Why Railroads Don't Need Expansion Joints

A friendly overview of thermal effects on railways.
💡 Go to brilliant.org/PracticalEngine... for a 30-day free trial & the first 200 people will get 20% off their annual subscription.
🚆 Watch the entire railway series here: • Railway Engineering
Errata: At 9:00, the left column of calculations is incorrectly labeled "SI." It should be imperial. Whoops!
Just as all materials have a mostly linear relationship between temperature change and length change, all materials also have a similar relationship between stress and change in length (often called strain). And this is part of the secret to continuous welded rail: restrained thermal expansion. You can overcome one with the other.
Watch this video ad-free on Nebula: nebula.tv/videos/practical-en...
Signed copies of my book (plus other cool stuff) are available here: store.practical.engineering/
Practical Engineering is a KZbin channel about infrastructure and the human-made world around us. It is hosted, written, and produced by Grady Hillhouse. We have new videos posted regularly, so please subscribe for updates. If you enjoyed the video, hit that ‘like’ button, give us a comment, or watch another of our videos!
CONNECT WITH ME
____________________________________
Website: practical.engineering
Twitter: / hillhousegrady
Instagram: / practicalengineering
Reddit: / practicalengineering
Facebook: / practicalengineergrady​
Patreon: / practicalengineering
SPONSORSHIP INQUIRIES
____________________________________
Please email my agent at [e-mail]
DISCLAIMER
____________________________________
This is not engineering advice. Everything here is for informational and entertainment purposes only. Contact an engineer licensed to practice in your area if you need professional advice or services. All non-licensed clips used for fair use commentary, criticism, and educational purposes.
SPECIAL THANKS
____________________________________
This video is sponsored by Brilliant.
Stock video and imagery provided by Getty Images, Shutterstock, and Videoblocks.
Music by Epidemic Sound: epidemicsound.com/creator
Tonic and Energy by Elexive is licensed under a Creative Commons Attribution License
Source: • Elexive - Tonic and En...
Video by Grady Hillhouse
Edited by Wesley Crump
Produced by Ralph Crewe
Graphics by Nebula Studios

[@PracticalEngineeringChannel]

🎅There's still time before Christmas to pick up an infrastructure bingo pad or a shirt at store.practical.engineering 💡 Check out Brilliant for the best way to learn math and science skills: brilliant.org/PracticalEngineering

[@michaelmayhem350]

It was a long 24 hour wait to comment here after watching your video on nebula. I wanted to add something important to trains that you didn't mention (probably because us trains are so archaic it doesn't matter) but when metal gets hot not only does it so all the tings you mentioned but it also experiences reduced conductivity. This why computers perform better with better cooling, less resistance to the flow of electrons means more electrons flow. I would imagine engineers in China & Japan have to consider this with their all electric trains.

[@paulmichaud3230]

Is that an electrical bonding strap? 0:57 If so, why?

[@youtubejawn]

haha you said nut

[@josephvanname3377]

But does Brilliant have a course on reversible computing?

[@ExiSTEIN]

@@paulmichaud3230 For Signal System!!!🙋‍♂🙋‍♂

[@mvcrailphotos]

As a engineer who works in railway track design, one thing I'd like to point out - especially since you mentioned a video on signaling will be coming soon - is that neutral temperature is typically high not because a broken rail is less dangerous than a sun kink. Yes, its more likely that a train could operate over a pull-apart rather than theough a sun kink, but largely because the signal system will detect a pull-apart as a discontinuity in the rails and thus the signal circuit - automatically turning the signal ahead of that track to red!

[@mikefochtman7164]

I've always been amazed at how railroads developed signalling systems. And how the idea of 'fail-safe' design was involved waaaayyyy back in the early 20th century. Broken rails, loss of power, relay failures were all considered and will set signalling 'safe'.

[@TSBoncompte]

ooooooooooooh, right, cause if you use disjointed rail you have to have a separate cable for the signal to travel on, at least in every joint. so even if pointy joins solved the expansion issue the added wires and cases and little welds sounds like a lot of extra work.

[@Tupsuu]

Im interested to know how that works. Im pretty sure I saw bare copper used as grounding for the rails last time I paid attention to it. Or have I missed something

[@mvcrailphotos]

@@Tupsuu It depends a lot of the specific railway line, that could have been traction power return cables, grounding, equalizers between adjacent tracks, signal infrastructure, etc...

[@Torchedini]

@@mikefochtman7164 That fail safe system was paid for in blood though. But since rail is that old most of the accidents aren't that highly publicized or live in recent memory like development of the airplane and space flight. It is kind of interesting to see how safety culture around cars is so low, but that is I think a question of high amount of owner operators/low barrier to entry. On rail most things are owned by large corps, which has similarities to aerospace as well.

[@petepi]

"Stress is what breaks things" is a truth that more people should apply to their own lives

[@FlyTyer1948]

Well said.

[@MrMadsci7]

Much truth

[@andrebartels1690]

Absolutely

[@silversilk8438]

What about eu-stress and the joy of overcoming challenges like winning a game of fuse ball or soccer/football?

[@janami-dharmam]

Stress causes strain and it is the strain that we cannot cope (find hard to manage). Stress is the cause and strain is the result. What is feel is the strain (the deformation in real life from what we are used to).

[@stevenrielly240]

As a trained metallurgist and engineer, I found this amazing.... what a way to encourage future engineers, Simply, softly spoken.

[@arctr00perecho]

Definitely! I am currently studying mechanical engineering at university and just finished taking courses on materials of engineering and strength of materials. It's awesome getting to see practical real world examples to the theoretical content covered in class. It's why I love Grady's content. I also did a research assignment on the development of the Hoover Dam in an introductory class to engineering and got hooked on all his dam analysis videos (proof reading this right now and find the ending unintentionally quite punny)

[@sandgrownun66]

So does CWR not need a breather switch, also known as an expansion joint?

[@andershenriksen6997]

Over the last few years here in Denmark, we've experienced some hotter and dryer than normal summers. We've had our cases of public trains being a bit more like a rollercoaster. We'll move up and down, from side to side. Oh, well. Sometimes, clack-a-di-clack might make you feel more safe; at least if you feel turbulensse in a plane, there're miles to the ground, ie plenty of time to find new winds. In a train: bang, the whole side is torn open because of a passing train, or you tip over - crash!

[@TonyP9279]

@@andershenriksen6997 Fortunately, rails can clickity-clack for a long time before it becomes unsafe, especially since some of that comes from parts other than the running rails, like check rails, brake trip mechanisms, conduits and parts associated with the transponder sensors, and other mechanics that make signalling and interlocks work. Of course it depends on the rated track speed: If it's high-speed rail, say ~200 km/hr, you don't want to hear any click-clack.

[@jarekferenc1149]

Same education, same opinion. I teach students the thermal expansion, also from the viewpoint of phase transformations, and I think this vid is an excellent example of application of knowledge. Knowing this is a must!

[@TheRealFumigator]

My 14 yo son was telling me about metal expansion he'd learned about in school and we got onto the topic of railroads. I mentioned they don't really put railroad down in sections anymore and he asked how they accounted for expansion. I realized-- I had no idea! So thank you for this great video-- I now know.

[@Paul--88]

Just on derailments

[@gregyjoebatin2558]

As a civil engineer, this type of content is therapy

[@skeetrix5577]

your quite the nerd

[@WinstonSmith0824]

And that's a great thing.

[@skeetrix5577]

I watched about half of it until I got bored and clicked away

[@deandennis2838]

I’m an electrical engineer and I find this therapy also.

[@Pingviinimursu]

I'm a software engineer and I find this therapy also

[@pschlik]

Fun fact: The problem with 'rock n' roll' (more formally, harmonic rocking) is bad enough that rulebooks disallow travelling at certain speeds on jointed rail. The rulebook I have says this danger zone is 13 to 21 mph, so if your train only has enough power to reach 20 mph (entirely possible, if normal, these days) safety says you'd have to slow to 13 mph on jointed rail. So CWR not only makes maintenance-of-way's job easier, it also makes operating the trains a bit easier and safer too.

[@erich1394]

r e s o n a n c e

[@royreynolds108]

@@erich1394 I think you both are correct. As a civil engineer who spent over 20 years in track design and work, the term "harmonic rocking" is the one used by the railroads with certain speeds over jointed track with alternating low joints. It is the alternating low joints that cause the problem not the joints by themselves. There was a lot of research done to determine this phenomina.

[@miinyoo]

So I think your point is; stay out of the resonance zone or transient through it as quickly as possible. Go slower or faster so you don't build up a resonance that can, not always but can go exponential. Blown mind all these little details that add up.

[@EnjoyCocaColaLight]

So.. faster than 21 mph is fine... less than 13 mph is fine... anything between 13 and 21? Immediately catastrophy! Like how certain hz may cause bridges to collapse (and therefore people shouldn't march over bridges) etc? Fuq'n science always messing things up >:(

[@Trump2Prison]

@@royreynolds108 Also bogey (or track) hunting. I forget which car type, one had a both a lower and upper speed limit when unloaded due to excessive gauge resonance that lead to $many derailments.

[@smartereveryday]

I really enjoyed this video! Well done!

[@richardnavratil9661]

Oh, my two favorite educational channels! Destin, Grady and Sam from Wendover would make a trifecta of digging into a problem/topic!

[@ParallelPages]

Love your videos too Destin

[@derekkimball6662]

Builder here. South Florida. I never understood why rail work here was always done in the summer heat! Thanks for clearing that up. We never see rail work in our “winter”.

[@Whammytap]

Spread the word! It'll make everyone smarter, and less resentful about working in the heat. ;)

[@plat2716]

@@Whammytap I imagine the workers know this already lol.

[@Whammytap]

@@plat2716 OP's comment seems to suggest otherwise, but they may be an outlier. Or I might have misconstrued the term "builder," I don't know the jargon real well. 😅

[@xavierburval4128]

I love that the primary solution for buckling is just “do it on a really hot day lol.” It’s the kind of thing that sounds way too dumb to work but is somehow the most genius solution ever.

[@craigslist6988]

for engineers it's obvious... really obvious. Honestly I think his explanation was misleading.. the direction of thinking is backwards. As an engineer you would think of wanting to tension the bar first, because that's an obvious well known way to prepare for thermal expansion. But the obvious problem is that as it cools the tension will increase. So you figure out your safety factor'd tensile limit, calculate how much strain (lengthening of the rail) that max tension will cause, then plug that into the themal expansion equation which will give a delta-T, which is the maximum range of temperature the rail can be tensioned to neutralize expansion for. This is the basic solution to the problem, but I will say there's a whole lot of work wrapped into "decide on the maximum tensile force" part. But moving on, since that temp range apparently isn't a large enough number to cover the range of temperatures we see outdoors on earth, they are stuck having to design the rail to survive some compression also, which again the obvious solution is by 'pinning' points down to avoid buckling. So avoiding buckling that way, when you've come at the thinking the direction I'm describing, is basically an afterthought. I'm skeptical there aren't a lot of other nuanced things involved in this solution too, but those are really getting into engineering weeds. But that's just the reality of engineering. The reason this wasn't always done or done earlier wasn't for lack of 'genius' engineers but lack of practical ability to do it. I wish he had explained more what that was because I can only make educated guess(es). Probably it was a series of factors that all had to be solved over time.. such as the quality / consistency of the rails, and the difficulty of welding pieces together well enough that they don't just instantly break on a cold day. Most people don't understand how complicated "welding" is.

[@charleswalters5284]

That's what i do installing 'floating floors' and such

[@sirsancti5504]

And then,you get 3 years to conpkete 200 meters..

[@TheKillGameR8351]

Honestly I though about doing it on hot days, and during all the video I was wondering what was the problem with that? All to find out it was actually the solution. Maybe I should have worked as a railway engineer

[@noop9k]

It is kinda obvious, not dumb at all.

[@MikeV8652]

If anyone wonders why the odd figure of 39 feet was the standard length of single sticks of traditional jointed rail, it was so that they'd fit into 40-foot gondola cars. Welded rail is now typically manufactured in long lengths (1,320 feet in the USA, which is ¼ mile) and can be welded in the field to be continuous. Such long rails are transported on special rail trains, like the one Grady shows half-loaded at 0:25. The yellow racks hold them in place, and the rails bend as the train goes through horizontal and vertical curves 😮.

[@royreynolds108]

Actually the modern rails are rolled in about 30-meter lengths and then welded into the quarter-mile lengths to be transported to the laying site and then welded by a portable electric welding machine or thermite welding process in track.

[@MikeV8652]

@@royreynolds108 The fabricating of quarter-mile lengths is part of the manufacturing process. I intentionally avoided saying that the rails were rolled that long.

[@ferky123]

The rails are only "held in place" in the middle. They lay on the rest of the structures.

[@MikeV8652]

@@ferky123 True that they're fastened down only at the middle, but the racks surround them to keep them aboard and untangled, to guide them to bend around curves, etc.

[@thomaswalters4365]

Didn't know that about the 39'. Makes a lot of sense. Had to transport the rail to the site somehow back then. Gons were 40 foot.

[@trainsbyben]

People get stressed sitting at a crossing waiting for the train to pass by because they have places to be. I get that. Me, I marvel at the engineering that isn't seen by the casual observer. The whole system from concept to application is mind blowing and has more facets that a jewelry cut diamond. Always enjoy your videos.

[@johnandrus3901]

If I remember correctly, what my associates in the track department told me is that in the Northeast where we are, they had to lay the rail at a certain temperature range. This is to mitigate the expansion and contraction which happens in our area because of the swings in temperature across the seasons. When our company was bought out, the new owners decided to lay rail in freezing temperatures and they ended up paying the price. Obviously, in Florida, as you mentioned, they lay the rail at a high temperature for a reason. My knowledge of rail only encompasses running my trains over the tracks and also how the shape of the rail works. Your video was very interesting and informative.

[@Sam-bn9ix]

Railway engineer here! A few points: Not all rail is made of the same alloys, particularly high load areas such as points and crossings. These use harder metals to reduce wear, such as manganese. Adding this into the equation means welding plain line to p&c is a particularly complex job. I am also glad you mentioned track circuit will be covered later on as that is a big downside to CWR with older signalling systems. Aluminothermic welding is spectacular when it goes wrong, search it up!

[@whuzzzup]

So let me ask a question what I wondered about often. So the pictures of buckling rail is because those rails were not installed properly? For example laid in cold winter without heating or applying pre-tension? And/or were the ties in the past worse in keeping the rail straight? Is it common for rail to be heated/tensioned in mild climates when a new track is built or does this only happen in more extreme climates like Texas?

[@Sam-bn9ix]

@@whuzzzup if installing CWR most of it needs to be installed and stretched to the length at the stress free temperature rail length. I.e in the UK all tensioned rail is stretched to it's equivalent length at around 27 Celsius. It's also important to note that this is the rail temperature not air temperature so it may not be very warm but if it is a sunny day the rail will become a lot hotter than air temp. This is when buckling occurs as the rail becomes very hot and has nowhere to move. Majority of this happens when a vehicle is running over the rail. As engineers know this we implement critical rail temperatures to ensure that beyond a certain temperature we know the rail is expanding we can gradually implement speed restrictions etc. to reduce these risks to public.

[@thomaswalters4365]

I can see diagonal tapers being used in between blocks to isolate block signals from each other.

[@iainstenhouse8399]

@@thomaswalters4365Those are expansion joints. I am as yet to see one used in that application.

[@salerio61]

Hi Sam, I was a railway engineer too back in the days of BR, worked out of Derby Research. I missed the part where he mentioned track circuits, but I am very interested to see the American slant on it.

[@Itslvle]

In Finland (North Europe) our temperature range from summer to winter is quite high, from -30C to +30C (-22F to 86F) is to be expected. If we did the “wait for the hottest day” trick, our tracks would snap in the winter from the enormous tensile stress. Instead we do it in the summer at night, as rail that has sunlight shining on it can easily reach 80C (176F). We haven’t used wooden railroad ties in over 50 years. Ours are always concrete, weighing 240kg (529lbs). Railroads buckling doesn’t happen because this is over-engineered enough so that compressive forces can always be overcome, but we can’t overcome the inherent weakness of steel under tensile stress after a certain point without making the rails ridiculously thick. Fighting buckling is economically much more viable if the temperature changes are high enough.

[@MarkRose1337]

In my part of Canada we range from -40 to +40, but I imagine our rails don't get much hotter in the summer sun. Concrete ties are very rare here, with ties almost always made from wood. I don't know what they do to mitigate the thermal stresses.

[@jballaviator]

I mean here in Mid continental U.S. we vary from -10F - +110F. Practically the same. Safety always takes a side seat here.

[@micnorton9487]

EXACTLY,, that's the way railroads need to be built in America too but corporate profit keeps over- engineering from ever entering the picture... Like the bridge in Minneapolis that fell into the Mississippi River from a combination of no redundant features and simply overloading the bridge... A stupidly cheap design to begin with,, where failure of any one part may have only resulted in failure of a section of the bridge but with shoddy maintenance and overload from numerous layers of roadway just laid over the old stuff, and it was like the world trade center... People with no knowledge of engineering like to say about the world trade center, there's no way jet fuel can melt steel and while that's probably true, the fasteners didn't have to actually melt to cause failure... the same as the bridge in Minneapolis didn't have to fail all at the same time, the under engineered overloaded structure ensured that with any extreme stress, the fasteners would snap along the whole length of the bridge... Kinda like the Towers,, those buildings were designed to minimum stress mostly associated with wind,, and with the added negative factor of using mob concrete, the internal structure of the building wouldn't even handle one floor breaking out of its place and falling on the next, because it's just not something you engineer for...... Now of course it is,, but I hope for this new world trade, they didn't use Donald Trump's dad's concrete company lol...

[@micnorton9487]

​@@MarkRose1337.. I think CN still has a lot of old style rails on some of it's track,, CN trains almost never derail because of superior maintenance while Burlington Northern tracks, I'm in Minnesota, are derailing almost all the time... Not in any one place, I just mean somewhere on the Burlington system there is probably one or two derailments a day.... And that's just Burlington Santa Fe RR corporation... They were also the line that dumped six or seven tank cars of benzene into the nemadji river in Superior Wisconsin about 15 years ago too...... The entire city of Superior and half of Duluth Minnesota had to be evacuated...

[@darkwing3713]

There's places in the US where the average range is -15F to 115F (-26C to 46C) with very strong sun. And they don't use concrete ties. The risk is mitigated by applying very large amounts of money to political campaigns.

[@rayyoon6265]

as a non-engineer minded person just wanted to say thank you for making this very easy to digest and understand. i truly appreciate learning from people who are able to follow the KISS rule lol

[@rayyacht4342]

BRILLIANT BRILLIANT BRILLIANT!!! Years ago I asked myself how it's done. A friend told me, expansion is controlled by rigid counterforce from the bed. But now YOU really opened my mind by detailed background - technically and by examples. Very good a teacher you are. English is not my mother tonge but I could understand anything. Thank's a lot and You have a nice life.

[@pyrokaren]

I developed a broken rail detection system at San Francisco BART that used ultrasonic waves traveling miles in the continuously welded rail. As part of characterizing the rail I measured acoustic properties as a function of rail temperature. It was amazing to see the rail temperature exceed 140 degrees F on medium hot days, like air temperature in the 90’s. The peak temperature was in the late afternoon when the rail web caught most of the sunlight. Definitely painful to touch!

[@greenredblue]

"Don't touch the 3rd rail, it's hot." "What do you mean? They're ALL hot!"

[@dwc1964]

As a BART rider, I thank you for your service

[@craigslist6988]

engineering any equipment that is outside in the world day and night makes you aware of temperature. You probably didn't have to think of this, but your counterpart designing rail lines through Montana has to take into consideration that at night it can be -30 (and lower with wind chill on metal rails..) and then get up to 130 the next day. That'll mess with your tolerances a bit.

[@pyrokaren]

@@craigslist6988 I vaguely remember hearing that over the temperature range in the bay area, a mile of unrestrained rail will change in length by 7 and a half feet.

[@eugenetswong]

That's amazing work. Thanks for your contributions.

[@chadportenga7858]

In the foundry industry, the reverse has to be dealt with - shrink. When we build patterns (tooling) for foundries, we have to account for the shrinking that occurs when the molten metal cools back down. Usually, it can be predicted quite accurately, but on occasion, it throws us a curve ball.

[@erich1394]

That sounds like a really interesting problem! How do you predict it? Are we talking like... computer simulations or basic algebra?

[@riccardoorlando2262]

@@erich1394 Well, for very regular shapes you can easily predict the cooling with essentially the formula that was shown in the video, applied along each dimension independently. More complex shapes might behave unexpectedly, though

[@johnhunter7244]

​@@erich1394ideally when the part completely cools down it is the same shape, just a bit smaller. The transition where it is not homogeneous temperature might have some non-linear effect which requires more complicated math or computer simulation. I'm not exactly sure. I'm sure it also depends on the geometry of the part. If you're just casting a ball, it should be super easy, but if you're doing something long and slender with an unsymmetrical cross section it might be much more challenging.

[@Saakk129]

Ok…?

[@christrunzo7305]

I would imagine that differential cooling based on the size and volume of a particular shape could lead to wracking in many cases.

[@kustom4935]

While studying mechanical engineering in college and now as an actual locomotive engineer for a commuter railroad, this is great information and very well presented. I’ve witnessed CWR being installed but your explanation of installing rail at its warmest neutral temperature makes perfect sense! Indeed, a rail pull-apart is much preferred over a heat-kink, not only from a rail perspective, but also the cab-signal code will drop to zero in a broken rail but may still function normally during a heat-kink which could be disastrous.

[@stripervince1]

Excellent info. I started my RR career in the 1970s, I was a signalman and then conductor for almost 5 decades. I remember workin out in the desert doing cadweld bonding and at 8am and 70°, there was a half inch gap in the rail, in one hour when the sun came up and it was 90° the gap would close with in minutes. Those oblique slider joints are fantastic, they prevent sun kinks... good video

[@Yikes5824]

This is the first question I ever asked on the “internet” - a Compuserve message forum - back in 1989, and it was answered by a Canadian rail worker. That’s when I knew we were in a new era.

[@Jablicek]

It was an interesting time back then in so many ways. I really loved seeing both the difference and commonality we all shared. That wonder at the new is gone now - our kids see it as perfectly normal.

[@davidbroadfoot1864]

Does Canada have railways? I thought they didn't even have roads.

[@kisaragi-hiu]

Rail is fascinating. I never considered that (fundamentally) you could just hold it in place and force it to not thermal expand and the stress isn't always prohibitively high.

[@julianemery718]

Ehhh, isn't that quite a risk though?

[@RCAvhstape]

​@@julianemery718Thousands of miles of daily operations suggests it works pretty well.

[@notbob555]

​@@julianemery718this video literally covered that question.

[@GilmerJohn]

Well, it still thermally expands. But it gets wider and higher but not longer. The width is also slightly constrained but the sky is the limit on height. The "returns" aren't really in whether this length & width constriction is good or bad from a metal fatigue point of view.

[@erich1394]

I feel the same way about the compressibility of hydraulic fluid vs water. It's just difficult to intuit whats going on when the forces get bigger, you know?

[@TheTrock121]

Love your teaching style. It's been over 40 years since I took Engineering Mechanics and truly enjoyed your presentation.

[@kendraowen2841]

My uncle was the engineer on the train the derailed in Crofton KY which resulted in white phosphorus burning. It was caused by bucked rail, a warning sign that had been knocked over or never put up, and a missing warning on the orders.

[@InvestmentJoy]

Awesome! I've seen thermite rail welding before and was curious behind the why's of it.

[@galvanizedgnome]

911 conspiracies are not welcome here.

[@NavajoNinja]

Used to do thermite welds. Works on 2" gaps, its basically adding molten metel between 2 pieces of hot metal. Usually the thermite burns at 1400° F

[@JimAllen-Persona]

Love watching thermite welding.

[@jceggbert5]

​@@NavajoNinjabasically volumetric welding?

[@Vulporium]

woah, cool seeing you here. I subscribed to you many years ago, I hope you're well.

[@fantabuloussnuffaluffagus]

I had always assumed that the expansion in CWR went into the curves. It never occurred to me that the track just loads up like a big axial spring.

[@PetesGuide]

You can actually hear it sing from the pressure of oncoming trains before you can see them!

[@jimonthecoast3234]

it's interesting when that force, breaks it's constraint, and huge buckles appear in the rail.

[@mvcrailphotos]

It's a well-known phenomenon that railway curves will "breathe" due to temperature swings too: in warmer seasons the curve radius is slightly larger (sometimes by inches) than in colder seasons as the rail expands and contracts.

[@EnjoyCocaColaLight]

@@jimonthecoast3234 Myes.. interesting

[@agilemind6241]

@@mvcrailphotosSorry but citation needed. Because the length of the rails is different between inner and outer curve rails, if those rails did expand due to thermal expansion it would not be an equal expansion between the two rails -> the outer rail would expand more. Thus risking warping the alignment of the rails and derailing the train.

[@stepheneyles2198]

This is something I've always wondered about but never got around to looking it up! Thank you Grady, for explaining it so well and concisely! 😊

[@DrDave-cp8vq]

I love how you came back around to neutral temperature at the end since you got me first thinking about it with the demonstration at 4:00. Great video!

[@dallassukerkin6878]

I am a SCADA engineer, working in the electricity and water distribution fields, and the first thing I became aware of when I started work was that those things we take for granted to keep civilisation running are *much* more complicated than most people know. It seems that railway tracks are no exception! :)

[@testales]

When I first drove from East to West Germany in a train, I stood in the middle of a crowded wagon but the moment we crossed the border was obvious as click-clack instantly stopped and there was instead a nice continous sound. Everybody was surprised that this is possible.

[@sabersz]

The joys of Continuous Welded Rail haha

[@user-yp7xd7gr6n]

As a summer student in 1970, I worked for Canadian National Railways on an extra gang laying ties, track and ballast on a branch line. One task was repairing a sun kink by assisting in the cutting of a section of rail and replacing the bent piece with a straight one. CN soon after started installing continuous rail on the main line. For the last 50 years, I have wondered why this was viable given the absence of expansion joints. Thank you for the excellent presentation - my question is answered.

[@bicivelo]

When I had the baize, AKA felt/cloth, replaced on my pool table the installer sprayed it with a thin mist of water to introduce the most amount of moisture to will ever experience. So, when stapled down it dried and tightened up the cloth putting it under tension l. So, no matter how humid it gets in the house, the cloth won’t get wrinkled keeping gameplay very consistent. I think this is very similar to installing railroad tracks on the hottest day possible to keep them under tension as well. It’s so cool how materials respond and react to their environments! Awesome video series!!!! 😊😊😊

[@jimwebb9328]

I spent a couple of years in Buenos Aires in the mid 70s. The rail system there had been built 50 years before by the British and then nationalized in the 40s by Juan Peron. Very little maintenance had been done since then. There was one rail trip of 200 km that I took regularly. Leaving Bs As the train moved at 50-60 mph (my estimate) and the clickity-clack was spaced out, indicating long lengths of rail. But there was a point beyond a certain town where the speed decreased to a relative crawl and the wheel noise was almost constant. The ride quality deteriorated to the equivalent of driving on a rocky dirt road after exiting a paved highway. It seemed like the rail sections were only about 10 feet long.

[@PapaDalbec]

My man, promoting math to the masses. We need more folks like you!

[@fredinit]

Check out 3B1B (3Blue1Brown). You may not understand the math, but you'll love how it's explained and demonstrated.

[@ethananderson4348]

This video came at the perfect time. We just started studying column buckling in university, and I brought up sun kinking in rail lines. Very interesting video!

[@stefchemacrae5540]

You've got such a talent for explaining engineering concepts simply, and your demonstrations are fantastic. I wish my undergrad professors had possessed the same commitment to ensuring students understood things intuitively as well as mathmatically.

[@Merennulli]

I am stunned by this. I knew rail was difficult to lay in general, but maintaining strain at the right tension for the whole length is incredible. Where I park for work is very close to a rail facility of some kind (I honestly don't know what it does) so I am frequently hearing train brakes whail and trains pass, and the clicking is almost non-existent. I just assumed they were more precise or something with the leveling of expansion joints, but this is amazing. For the people who are going to weigh in on my ignorance of the rail facility beside the parking garage - it's about half a kilometer from the passenger rail station, but regularly has both freight and passenger vehicles that are definitely not meant for local unloading. It has a small spur that rarely gets used that crosses the road where passenger trains pull off onto and then almost immediately back off of. My assumption is that it's some kind of switching facility, but I don't really know.

[@cyan_oxy6734]

Rail is still far easier to lay than build streets.

[@Merennulli]

@@cyan_oxy6734 Rail costs 5x as much per mile. The problem with streets is the vehicles on them are far less efficient.

[@roger5059]

@@cyan_oxy6734 if you compare roads and rails with the same amount of tracks/lanes, the rail will be much more expensive, but will also have larger capacity and the maintenance cost per capacity can be lower.

[@sublivion5024]

@@Merennullistreets and railways are fundamentally different things - streets are supposed to be walkable and places for business, leisure etc while railways are transport corridors. sure a street’s central road can be used as a transport corridor but that is only a side effect (and usually undesirable).

[@sublivion5024]

@@roger5059 a rail track has the same capacity as up to 20 lanes of road, so yeah significantly more expensive but the potential is much greater (and it’s a lot better for the environment and people’s health, and these things are costs too even if certain governments want to ignore them)

[@jordanwhitecar1982]

As a train buff who's talked to some railroad engineers, another thing that's sometimes done (at least in california) is put an intentional slight wave in the rail (a few inches to each side every few hundred meters) this does allow the rail some side to side action when heated or cooled. Instead of preventing the buckling from happening, it's given a controlled outlet

[@paulheywood5205]

In Victoria (AUS) the 'high speed' passenger trains that normally travel at 160kmh (100mph) are restricted to 95kmh (60mph) when temperature reaches 35*C (95*F) for fear of track buckles!

[@rabbit9464]

I run a machine that lines the tracks, and I'll admit that I have done this before if there was already a sun kink and I had to do something with all the extra rail until it could be cut and re-stressed. It has always worked out great for me, and I wondered why it isn't a more commonly used method in a pinch. In a perfect world we would use our laser-lining system every time to align our tracks for long stretches, so any bit of rail movement can be immediately spotted. In the real world, giving the rail some emergency room to grow seems like a good idea.

[@wmg33]

Meters? Bro this is America. We measure in feet

[@DaTT78]

@@wmg33 In Canada, this is one of the few industries that still use the imperial system...mostly.

[@charleswalters5284]

​@@wmg33you mean USA ?

[@1471SirFrederickBanbury]

I loved going on the east coast mainline from London to York a few months ago, but one of the best parts was seeing Bullhead Track(!) in York station. I’d never have guessed that I would see chairs and keys on rails in the 21st century, but there it was, in 60’ sections, leading right into the railway museum. It’s still used on heritage lines though, and with great success.

[@williamcolley4182]

You would have been everyone's favourite teacher. I'm sure you have inspired lots of people with your engaging style. More power to you my man

[@jacksonbuckner5756]

What a great video explaining sun kinks! It might be fun to know, in the world of Swiss narrow gauge, we actually have another way to manage the stress in our rails: "Bogenatmung" or "curve-breathing." Essentially we allow our narrower curves (below a radius of 120m) to expand and contract by a few centimeters over the course day. In other words, we minimise stress by allowing for a controlled lateral deformation in our track geometry

[@sharkheadism]

Thanks for mentioning this, I wondered if that was a plausible anywhere

[@erich1394]

Railway engineering is a gift that keeps on giving! Every time I feel like I've learned all the basics, there's a whole new thing. Love it! Thanks for being a great teacher

[@daisei-iketani]

I have zero maths background but thoroughly enjoyed this video! I always wondered why certain stretches of my local subway have the clickety clack sound while other stretches are perfectly smooth. As for the bullet trains, they have a perfectly smooth hum for the entire run where passengers ride! Thank you for posting this info!

[@Scubongo]

As a metalworker, I've asked myself this question so many times in my life, and you finally answered it. Thank you so much for that! I had no idea.

[@robertlucht4657]

Back in the 1890's at the Salt Creek Oilfield in Wyoming, the built one of the first cross country pipelines. That pipeline was not buried, but rather it was just laid on the surface. It worked fine until the first winter. When the cold of winter came, they had over a dozen places where the pipeline parted, caused by thermal contraction. That was 4 inch threaded pipe with 8 round threads, which pulled apart in the cold. I enjoyed your presentation I always wondered how the railroads overcame the thermal expansion problem.

[@richardsolberg4047]

Trans Alaskan pipeline overcomes that with jogs in the line every so often .

[@Silver_Lloyd]

Always love these videos, My Dad's a civil engineer and I've driven him mad with questions about every last thing I could see out the car door window. He eventually got completely sick of me asking, I'm thinking of getting him your book for Christmas but I worry it might be a bit redundant for him- pretty sure I'm more enthusiastic about his job than he is lol.

[@IshtarNike]

Buy it for him and tell him it's his present to you and he can Venmo you by 12 midnight before it becomes boxing day. 😊

[@JimAllen-Persona]

Hey, my f-I-l is a Chemical Engineer and he loved the book.

[@just_some_commenter]

Engineers love redundancy, go for it!

[@AlfredBarron]

excellent video! Always interesting to hear how other engineers get around problems in clever ways - I hadn't expected the idea of only installing on hot days!

[@ronstar8857]

I can say that I, am a wanabee Engineer. I find your videos very educational as well as entertaining. I have always loved working with my hands and re-engineering things around my home and business. The world owes so much to engineers, such as you, who dedicate their lives to making things better and most importantly, safer for all of us! Thank you!

[@BarderBetterFasterStronger]

I can say that you, don't know how to use commas.

[@paulturner5769]

Back around 1965-1968, British Rail built a test track near my school in Nottingham, England. It was the world's first pre-stressed railway line and astonishingly quiet for back then. As I recall, the story was that the rail was treated to expand thickness-wise rather than length-wise, allowing lengths of rail to be welded together. Maybe that was mis-direction or maybe it was a patented technique so a work-around was found by other railways.

[@Blockbuster2033]

Nah, it's the same thing. If you put compressive force on a material in one direction, it will shorten in that direction, however, it will expand in the two other directions. The reverse is true for tensile force where the material will stretch in the direction of the force, but get "skinnier" in the two other directions. So on a continuously welded track, instead of the track getting longer when heating up, it will develop internal stress and it will also get "thicker". The effect is really really small tho, and has no effect on the overall performance or design of a railroad track, that's probably why it wasn't mentioned in the video.

[@janami-dharmam]

when it expands, it expands in all direction. The expansion is proportional to the length and hence the expansion in the thickness is rather small.

[@angelikalindenau943]

Britain and railways? Those were the days...

[@janami-dharmam]

@@angelikalindenau943 I have traveled by the british rail in the late 70s; also on the tube during the same time. Some people were so friendly and some were so racist then...

[@gregwaters7830]

When it expands due to heat, being a homogenous material, it expands in all directions equally, but if it shrinks or stretches due to co-linear forces, it gets wider when compressed and skinner when stretched@@janami-dharmam

[@JarrodBaniqued]

I can’t wait to see the videos on signaling, communications and electrification for railways. I think the cherries on top of them might be explanations of why ERTMS and 25kV 50 Hz AC are the gold standards of railways. Keep up the good work, Grady!

[@TheBackyardChemist]

50 Hz AC is extremely convenient if your entire grid is 50 Hz AC. Just install a few transformers and you have traction power! No need to rectify it to DC or convert the frequency.

[@LordWaldema]

cries in 15kV/16.7Hz

[@lightningdemolition1964]

I can't wait for the video. Why does it matter the frequency for traction power?

[@ReddieSysteem]

ERTMS is not the standard yet, but in the future it will atleast allow trains from different countries in Europe to cross the border without any effort. Most rail networks or trains in Europe do not work so well in other countries. It will also vastly improve signalling and reduce maintenance and faults.

[@royreynolds108]

@@lightningdemolition1964 If the traction power frequency in AC is different from the mains or grid then it will need to be changed just as changing AC to DC as well as the voltage if DC is used. There are several industrial (mining) railroads in the US that use either 25K or 50K AC 60 Hertz (our grid frequiency) for traction power. The Northeast Corridor (Amtrac) traction power is 11K VAC but is 25 Hertz so the 60 Hertz has to be changed to 25 Hertz. This was done in the 1930s well before modern AC systems. The electrification of the commuter trains out of San Fransisco will be 25K or 50K AC.

[@BillDavies-ej6ye]

8:48 In the table, there two system of units names. However, SI signifies the International of Units (Système International d'Unités), i.e., metric units. The US system is named Customary Units, often but wrongly called the Imperial System, which are British units defined by Act of Parliament in and introduced from 1826. The US standards were derived from various earlier English standards.

[@Tvimadur]

I was wondering the same. Maybe it was supposed to say "non-SI" on the left side and "SI" on the right but later mixed up for some reason.

[@zuzik7714]

​@@Tvimadurtechnically right side also isn't SI :) Celsius are not in SI tables - but since there is deltaT math checks out ;p

[@peterwilson4546]

Started building track 51 years ago. Great video ! A few things left out are the use of rail anchors in tracks with standard tie plates, pandrol cliups as well as well as others do the same job. Ballast shoulders are an important consideration to prevent sun kinks.

[@SnowTiger45]

I thermite welded ribbon rail for the CNR for 2 years (1981-82). We did lots of de-stressing and also eliminated many reoccurring heat kinks by installing cement ties with pretzel clips which allow for the expansion and contraction of the rail without dislodging the ties. Every time a thermite weld was done (we were doing 12 to 14 per day), the temperature of the track was recorded and the gap between rail ends was determined (calculated) prior to placing the thermite mould. Sometimes we had to cut a section out of the rail because of expansion. Sometimes we'd cut out a section and then find we'd have to cut out even more. Sometimes the blade got jammed into the rail due to the heat expansion. Other times we had to force the rails away from each other and then weld the rail joints together. We used a hydraulic clamp pusher/puller to manipulate the rails. Subsequently, I ended up Joint Welding in my 3rd year with CN to repair old rail joints that were battered by the train/car wheels. As well I welded Magnesium Frogs which are the cross-over switch rails (Shaped like an "X"). They'd have to be built-up with weld then ground to the right camber, roll and smoothness. It was all interesting work and paid very well.

[@juhmickm4410]

You might already be familiar, but rail (track) structure interaction analyses are a fascinating task that bridge engineers need to complete when designing longer spanning rail bridges that utilize direct fixation rather than ballasted track. It’s a fun nonlinear finite element analysis that considers the structure, the rails and the connections (with nonlinear springs to capture when the track slips from its hold down clips) to make sure the rail isn’t over stressed.

[@Hyce777]

Great explanations! Was genuinely waiting for the joints that we do still have to not be covered but you nailed it. Cheers!

[@avengingterrier3244]

I've lived in both Texas and in Michigan (where we get very cold winters) and have seen gaps in rails, and being an amateur engineer, have wondered how this worked - thank you for explaining it.

[@kevinwallis2194]

When i was doing fire sprinklers for a living, we had what we called expansion joints for out pipes when going between sepperations in building, but it was more for earthquakes, not temp.

[@quetzal4042]

In July 2002 I was on the Amtrak Capitol Limited when it derailed just north of Washington due to a buckled track on an extremely hot day. I could feel the train lurch to one side just before it derailed. Subsequent reports said the track had moved 18 inches out-of-line. You'd think the Northeast Corridor, of all rail lines, would have been properly engineered!

[@josephastier7421]

Or that they would know the risks posed by very hot days and inspect their tracks.

[@joepatriot363]

Please read my comment and you will realize what took place.

[@ps-ic8pm]

I used to ride the MARC commuter train to DC, using the same tracks as you on Amtrak, using wooden ties (the Amtrak lines up to Baltimore and points north used concrete ties, like our friend from Finland above, and never had these issues). In the last few years I rode about, 5-10 yrs ago, CSX who operates my line had a rule (for passenger trains only!) that if the expected daily temperature increase was more than 20F then they would impose heat speed restrictions on the ride home from work, which happened frequently in the summer. I guess they didn't care if freight derailed, as you can't get sued by freight :) Regardless, hearing or seeing texts warning of heat restrictions always got a collective groan from the riders. I often said that the B&O railroad, the oldest in the USA, hasn't figured out how to operate trains at speed when it's hot out or the train can derail, is ridiculous!

[@quetzal4042]

@@ps-ic8pm Right, I realize now that route isn't actually part of the Northeast Corridor until you get South of Union Station. The derailment was just North of Union Station on a freight line running northwest to Pittsburgh owned I think by Norfolk Southern. So I guess that explains the poor condition of the track. But as far as speed I don't think we were going more than 15 mph when it derailed, although supposedly the engineer saw it and tried to stop but not in time.

[@LaplacianDalembertian]

CWR is extremely costly. I guess somebody likes to take a short path and reduce the costs where it's impossible.

[@JohnWick-cz1ti]

As a medium carbon steel rod I find this content traumatic.

[@aussietaipan8700]

Here in Melbourne Australia, our rail system has gone under major upgrades with many level crossings removed. One of the other upgrades is continuous rail and replacing wooden sleeper with cement sleepers. Many of the old wobbly rail routes are now quiet and much smoother.

[@SignalLightProductions]

As a locomotive engineer who flunked out of civil engineering, this answers a lot of questions. I only deal with jointed rail but have always wondered about CWR. It being in tension makes sense. Thanks!

[@Frank-mn4hx]

Tis is a good video about the construction of the tracks. And here in The Netherlands they use Thermal welding to join the sections of track together. The rails are each about 25 meters long when they are transported. And we us a special machine to place the tracks on the sleepers. And we first used wooden sleepers. There are still wooden sleepers in use. And we use solid concrete sleepers. Also we use sleepers in the shape of a dumbell. Meaning 2 concrete blocks under the track and these blocks are connected with a thick metal tube. This is done to safe of material and cost. It's just as good but, more easy to install and has a lower weight so, a person like me can move it if needed.

[@woodwhittlingbushcrafter7278]

I acctually work for a railway company in engineering builing tracks, I've been thier for awhile and I'd have to say i love this content. Keep up the good work

[@buentaste]

I am sooo glad I found your channel. I am very determined to build my own train track. Not a long way just about 80 foot long and all this information is super helpful I would used expansion joints instead since my track does not need to be extra strong or worry about clicking.... clickety clack . Thanks 🙏

[@harkeeratsingh2509]

I am not an engineer but I use railways daily. And I have learnt something new today and I loved it. Thanks for the amazing knowledge. ❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️❤️

[@FakeMaker]

Same. I use trains and trams every single day to commute and travel. This series made me appreciate all this infrastructure and all of the people who make it possible a whole lot more!

[@dan725]

LOL hahaha as a Korean, I totally was going “칙칙폭폭! (chikchikpokpok!)” at the beginning of the video in my head. Then all of a sudden, you actually say it!! HAHAHAHA!!! I was NOT expecting that! LOL!!! Growing up, my little brother and I would say the same thing when a train would pass! What another amazing video! Your videos also inspire vibrant comments that’s just chock full of knowledge all over as other people working in the field start chiming in! Grady, you’ve fostered such an amazing community here! Your channel is such a wonderful treat for my family! THANK YOU, and THANK YOU ALL other commenters here dropping these amazing supplementary information in the comments!

[@nakinajay]

I ran a holland buttwelder from Winnipeg to Armstrong Ontario. That machine was definitely something else. Worked for 5 years then I got tired of living out of suitcase working from Ontario all the way to peace river Alberta. The rail is supposed to be in neutral stress or compression at 90* and we have a chart to determine rail movement over a certain distance of free rail. It’s all dependent on temperature. After 115 degrees rail temp we were not allowed to weld or cut the rail. I’ve seen the track jump so much that it ripped the rail saw off its mount. When the rail is cold it pulls apart and when it’s hot, it pinches the saw. The railroad is so interesting

[@raymitchell9736]

Clearly explained as always. I like your narrative form how you take us on a journey and you breakdown the engineering that goes into the system, in this case the railroad. We take it for granted as a thing, but I didn't consider that the railroad technology is ever evolving, they just look like railroad tracks... until you look closer! And I like that you explain what the problems are and how they are being solved.

[@IsawU]

As a continuous welded rail with a sun kink, I can confirm this is accurate.

[@prozmystery]

One thing I want to appreciate about Grady's videos are the choice of words he uses which makes it so easy to understand even from a non-civil engineer's perspective.

[@drkashik]

Great video! I love that you included the math behind why this works. I'm a prodigal chemical engineer by education and I've always loved the elegance of a mathematical proof of a practical physical application.

[@olmynuwen]

This is something I have often wondered about. Thank you for explaining it so clearly.

[@GermanZindro]

8:52 the left side should be marked "imperial", not SI, since neither Fahrenheit nor psi are SI units.

[@SidewaysCytlan]

Yeah, that confused me. Maybe "SI" means something different in this context? I have no idea, but it's definitively not in the International System of Units. Edit: Seems like it's just an honest mistake! It's noted in the description.

[@Arkantos365]

Thank you! (I searched that comment, I also ticked seeing SI and °F)

[@dropshot1967]

You seriously have a knack for explaining enginering problems and their solutions in a way non engineers can understand (at east the principles) Thank you.

[@dbayboyds409]

Loving the balance of maths and practical demonstrations. You’ve really nailed it. Thank you for another engaging video.

[@tobybrowne7906]

I’m loving this train series keep it coming! I’m a train driver and I’d always wondered how the track could be laid continuously. We do get blanket speed restrictions on hot days though and I have seen rail let go and look like spaghetti.

[@KaiHenningsen]

Another thing that's relevant in that space is that many high-speed rail lines use ballastless or slab track, which is usually a bed of concrete that holds concrete sleepers to mount the rails on. There are a number of variations, including elaborate constructions to reduce noise. They are estimated to last about twice as long as ballasted tracks but are rather more expensive to build or repair. Obviously, that way, it's easier to ensure the rail is perfectly straight and has the right amount of longitudinal stress. As an aside, high-speed signalling systems can't use track-side indicators, as the engineer will not have enough time to be sure to notice them, and needs too much distance to stop the train then anyway. So signaling systems for these trains are all using radio or other electromagnetic systems, and instead of trackside masts with optical indicators, the relevant information is shown in the cab on a screen (and enforced by the on-board computer: too much speed, it may just stop the train; and pretty much all restrictions are about top speed: the equivalent to a stop signal is "keep it slow enough so you can stop in the distance left ... slower ... slower ... max speed is now zero"). Over here in Germany, the speed at which high-speed traffic rules start to apply is 160 km/h, around 100 mph. Level crossings are also forbidden then. We have two high-speed signaling systems: the old, locally-developed LZB (line train control, so named after a cable going through the middle of the track), and the newer European ETCS (from level 2 - level 1 is for slow tracks and uses traditional signals).

[@merlinbass8870]

Very good! Sun Kinks a real phenomenon. In mountain country generally on curves. Enough to laterally displace concrete tie track structures. Cold weather rail contraction also can result in the rail pulling apart. I've seen it break in multiple places in severe and unusual cold temperatures for a given district. Old timer Engineers used the rhythmic 39' clickity clack as a subtle gauge of speed and distance.

[@kenoliver8913]

You don't want even a tiny sun kink when the train is travelling at over 350km/h. I'd lay odds an awful lot of thought and an awful lot of testing in all seasons had to be done before the first Shinkansen was built.

[@keithfriedlander7699]

Superb and comprehensive explanation which is neither too dense nor simplistic. Well done!

[@TheTalkingMoose1]

This is an incredible video, so succinct, so well spoken. You inspire so many people with your thorough and thoughtful videos.

[@AdmiralJT]

As a civil engineer designer (in oil & gas field), always excited to learn about things outside of my normal work flow. Train related things are always cool.The SA sewer lift was cool too, our W/WW team is just over the wall so I hear alot of stuff. Weird being in this line of work and not liking math, thank goodness for spreadsheet calculations lol

[@Saakk129]

Wow, such amaze

[@TexanMiror2]

Fascinating how complicated something so "simple" is. Thanks!

[@richardlewis5316]

Having spent many hours on trains I certainly appreciate the lack of clickety-clack!! I often wondered how the expansion and contraction was catered for . I wrongly assumed there were sliding joints similar to your old/new joint. Its good to have found a video which explains it! Thanks

[@papanga1197]

I already passed my steel design course but that demo on the Pcritical formula just changed the way how I looked at it. Thank you engineer!

[@ammarmustapha8049]

Great video! It would be interesting to also show this effect when the railway is placed on a long viaduct, an effect usually referred to as track-structure interaction. We don't always place a rail expansion device on bridges as they are very expensive to install and maintain. Instead, we analyse the effect of the bridge expanding and contracting under the track (yes it can get quite complex!) and then assess the additional stresses from this effect. We also analyse the additional stresses in the rail due to the braking and acceleration (and vertical) forces from the train on the bridge, which invokes the foundations of the bridge and eventually impact the rail stress. I've worked on this for more than two years and I keep learning new things!

[@hannahranga]

Interesting, the RR I work for has a few bridges with them.

[@jeanladoire4141]

As a blacksmith and metalworker, it's always nice seeing different uses of steel. When i was making grids and frames, heat was a major factor in achieving precision. The precise order at which you do the welds will move your steel around, and pull or push/buckle your bars. Welding tends to pull as the metal cools down, i think it would help against the buckling of rails

[@martybishop8484]

That’s an interesting insight. So your shrinking welds work a little like scars, maybe? A healing cut can pull and tug your skin in small, uneven ways. That’s what makes a scar so noticeable, even more than the line of fresh scar tissue. (I think that’s why patients with visible cuts typically see plastic surgeons for revision after healing. Even if the original cut was perfectly patched, the recovering tissue will shrink and warp at the edges, just as you describe in the welding process.) Now I can’t stop picturing a nose job done with a welding rig. Mask down, sparks a-flyin’. Celebrities the size of Lady Liberty getting an overhaul to carve a couple hundred pounds of copper out of the ol’ schnozzola. Hammer out that crease between the eyes, while you’re at it. A little Bondo® around the lips, please?

[@jeanladoire4141]

@@martybishop8484 uhh yeah i suppose that's similar. But we also learn that heating up a bar of steel that has been compressed or held into a frame, will shrink and pull your whole structure. The reason is, as you heat up it expands, but of course applies pressure, just like shown in the video (in the press). Except at some point, with high enough heat, the steel just gets compressed and actually becomes shorter, losing the elasticity. So when it cools back down, it actually wants to keep its compressed state, and as it shrinks it pulls the whole structure with force.

[@Gottenhimfella]

@@jeanladoire4141 That's because when it's compressively restrained lengthwise, the extra volume of steel arising from heating it can only go into width and height. When the temperature returns to normal, the steel shrinks in all three axes, including length which will end up shorter than it began (whereas the width and height will remain increased, albeit less than when hot). This is seen more clearly with sheet or plate steel, if the heating is done quickly in local spots, to cherry red. When they have cooled, you can feel the bump on both sides of the plate where the "upsetting" of volume has thickened the steel locally. In this instance, the restraint is exercised by the surrounding cold metal (no need for a frame). This technique can be used to correct distortion arising from other processes, including welding, by shrinking in carefully chosen locations.

[@bradlevantis913]

Really enjoying this series. Inside a commuter train a fair bit and they switched to the continuous rail a few years ago. It is interesting to see the engineering involved in that switch. Side note, I don’t miss the noise but I do miss the smell of the treated wooden railroad ties on hot days. (they use concrete now)

[@JeffMTX]

Creosote :)

[@maestrovso]

The first time I pondered about this was riding the Shinkansen in Japan. That was the first time I notice their tracks has no expansion joint. I search the web as what is the secret behind this to no avail. I got it immediately as soon as you started to talk about modulus of elasticity. The most impressive thing about these rail system is the engineering structures go into withstraining the rails from buckling, especially the extremely high speed train in Japan and other countries. Brady's channel is one of the smartest on KZbin.

[@RailRide]

From my YT explorations, it appears the Shinkansen lines use slip joints like those shown at 12:50. This I learned from a video from a Japanese hobbyist trying to replicate the unique sound of trains crossing these joints...with his model trains (he got close, but was no small feat owing to...well, size and materials).

[@RandyK1ng]

Grady, thanks for another great episode. This series on the rails is exceptionally timely, given all the difficulties being experienced across the country with failing components. By no means does it absolve the railways of their responsibility for solid monitoring and maintenance, but it gives us all insight as to the challenges facing them in doing so. There's nothing like "sunlight" to expose the issues and get the right people to take appropriate actions. :)

[@namrnam5413]

2:38 i didn't expect to listen to the sound of a train in Korean. The sound ChickChick Pok Pok (칙칙폭폭) is indeed the sound of train, but it is equivalent to the steam engine train sound. Although the word is from the sound of steam sound, still we all use that word for train sound. Thank you for the wonderful video! And i really appreciate using the metric systems

[@user-cw9qn1nb2n]

In the UK that rail-joint sound is often mimicked as Diddley-dee, Diddley-daa. And going over pointwork is diddley-diddley-diddley-diddley-dee, diddley-daa! Hope that makes you smile.

[@PatricioTorre]

Excellent content as usual. Practical engineering is key to see how all these equations that show up during your studies solve real life challenges in the most clever way. Congratulations and keep it up!

[@Fig.220S]

This continues to be one of the most valuable channels on all of KZbin. Thank you, Grady

[@SgtLion]

The high neutral temperature thing is super cool, I would have been a long time thinking of that. Having the rail typically in tension also helps prevent it bend from other forces too. I wonder why they don't use that technique for more installations that have to deal with thermal expansion? Maybe significantly more complex to simultaneously apply heat/high tension and install say, a bridge girder.

[@ArneChristianRosenfeldt]

We tension steel inside concrete.

[@royreynolds108]

@@ArneChristianRosenfeldt Yes, for prestressed concrete members. The stresses in the steel rods for prestessed or even poststressed concrete are 160,000 to 190,000 psi. This is to keep the concrete cracks from opening up. All concrete cracks, you just don't wnat it to come apart.

[@Real_Tim_S]

A comon hazard in engineering is (literally) overconstraint. If you oveconstrain a problem it can become intractable (unsolvable). A lot of having a good experienced engineer, is the experience part. Knowing that for say a sidewalk an expansion joint is the correct solution for the problem for a material that doesn't like tension but can deal with compression. For something like a bridge grider, managing expansion becomes cheaeper than constraining expansion - those expansion forces need to be transfered to something, and in the case of a bridge it's the piers and foundations at each end. Creating a bridge pier which is strong enough to constrain the expansion of the bridge girder would make the pier huge and expensive. Enigneering is all about making the equation ballance out - and there are a lot of variables, not just loads and life-cycle, but cost and installation complexity. This goes for every industry that uses any type of engineering (structural/mechanical, electrical, chemical...).

[@stvrob6320]

It wouldn't be practical to constrain a bridge girder in this way. A rail can be restrained to the ground every foot or two. A bridge girder needs to span some distance to be useful.

[@Kandralla]

It's the wrong tool for most applications. The thing you have to remember with railroads is that: 1. The structure is entirely exposed. 2. The owner of the track knows exactly how many axles have passed over a section of track and how much weight each of those axles had on it when it did. 3. The thing that ages the track is well controlled (a little side to side stress and a whole lot of compression/expansion cycles from a single direction). They tolerate the additional stress on the structure because they can figure out how long they can wait between inspections, which are easy to do because it's literally just lying there on the ground. If a section needs replaced it's a relatively quick and easy operation. For them they're trading shorter rail life to get longer train life. Doing this on a bridge, or a building, where lots of the structure is hard to get to, has been subjected to unknown loads and cycling, and for which repair would basically remove the structure from use for a long time, is a much different problem... and what would the benefit of drastically increasing the cost of the build and reduction of life be? What would the neutral temperature even be in something that experiences a wide range of temperatures across it's volume? It's not worth it, you just design so the building tolerates the movement and there are points in the structure where the expansion loads can be relived and once the dry wall is up no one gives a second thought to the fact that the structure is a bunch of different elements that move with respect to each other depending on the day. Similar situations occur with cars and airplanes. Weight comes at a premiums on an aircraft so you design the thing to just barely handle the loads and spend a lot of time thinking about how long things will last and how you can make impending failures detectable before they happen. For a car, where a couple extra pounds has little effect on the end product you over design a lot of stuff and move on to more important things than min/maxing weight trades.

[@richvail7551]

Excellent explanation. Been on the rails since 1990 and this is the first time learning how CWR works. In short, it’s a better product if it breaks when cold rather than hot. I’ve seen both issues happen and the choice is obvious when you see yourself heading towards a sun kink vs hearing the sound of the broken rail as you drive over the break. It’s horrifying to see yourself heading into wobbly rail cause you have no way of knowing if you’re going to make it alive.

[@raven1378]

Hey Grady! Love your vids. I always learn something from them, and you do a very simplified explanation of how things work. Keep up the great work and Happy Holidays :)

[@bcreason]

Thanks, I was traveling around Europe by rail earlier this year. I was wondering how they managed thermal expansion. The welded rails are definitely a quieter, more comfortable ride. I did notice something that you didn’t mention in the video. All the ties are concrete. No wooden ties anywhere. I assumed the ties were steel reinforced, being much stronger than wood. I figured this is what was keeping the rails from buckling. Around three steel reinforced ties per meter, using bolts to hold the rails instead of spikes in wood. This setup could withstand a lot of buckling force.

[@matthewmaxwell-burton4549]

Nope wood is still very common, and has some benefits over concrete. It depends on what they are looking for.

[@bcreason]

I didn't see any wooden ties in Europe. We were traveling through France(TGV), Switzerland, Germany, Italy. Maybe they are still used on routes that have slower trains. I didn't hear a single "Clickety Clack". It was all welded rails for the trains we were on. I don't think wood is strong enough to resist the buckling forces on welded rails. I can envision spikes pulling out and even bolts coming loose as the bolt holes expand from wearing away the wood as the wheels pass over. @@matthewmaxwell-burton4549

[@kentslocum]

I'm really enjoying this deep-dive series on railroads; not just because I like trains, but also because I've always wanted to know the answers to these same questions!

[@markmark2080]

An older average gear head here, I never cease to be amazed at the actual complexity of things that looked "simple" when I was young...

[@djmudflaps2382]

This video was pure joy (for me) as it was a revelation on things I really enjoy (trains). Let's hear it for engineers: softly spoken, confident but not arrogant, justifies everything with facts and data and shows a certain degree of humility as well but they also get things done. The best type of American is an engineer.