Hi all, I’m back today not so much with a guide, but with a theory that we’ve kicked around on the forums before. As of a few days ago I actually published a video (the first in months) that covers what I want to discuss today. This post will be a continuation of that video, as we’ll be looking and testing what level of heat resistance we can expect by keeping a fire chest inside a container that would simulate a UL 350 class safe, while in a fire.
Let me break that down a little more clearly, here’s what I’m testing: Assume there is a fire, you have a safe, and inside that safe, is a fire chest. We are testing to see how the fire chest would do in this circumstance.
So how do we accomplish that? Put a safe inside a safe and have a bonfire? That would be nice if I had the funds. However I’m going to be doing a much more simplified version.
Hank Hill has got nothing on me.
SETTING UP THE TEST:
That’s right lads and lasses, I’ve got propane and propane accessories ready to go! This experiment will consist of a BBQ grill, a half hour rated sentry safe, one internal digital thermometer, one external digital thermometer, my neighbor believing I am crazy, and a ton of Pokemon related testing supplies to be monitored in the safe while it cooks.
Testing supplies are as follows:
- Single toploader (interior card sleeved)
- Snap case (interior card NOT sleeved)
- One touch magnetic holder (interior card sleeved)
- Screw down holder (interior card sleeved)
- Card saver one with the following interior configurations
a. Gaming sleeve
b. Double sleeved
c. Single sleeved - Stack of card saver ones single sleeved (3-5 cards)
- Gaming sleeve only cards
- Single Raw card
- PSA slab with no (exterior) sleeve
- PSA slab with (exterior) sleeve
- Individual booster pack
- An entire booster box
I took these testing supplies as a part of community feed back of things that might be kept in a fire chest. So lets set up our test and explain how it’s going to work, as well as the hypothesis I had going into this.
HYPOTHESIS:
If a one hour UL 350 rated safe’s interior only reaches 350 degrees at a maximum, then a (half hour) 1550 temperature rated fire chest should be able to last at a minimum four times longer (1550 / 350 = 4.428). Additionally the rate of heat transfer should not exceed an internal temperature to the fire chest of about 90 (350 / 4 = 87.5).
ANALYSIS OF HYPOTHESIS:
I must admit, my hypothesis operates completely in a vacuum, and does not deal with a multitude of variables. Also, the hypothesis only takes numbers that we have verified either through UL or an independent claim made by a company (in this case, sentry safe). The division I do is based off of a simple proof of math, not by thermal dynamics or any degree of thermal engineering. To be blunt, I’m optimistically operating off of numbers that are simple to apply without a full understanding of how heat and heat transfer works. But we have to start somewhere.
THE TEST:
Lets put the safe on the grill and take a look inside. As we can see, all of our supplies are ready to go and in the lower right hand corner we have a digital thermometer that will measure the internal temp as we grill. It has the ability to measure up to 158 degrees before it fails, in theory we won’t come close to this number.
After we close up our supplies, we can lock the compression handle, as well as the original locking system in place. Now we can place our external temperature probe outside of the case. She’s all set up, and ready to get fired up.
For this test I’ll be lighting the grill as quickly as I can and then shutting the lid to the grill to trap as much heat as possible so I can simulate how an interior safe might function. I’ll be keeping a close eye on my external temperature probe to monitor and adjust the heat level as needed. I’ll also record my results every five minutes so we can see how the temperature fluctuates.
DATA GATHERED:
The spreadsheet with the testing data is here. I would like to make a few notes here about the data and some of the issues I encountered while testing.
First and foremost, windy day to do this, probably way too windy. While I only recorded the results for every five minute increment, the level of fluctuation was pretty ridiculous. The wind would typically push the heat point down into the 280’s for much of the test and as a result I was forced to continue to add as much heat as possible to maintain as close as I could to 350. My average temperature was around 325, but that was a recorded average. Actual average was probably much lower.
Next, I would like to take a look at the length of the test. While I would have loved to have done a four hour test to really try the hypothesis I had in mind, I opted for a shorter one hour cook time with a roughly half hour cool down time. I did this for three reasons: 1. I didn’t want to grill until midnight. 2. A one hour grill time should have easily given me favorable results, and opened the door to more testing. 3. I wanted a cool down time as this seems to be more realistic with a scenario a collector might actually encounter.
I really want to hit on the cool down time here. House fires don’t operate in a vacuum, your safe won’t go from an 1100 degree inferno to ideal room temperatures in an instant. Cool down should be considered for anybody who is serious about this. Your safe is going to be really hot to the touch after a fire I promise you won’t be able to open it right away.
RESULTS:
The moment we’ve all been waiting for. So after a 28 minute cool down time I popped the grill open and took a look at the safe. You can see what it looked like here. Not super pretty but it doesn’t need to be. I found that the key I had still was able to be inserted into the lock, however the actual locking mechanism itself had failed, meaning the case was just unlocked. Not a huge deal, but it seemed odd the metal piece had failed. The compression latch was still in place… time to take a look inside.
I have a more detailed breakdown of the individual testing results on my YouTube channel here. But as you can see, it didn’t go as well as I had hoped. The internal thermometer had actually failed which tells me the internal temperature had surpassed 158 degrees. Upon further review of the case I found that the interior seal of the case had actually come right off some time during the testing. I didn’t see a trace of glue residue which tells me the case might not have been manufactured properly from the start. I also took a look at the underside of the case that was the only direct heat contact surface. You can judge for yourself how it did. Another thing I noted while filming is that the case has several small “holes” in the case itself I’m sure it has something to do with releasing pressure. When the plastic underside melted, those became somewhat see through, to the point where I think there may have been direct exposure. Pretty lousy design if you ask me.
CONCLUSIONS:
So back to the drawing board right? Well, not so fast there are a lot of things that we can learn from here. First and probably the most important thing to learn here is that heat transfer can’t be defined as a simple ratio of degrees. The more I look into it, it actually appears that the exterior temperature a fire chest or safe can take really does not matter all that much. It’s about how long a safe/chest can sit in the heat. From my experiment I can assume that a half hour fire safe is just going to last a half hour even in “low” heat condition compared to a fire. This will be a great starting hypothesis for another test.
The next conclusion we can make is that some Pokemon collectibles will do far better than others. I would have bet the ranch that the shrink film on the booster box would have been the first to go, and the booster pack would have been unrecognizable. Turns out they do just fine in high heat conditions. Also raw cards in gaming sleeves do very very well, where as anything in plastic does not.
We can also conclude that sentry safe may not be the go-to brand we want when it comes to reliability. Both a mechanical locking failure and a rubber seal failure are pretty critical things. To be fair, this was cooked longer than its intended use, but even so it had no direct contact with the flame, I would expect more from sentry safe.
My last conclusion is pretty simple: I need to do more testing! I would say we figured out what we screwed up here on test number one. We need a way to actively measure the temperature inside the case. Weather conditions really should be favorable. And a new brand should be tested…
Thanks for reading guys! I’ll be posting more test results both here and on YouTube so keep an eye out for those!
