Altitude Chamber: Violent Implosions

Shizznit's gettin' real yo. Today I got lucky and avoided injury when I had a violent implosion of a 55-gallon steel drum in my basement lab. Kids, don't try this at home!

There's a reader quiz at the end so stay with me...

Let me back up a few steps. While waiting for delivery of some additional components for the HAPP structure, I continued working on a vacuum chamber to simulate high altitude conditions. The idea is to test fire the 3D-printed jet nozzles in low atmospheric pressure and measure the resulting jet force, which will be different than the force produced at low altitudes. The controls software needs to know the jet force as a function of altitude so it can accurately control the HAPP's rotation at any point during the flight.

Previously I found the most accurate way to measure jet force was using a rotating platform with a well-known moment of inertia. This means the altitude chamber must be large enough to hold the rotating test apparatus. So my thought was to use a standard 55-gallon (208 liter) steel drum, hang the test apparatus inside, and run live fire tests after pumping out as much air as I can before the drum implodes. Why, you ask, would a steel drum implode? Answer: Because the sides of the drum get squeezed with over 38,200 pounds (17,300 Kg) of force once the air is pumped out!

Air at sea level is at 14.7 pounds per square inch (PSI), which means the net pressure on an empty drum is 14.7 PSI. That may not sound like much compared with your car tires, for example, but the problem is not the P. The problem is the SI. The sides of a standard steel drum have a total surface area of approximately 2600 square inches. So 14.7 x 2600 = 38,200 pounds of force on the sides of the drum. Anyway, I figured I could run tests at various "altitudes" and go as far as possible before the drum started making strange sounds, at which point I would back off.

Of course I needed a transparent, removable cover to insert the test apparatus and make observations during the tests. After doing some math I decided a 1.25" thick sheet of cast acrylic plexiglass would suffice.

This cover also needed an airtight gasket of some sort, which I custom-fabricated to match the steel drum's rim. This was easier said then done. After lots of tinkering I finally came up with the idea to use a router to cut a circular channel in the face of the plexiglass, then fill it with liquid silicone that vulcanizes at room temperature. Voilà, instant gasket.

Clockwise from top left: Router and protractor on plexiglass slab;  Pouring RTV silicon into the trough;  Peeling away protective masking;  Attaching hangar for monofilament;  New compact test apparatus;  Steel drum;  Apparatus hanging from plexiglass cover.

The other issue was how to control the test apparatus once it's inside the vacuum chamber. Data is logged to the onboard micro-SD card, but I needed some way to give commands, especially the commands to start test runs and then null out any rotations between tests. My solution was to enable WiFi-based control of the Arduino on the apparatus and "drive it" from my smartphone. I found a great app called Arduino Manager from a guy named Fabrizio Boco that greatly accelerated this process (thanks Fab!). After downloading the app I had the following control panel up and running on my iPhone in literally half an hour. You can read the labels on this screen shot and get an idea of the various functions I created.


iPhone controller made using Arduino Manager

I rigged up a standard vacuum pump and I was ready to rock. Being the (mostly!) sensible type I ran a series of tests at different simulated altitudes, starting with sea level and gradually evacuating air from the drum in steps of 2.5 inches of mercury (the units used by my pressure gauge; about 1.2 PSI). The rig was performing flawlessly and the iPhone controller was working great.

I was wearing safety glasses but I kept my ears exposed despite the noisy vacuum pump; I wanted to listen for any hint of complaints from the steel drum so I could back off the vacuum if needed. I assumed I would hear something as I approached the failure limit. 1500 meters altitude... check. 3300 meters... roger. 5500... five-by-five. 8400 meters... good to go. 10,300 meters... all systems nominal. Then BANG! No warning, no creaking metal, just a violent implosion and a 50-pound plexiglass sheet airborne in my lab. Duck! The carbon fiber air tank charged at 4500 PSI was chipped and is now unsafe to use, but fortunately it didn't rupture. All kidding aside, that could have been catastrophic.

Below is what's left of about two weeks of work. I don't mind rebuilding, but I'm really pissed I did not capture the implosion with a GoPro at 240 frames per second. You, dear reader, could have had great fun measuring my reaction time as I jumped back liked a scared cat. Sorry to deprive you!

From clockwise at top left: Imploded steel drum;  Contents barfed out of the drum;  Pressure fitting sheared off;  Bits and pieces of the apparatus & nozzles. Glad those red LiPo cells didn't go nuclear!

From clockwise at top left: Imploded steel drum;  Contents barfed out of the drum;  Pressure fitting sheared off;  Bits and pieces of the apparatus & nozzles. Glad those red LiPo cells didn't go nuclear!

READER QUIZ: Calculate the crush force on the steel drum given the simulated altitude of 10,500 meters at the instant of crush. First correct answer posted to the comments below wins... well, probably something cool. Someday.

What doesn't kill us makes us stronger. I shall return to this phase of the project, and it will be bigger, faster, and stronger than before. We can rebuild him. We have the technology...

Would this man quit after a minor disaster? No!