No-one knows how explosions work (yet)

An update from Sam, the scientist in the video! He says: "It's been great reading through your comments and seeing people excited and intrigued by our research. I thought it'd be useful to answer some of the common questions that keep cropping up: 1. Yes, high speed video/photography of explosions is nothing new. They did this all the way back in the 40s and it's famously how GI Taylor estimated the yield of the Trinity test. Our high speed video is not what gives us the new scientific insight, but it does help give us a steer as to what is going on, and it's a fantastic way to demonstrate what an alien world it is so close to an explosion (in time and space!) 2. Our pressure measurements from the MaCE rig do give us the new scientific insight. If you want to see an example of our recorded data then please feel free to read our recent paper (link in description), Figure 5. We're measuring pressures higher than the strength of normal steel that are applied and removed in 50 microseconds. For comparison, a blink lasts >100 milliseconds, so over two thousand times longer! 3. Scientists know a lot about nuclear explosions, that's true. There, the energy is released effectively instantaneously and effectively as a point-source. With high explosives, the reaction rates are comparatively slower. That means we have an ongoing chemical reaction that changes if/when the blast wave and fireball come into contact with structures. The exact loading applied to said structure is a function of the pressure-volume-energy state of the fireball, so there's an intricate dance between the two (pressure and reaction rates). This is where current models break down. Yes, even the sophisticated physics-based ones. 4. We're only just discovering the extent of our ignorance, because until now we simply haven't had the experimental data to compare to. We do know that these secondary reactions are significant, which makes explosions (from high explosives) so difficult to simulate, because we've measured explosions in normal air and in an almost pure nitrogen environment. 5. Yes, the mottled/bumpy surface of the fireball as it expands outwards is a genuine physical feature, and seen at larger scales too (see photos of the "Minor Scale" test). When we're performing tests that we intend to publish we form our explosives into a 3D-printed mould, but even then we still see these features. 6. We filmed at 250 thousand frames per second for this video, but our camera can go all the way to 10 million frames per second. Thank you once again for showing an interest in our research. Cool, innit?"

It does make sense why it's difficult to observe what happens in an explosion up close

3:17 Tom: “What’s the use for this?” Researchers: “Uh, because it’s rad?”

I’m imagining the researchers pausing for moment when Tom asked what the purpose of this was and they had to fight the urge to say “because explosions are cool”

You can always tell it's going to be a cool video when we see High Visibility Tom Scott appear.

The best of these videos are when the chaperone is as giddy and excited about the tech as Tom is.

If you go to 5:06 and use the . button to go frame by frame, you can see that the ground lights up 1 frame before the explosion is visible. This is because the camera scans in "lines" of pixels top to bottom. When it was at the explosive material, it had not gone off yet, by the time it got to the ground, it had.

Tom is the only person that makes a saftey vest look like reasonable casual wear.

The actual data from this test didn't make it into the video (it's just a LOT of numbers), but there's a paper in the description for those so inclined! And this week's pinned-comment plug: the Technical Difficulties are back! A new season of four adventures from me, Chris, Gary and Matt over at youtube.com/techdif - one every Thursday.

Always amazes me that a scientist 100 years ago can think up theories that we can only test today.

I love it when a sciency person says, "We don't know what we don't know", it's both so true, and so honest.

Oh hey, this is something I actually do active research and development on! We’ve actually gotten pretty good at making computer simulations of these kind of explosions using multi-phase physics models. The Kingery-Bulmash model that Tom mentioned is still used in places, but it’s ridiculously simple compared to real simulations.

It’s amazing how many things there are that people had theories about how they worked 100+ years ago, and yet we’re either no closer to actually knowing or we’re JUST getting into it, like this

5:51 interesting to see the shockwave resonates the dust under the table. Wonder what impact this has on the signals within the data cables which run out from here and if a deflector would be beneficial

Nothing a scientist loves more than someone who obviously appreciates their work, you could see how chuffed he was at Tom's genuine amazement.

tom scott is THE cool guy that doesn't look at explosions

Sam was my lecturer in the blast protection module of my civil engineering degree. Fascinating subject and a great teacher.

I love how jealous Tom looked when he was told the camera recorded a quarter million FPS. Like, that was just as much a highlight for him as the explosion

The last two utterances in this video encapsulate why I watch Tom Scott videos: his genuine desire not only to allow people passionate about interesting things to explain them, but also to *get excited along with them*.

That feeling when you're just graduating from Sheffield and you didn't even realise that your uni was doing thing.