its light weight, it stays put, etc, etc.

http://avion.gradeless.com/1972Avion/1972_AVION_12.jpg

I agree , now to find that " right kind"

[Wabbiteer]

Jim... to the best of my knowledge, cross-batten baffling is desirable as long as they are not made water proof "enough" to keep future water leaks from accumulating. It is the lack of air movement helps make the 'fill cavity' fiberglass style very effective.

[vhord]

I am just beginning to insulate my walls. Before I got too far I thought I would do some somewhat unscientific tests. I ended up with some confusing results.

In the following picture I have three areas.

#1 - No insulation
#2 - Reflectex over 1/2" foam strip spacers
#3 - Same as #2 but with a sheet of 1/2" foam overlaid



I borrowed two different handheld infrared thermometers from two of my A/C customers to attempt to test the effectiveness of the different insulations (surface temps). The afternoon west sun was beating down on this side of the Airstream at the time. To the touch it appeared that #3 was somewhat cooler tan #2 and that #2 was definitely cooler than #1 which is what I would have suspected. However with the infrared thermometer it indicated that #1 and #2 were very close in temperature and #3 showed to be the hottest! I am not sure what to make of that. One of the A/C men mentioned that these thermometers are influence some by the color, texture, and etc. of the surface being tested.

Is there a better way to test surface temperatures?

[vhord]

After further research I think I know why the infrared thermometer did not show the results that I had anticipated (I have a very limited amount of knowledge in this field!). After reading this I discovered that I would need sensors with different settings to get an accurate reading from the different materials.

I suspect that the handheld thermometer that I was using was going through the insulation and reading the temperature of the outside skin. The bare skin #1 released its heat into the inside where as the insulated #3 was actually holding the temperature against the outside and not letting it enter the interior.

I guess it was doing its job. I just need to figure out how to measure the inside surface temperature. Maybe just holding a thermometer against the different surfaces would give me some idea but not exactly accurate. Whatever I do I will have to wait for the afternoon sun to test it.

[richinny]

can you mount a small piece of aluminum on the test surface and measure the temp of the metal? the mounting would have to leave the back of the metal bare.

[vhord]

Duh. That sounds like it might work. Is this what you had in mind? Now just to wait for the afternoon sun to heat up.

[richinny]

ummmm, maybe a needle and thread? top of the metal up & down so gravity holds it from slipping?

thin piece of duct tape in place of threads?

paper clip with one end hooked into insulation and other end holding clip?
... just hit the reading away from the clip.

[malconium]

Vernon,

I think to get the most accurrate result you need to have the small piece of aluminum in about the same relationship to the rest of the wall materials as it would have in the actual assembly. For the bare aluminum (#1 case) you could use some foam spacers around the edges to supend it about 1-1/2" from the outer skin. That would simulate the normal air gap with no insulation. For the foil only sample (#2) you could use some more foam spacers around the edges to suspend the small piece about 1" from the foil. Is the foam in case 3 right next to the foil or will it be right next to the inner skin?

At any rate it is important to simulate the air gaps and to keep your test piece of aluminum out of direct contact with the other layers in the same way that it would actually be in the final installation. It is also a good idea for any air gaps to be closed up around the edges so that ambient air can not get in behind and cool the small piece of aluminum from the back side.

Malcolm

[vhord]

First I would like to make it clear that I am not at all attempting to improve on Zep's excellent research on this topic. (I wish I could but I am not capable) I just wanted to make some practical application tests to see what I came up with.

Anyway I was not real pleased with my results but I am going to share them with you in case someone else might be able to get something out of them.

I took Malcolm's advise and spaced my test pieces away from the outside skin to simulate the finished product. I sealed the edges with tape. The one in the picture that is peeling away was re-taped. My test pieces were small (4"x4") and that may have hindered any true readings. On samples #1, 2, & 4 I cut a 3-1/2" hole behind the aluminum sample to provide the proper air gap.



#1 - no insulation
#2 - Reflectex only
#3 - Reflectex and foam board
#4 - Foam board only

To measure the temperatures I used an infared temperature gun and also a temperature pen. Okay, for what they are worth here are the results that I observed at 3:30 p.m. yesterday afternoon with full sun beating down on the outside wall.

Inside temperature - 89 deg.
#1 - 95 deg.
#2 - 91 deg.
#3 - 92 deg.
#4 - 93 deg.

I was surprised that there was not a larger difference. I also observed that the ribs next to #2 were 113 deg. and the ribs next to no insulation were only 107 deg.

Outside skin temperatures
#1 - 113 deg.
#2 - 120 deg.
#3 - 123 deg.
#4 - 120 deg.

I checked the temperatures again at 4:30 p.m. and got about the same results but slightly higher temps.

The Reflectex certainly was reflecting heat but I do not know why it did not make a bigger difference on the inside. I am still going to use the Reflectex but these results made me think that I was mainly baking the outside.

My experiment was sort of a bust. So much for a shade tree engineer. I think that I need to leave that for someone else.