Door insulation redux

[Garald]

2 minutes ago, SteamyTea said:

You can look at a list of k values and take a guess.  See how well it fits.

https://www.engineeringtoolbox.com/thermal-conductivity-d_429.html

 

 

Well, if we can talk of how it fits or does not fit experimental data (what data, and how?) then that data can itself be used to give us an idea about the k value, no?

[SteamyTea]

5 minutes ago, Garald said:

Well, if we can talk of how it fits or does not fit experimental data (what data, and how?) then that data can itself be used to give us an idea about the k value, no?

Yes.

You can set up a Solver in Excel to run though possibilities.

 

The only data you have is door area, temperatures and door thickness, plus a list of possible value.

 

 

[Garald]

3 hours ago, SteamyTea said:

Yes.

You can set up a Solver in Excel to run though possibilities.

 

The only data you have is door area, temperatures and door thickness, plus a list of possible value.

 

 

 

Right, but, if the only relation I have is the trivial one giving me heat loss in terms of those data, then I cannot conclude anything about which value is correct.

There's another data point: the temperature of the air in the room. I suppose I could set aside the curtain and, after a while, measure again the temperature of the door (which will then be in contact with the main air mass; the temperature of the door will necessarily go *up* when I remove the curtain). What I would then need to look up is how well transfers from an air mass to an adjacent steel surface; with *that* in hand (but where do I look it up?), I could estimate how well the door conducts heat, in comparison. (There will be three different temperatures: room air temperature, temperature of the inside of the door, temperature of the outside of the door.) Then I would not need to conjecture what is or isn't in the door - or, better yet, I could confirm a conjecture among several.

[Garald]

Wait, shouldn't there be a range of R-values for steel doors in metric units? I found a source which gave an R value that didn't survive a sanity check (see below), and a second look showed it was a US source. Of course it may be that the core of the door is not insulation but a honeycomb (which would be energetically stupid of whoever installed it; maybe they thought that, since this room was a laundry room back then, it didn't matter).

 

Even before I put on the curtain, the inner surface of the door was clearly colder than that of the windows (I can check this again, setting the curtain aside) which would seem to imply that its R was lower than theirs. (The only other explanation I could think of would be for heat to transfer much less effectively from air to steel than from air to glass (is that true? is there any significant difference? how can there be one?), but that would *help* the steel door on the other side, so the effect would be halved.) The R-value of the window is probably about 1.3, so...