Since everyone keeps bringing it up, steel being an alloy has nothing to do with this phenomenon. It is not an unfair comparison. Pure iron will form austenite just at a slightly different temperature compared to steel and the same shrinkage will occur
You did this demonstration the hard way. When I was in college, I attended a chemistry magic show. A steel wire 4 feet long was snugly strung between two insulating poles. The wire was connected to a variable voltage power source. The power was slowly turned up, causing the wire to heat. As the wire heated up, it expanded and sagged down until it reached that critical point, when it then rose back up. This could be repeated as many times as you like.
An interesting not-related bit: you probably know that ice expands, but as it cools down water actually already begins to expand at 4 degrees celcius, so before it turns to ice.
@@Tvngsten if you put water in a solid shell without any gas gap to stop any expansion and take the temperature well below freezing it will not freeze. On opening the shell the water will stay liquid untill the surface tension is disturbed where upon it will quickly freeze!
@@waynethomas3638while you're right, it's not the whole truth. Minute physics made a video about that (/watch?v=_bcfxty39Cw). If or how much of the water freezes depends on the temperature. Water in an infinitely strong container would still freeze entirely if the temperature is low enough. And you don't need pressure on the water to supercool the liquid. Any perfectly smooth and clean container works (/_9N-Y2CyYhM?si=BS8M5CbByTlhBz88)
I love how every time I hear ~730°C in the context of steel I have to remember my material science teacher hammering 723°C into our brains. That number stuck.
Latex rubber shrinks on heating as well! It's the basis for a sun-tracker. Matched elastic bands pull evenly, platform doesn't rotate. As the Sun moves in the sky, one band moves into shadow, where it cools and expands a bit, causing rotation. The platform rotates until both bands are equally illuminated.
@@vx-iiduit’s a low cost solution though. I doubt maintaining electric motors and putting down electrical wires for something out in the middle of nowhere is very economical. Replacing a rubber band is probably just cheaper.
This "Shrinks when heated" thing happens with water, too. When water goes from a liquid to a solid, it expands. Water just uses a different mechanism for its' volume-alteration.
As a industrial blacksmith I always wondered why when you heat a piece of steel with a oxy acetylene torch the steel bends away from the flame but bends back when past 900deg c Thanks
As a novice knife maker I am trying to get my head around hardening and tempering steel. Your explanation as to why steel expands and contracts provided me with a better understanding of the hardening process including the affects of altering the carbon content. BONUS! Thanks
Needs an inclusion of an Iron-carbon phase diagram! What's being addressed is the AC1 line. Also a TTT diagram!!! Time, temperature, transformation. It's the recipe book for heat treating different alloys.
translation to fellow members of "we leardned dat shit on wallpers" gang- ttt=c-shaped diagram. IMO c-diagram sounds nicer, but I probably just spend too much time in TTT lobbys back in time.
What a great explanation. Just enough detail to understand what's really going on and a physical demonstration that I've never actually seen done. Awesome job man.
I've never salt tempered, but i have oil tempered small parts before to minimize warp and cracking. I wasn't too technical about it, just used one of the cheap counter top electric deep fryer and set it to 375 F for the quench and then raised the temp (with the part still submerged) to 420 F for the temper.
Fun fact: Steel shrinks as it cools down to a smaller size than it was before you did the whole heat/cool process. The amount that it shrinks is visible with the naked eye.
You're burning the carbon out of the steel. If you could check the carbon content of the steel, you would find that the percentage has changed. If you were going to try to heat treat the steel you would have poor results. That's why when heat treating tool steel we use some sort of controlled atmosphere like stainless steel tool wrap. If we didn't do that, all that effort that we put into making the precision part would be ruined.
Decarbonising is a thing but the rate is pretty much the inverse of carburising so you’d have to soak it for a considerable amount of time for a considerable amount of carbon to diffuse out. Cheers
The preview picture was hammering into my head: "HOT STEEL STINKS". A few times, before my brain started doing what lf is supposed to do -> "HOT STEEL SHRINKS".
So does water below 4 degrees Celsius. Water is at its densest at about 4 degrees. If you heat it above 4 degrees, it expands, but it also expands if you cool it below 4 degrees. So if you have 1 degree cold water and you heat it up to 3 degrees, it will actually shrink in volume.
There are two states of hot steel. Face centre cubic and body centre cubic. It goes from one state to the other and then back again, thats why it shrinks.
@@artisanmakes Sorry, but I Didn't watch it. Just remember from my apprenticeship days, many years ago, that this was the case. Was just putting it out therei n case people didn't know. 👍
that dip is extreme with cf reinforced polymers (just at lower temps). its why even though they're marketed as "engineering grade", they are rarely used in end products for real-world use. its a much more extreme dip too. some (like cf reinforced pekk-teflon alloy) dip so hard that it contracts to be smaller than room temp briefly at the 150c range.
At first all kind of thoughts of why, what variables, and what's been proven currently. I think you answered 99% of it. You certainly could go a few depths deeper into the exact meteorology but even I would be lost in the woods. Let's just say that you definitely hit all of the high notes. Thanks🤙🤙🤙🤙
I was really doubtful of the headline as I’ve used an acetylene torch and liquid nitrogen to fit steel parts but that little window of expansion is interesting and I didn’t know about it. My laser maxes at 1900F which really messed my last bronze casting up
a shadow line from a light source cast onto a measuring surface would probably be a much easier way to do this considering the act of manipulating the metal though the hole while it's hot with that much leverage would very possibly reform the metal and cool it down at the same time.
As far as heat expanding the hole, if it was a thin ring, yes it would expand the whole thing, increasing the size of the hole, but with this setup, large plate and heat is exclusively being delivered into the plate in the bore of the hole, the metal around the hole is being held by all that cold steel around it so the expanding metal will ever so slightly stretch the rest of plate but mostly expand into the mostly free directions inward and up/down making the plate thicker in that area, so the hole should get smaller. Obv the expanding metal is running into itself in a constrained ring so i dont think its even close to linear like normal expansion of basic objects. No clue how to calculate.
Shooting compressed air between aluminum plates has to be my favorite "quench". Forces thing things flatter. Harder to set up though and you've got to be quick and have enough air.
Salt bath for quenching looks fun! Never seen that before. Wonder how hard it'd be to make a simple heat chamber from ceramic bircks and nichrome to hold a pot for melting salt to a precise temp! :)
Not hard in theory, build the physical furnace or chamber thingy, make your elements, hook em up to a pid controller and power supply. Pull your hair out trying to figure out PID tuning lol idk if pid is actually hard to get going and tuned in that kind of application, I just know I had a hard time following along on a pid drone control video.
@@miles11we It's not too hard to PID tune, PyKiln algorythm isn't bad and can be autotuned, and any semi-decent commercial controller these days have fairly reliable PID tuning built in, cheaper ones not so good lol There's a few other open source projects that have PID algorythms of decent quality too :) I've built my own kilns and furnaces, using commercial and diy solutions, it's a lot easier than it sounds
@@matfan81 Yes, but as far as I'm aware the same open source resources for PID tuning drones automatically exist, that's literally what caused the explosion of cheaply available commercial drones in recent years, the ability to quickly and reliably do the tuning using easy software that's available to anyone
bless your resilient heart. you made it happen, no matter the stupidity... i'd say, the shrinkage is due to temporary loss of magnetism...you lose the valence first. and then the next ring of electrons. so it collapses at a given temperature, meaning, some of the electrons vacate...allowing the atoms to squeeze together...maybe.
This is a very well understood phenomenon and it is all down to a shift in the arrangement of the atoms in the crystal lattice. The loss of magnetism is down to the formation of austentie, which is not magnetic. Ferrite is.
I'll put my guess here before watching the explanation: It probably has to do with the cementite component of the steel dissolving, allowing the crystal structure to become much more compact
It's be much more work than an electronic measurement tool, but you should also be able to estimate temp by cooling the ball in a known volume of room temperature water based on the amount of water lost to evaporation and the temperature of the remainining liquid.
Makes me think of the videos of hot rivets in old skyscraper and airplane construction. I can envision putting in a rivet and flattening the end could make for a tighter fit.
Stumbled across this the other day. Just wanted to say that your analysis is spot on. However, I did want to make one minor correction. Carbon steels like 1045 are not alloy steels. AISI 1045 is just a plain carbon steel with 45 points carbon. Same for 1018, 1010, etc. Even steels like 12L14 or 11L55 are not alloy steels. They're just leaded steels to allow them to be free-machining. (Easier on the tooling.) Alloy steels have elements like nickel, chromium, molybdenum, vanadium, etc., in them--that are above a certain percentage. And be careful with 316L. That L doesn't mean lead. Just means low carbon. 🙂🙂
I don't work with steel much. On the motorways sometimes in the morning we have to remove a length of crash barrier for access.at the end of day after sun's been out the section of barrier won't fit cos the long runs each side have expanded in the English sun . That's only a couple of degrees c but can be miles long. Result is 10 to 15 mm each side. Sorry its long but I was interested first time I came across it
Wow Thanks I had no idea this was a thing, makes me wonder if you got it to shrink temp and quinched a knife blade in a vacuum or an extream press when cooled ,would it retain its dense properties for edge retention ? Any how thanks I did not know.
It’s because steel has carbon in it too. The other metals on that graph are all pure, but steel is a combination of iron and carbon. This means that the crystal structure of the steel can take multiple different forms. I believe that the contraction is because of the shift between martensite and austenite.
Am I right in thinking that the cooling you get from quenching comes from the energy being used in the phase transition from water to steam? Oil boils much higher than water, is that why it's slower? Or does it have to do with the conductivity of the liquids and heat capacity
I feel like making the hole progressively larger would have been way easier and more accurate than sanding the ball. (And by "feel like" I mean "experience tells me." )
Another question: There is a swimming pool with water in it. There is a canoe floating on the water in the swimming pool. Now, which will raise the water level in the pool the most, tossing a penny in the water or the canoe, and why?
Hmm, I was going to say that each would displace the same amount of water, but that's not correct. Volume and mass BOTH matter. So there's math involved to find the answer. Or, just try it both ways and measure the result, because I hate math.
@@TehButterflyEffect no math; u just need to know if zinc is denser than water, which it is. penny in pool will lift water level base on penny's volume. penny in boat will exert downward force equal to its weight that has to be countered by buoyant force, which would require weight of water equal to weight of penny. water is less dense than zinc, so more water has to be displaced to create buoyant force to offset weight of penny than the penny would directly in pool.
So what happens if you pop the hot ball in the hole and cool it? Will it deform the hole? Will it have a dent around the center where it tried to expand but couldn't? This is assuming you keep the steel "ring" cool so it can't expand as it heats up.
At a risk of sounding a bit preachy (as both mechanical engineer and a high-school teacher), might I offer an explanation I give to my students? The basic sctructure of Fe-C alloy (namely steel) is BCC. Steel at equillibrium at room temperature is 1 atom of carbon per 8 atoms of iron, the rest of carbon lies between crystals of BCC steel. Heating the steel causes the Fe atoms to drift further apart and at some temperature (usualy best gauged by magnet, since BCC crystalline structure is magnetic and FCC, ie austenite, is non-magnetic*) Fe atoms drift so far apart that 5 more atoms of carbon can squueze in the space to create FCC structure. Rapid cooling will 'freeze' the crystalline structure of the material (slow cooling would allow Fe atoms to 'squueze out' those 5 carbon atoms back out), but after cooling Fe atoms will be at a greater distance apart then they should naturally be (due to those pesky nuclear and gravitational forces which are at an equillibrium in BCC structure), and the force attracting Fe atoms to each other will cause internal stresses in the material causing the brittleness (once the carbon atom is squeezed out of FCC, but not other 4 the ballance is diminished and the crystalls start to break). I explain the tempering as a process of allowing Fe atoms to slighlty drift further apart, allowing enough time to squeeze some of the carbon out of FCC crystalls but allowing them enough time and space to revert to BCC, while majority remains in FCC, and expelled carbon atoms lie between those structures- and we name different FCC, BCC, free carbon structures as bainite, martensite, &c). Or am I getting something wrong? EDITED to add that the dip in the graph (increased density) IMHO occurs as all the BCC crystalls are converted to FCC, since I assume that BCC crystall +5 atoms of C will occupy more space than a single FCC crystall- as carbon gets absorbed in FCC steel crystalls instead of lying between those BCC crystalls the steel will shrink
This is largely incorrect. Carbon only occupies interstitial positions, not principal lattice positions. The BCC and FCC iron crystals only have iron atoms on those principal lattice positions, the carbon occupies special positions in between. So when you look at a BCC crystal and see 8 corner and 1 central atom, those are all iron, for FCC you have 8 corner and 6 face atoms which are only iron. The transformation from BCC to FCC upon heating is related to a free energy difference between the two crystal configurations and is a diffusion controlled transformation. Also technically BCC iron is only magnetic below the Curie temperature, there is a small temperature range above the Curie point where iron is still fully BCC but no longer magnetic. This change is not readily detectable as a structural difference by most methods but is apparently observable by neutron diffraction, if you look at really old Fe-C phase diagrams you may notice a phase called beta-iron which is no longer referred to. Beta-iron is the non-magnetic BCC phase that exists under the austenite range. Tempering is performed to improve toughness of martensite and/or bainite which is BCC or BCT (depending on carbon content) not FCC.
Would water cooling the block not have been pretty easy to make sure it didnt expand when you were fit testing? Like you could have just put it into a bath and done this sideways for minimum effort.
We do annealing / decarbonisation of electric steel for making fan motor. We have been doing it for so long. The size of the steel doesn’t change and is same after slow cooling. But suddenly this one odd time the size of steel sheets shrunk. The Dia went from 127.3 to 127.1 mm which is a big failure as we work in the microns. Please help how did this happen? We heat the steel till 750 deg C in H2 and N2 environment. What was the deviation or failure that caused this? I can’t figure it out and it is imporant
isn't it an unusual comparison, comparison pure Al, Cu, Ti against a Fe-C3 alloy? there might be alloys of the other elements as well which have a lattice change with temperature
I don't think so. Pure iron will do this exact same thing, just at slightly higher temperatures than carbon steel. Most metals don't form these allotropes like steel does, so they wont observe these lattice shifts. I know that cobalt does, but instead of shrinking, the lattice expands further with the phase shift
could you make a video about casting steel and iron? i have done a lot of research and there is basically no hobbyist steelcasting on the internet, but it could be done. i think you could get a LOT of views making cast crucible steel tools and billets, and it is achievable with coal coke and preheated forced air.
Would quenching in boiling water bring any noticeable reduction in cracking/warping? Since the energy required for the phase change from liquid to gas is a lot more than just heating up liquid water, I'd imagine it would cool a bit slower.
From what I have read. Boiling water causes an instant steam blanket around the part and insulates it from rapid cooling, so it ends up slow cooling and not hardening.
What’s he’s describing here is shrinking while it’s hot. What you encountered is what’s called grain size reduction. The individual grains that make up metal restructure themselves during the hardening process. They get smaller and tighter together, and you end up with a minutely smaller part in many cases depending on the metal you are working with!
The 'Black Art' of high temp steels and casting...nice! Talk to pattern makers about this to get a real insight to the issue. I spent a few years working in an engineering shop that had castings/forgings made regularly and got to spend a lot of time with pattern makers. For castings pattern makers have different sets of rules (that are have scales greater than 1mm = 1mm) to 'build in' the shrinkage of molten metals back to the 'design size' Also, depending on what way long, irregular items (like "Banbury mixing shafts") are poured and cooled, different shrinkage occurs along the length of the item (Don't ask how many mixing shafts were scraped before we worked out what the F@#% was going on....LOL)
This is not a product of oxidation, this is a change in allotrope which very few metals have and iron is almost unique in that it shrinks when it changes
And if you were to try this with type 304 stainless steel, 316 stainless steel, or actually any other 300 series stainless steels, the thermal expansion would not have this kink. These are austenitic (FCC structured) steels, even at room temperature, and do not experience a phase transition as they heat up.
Depends on how hot you get them. If you get them hotter than a certain temperature (I honestly can't remember, but I think it's around 5,000°), then it cooks the stainless and changes its properties.
@@TehButterflyEffect Uhh. 5000 °F (2760 °C) is not the melting point of steel, it is the _boiling point_ of steel. (Give or take some, since that depends on the chemical composition of the steel.) So while technically it's correct that by _bringing steel to the boil_ you'd change the chemical composition, because chromium would evaporate and burn out of the melt, and it could become non-stainless, it's a moot point, because it is something different than a straightforward crystal structure change. It's an academic scenario and something that'd be really challenging to put into practice, outside of a lab with very specialized equipment.
I feel like you are trying to make this simple for people but this is a very deep subject and I am confused about some of your findings. With the thermal expansion diagram you show at 1:00 you are comparing an alloy against pure elements, steel isn't behaving weird, it's behaving like an alloy. The downward bump you see is the iron and carbon rearranging themselves into a more stable geometry, it's steels first phase change, that change uses up some energy and that is the dip. When dealing with phase diagrams you need to make sure that you are using the diagram with the alloy you have. You'll notice that each type of steel will have 5-6 or more different configurations that will form at certain temperatures. In your experiment it looked like as soon as the steel looked the color red you wanted you immediately tried your experiment, if you redo this experiment I would recommend getting the steel up to temperature and leaving it at that temperature for 10-15 minutes, that will make sure the heat is the same throughout the steel sphere. If you just get the outside up only the very outside of the sphere will undergo the phase change while the majority of the steel stays the same. Sometimes you want a certain crystal structure so once you get the steel to the temperature needed and give it some time for it to rearrange itself, you'll need to cool it quickly so the steel wont spend time in any of the other phases that require a lower temperature. This might work for your experiment, get your steel sample up to the temperature you would expect it to shrink at and quench it in water once you are sure the sphere is the same temperature throughout, this should lock the crystals in that configuration. There are different quenches; you mentioned water, oil, salt, those will cool at a different rate because of their thermal conductivity and their boiling point. I believe salt melts at around 600 degrees, so if you quench in that it will help pull all the energy until the salt hits 600 degrees and melts, that means once the metal hits 600 and the salt hits 600 it will start to cool slowly again which can help remove some of the stress. Water is 212 degrees, so like a car hitting a brick wall, oil is hitting a brick wall with a seat belt, and salt is like hitting a brick wall with seat belts and an air bag. You can also use an annealing furnace to control a lot of this, you can effectively control the temperature up and down to chose different crystal structures you wouldn't normally see in nature. I don't know if you'd be able to notice any size change at that first phase shift, we are talking about atoms rearranging themselves, so if there is a change I don't know how you would even be able to measure it, i'm guessing there would never be an application where that tolerance is needed unless you're dealing with huge distances or large amounts of material. I would think though that the thermal expansion up to that phase change would be way more of an issue. Your experiment is an interesting idea, it's really awesome that you designed your own experiment and then proceeded to test it. It really shows that you understand the tools and processes you need to be a great machinist if you go in that direction. I spent years in school studying metallurgy and it's absolutely fascinating with tons of job opportunities and pretty decent pay. I'd say keep working on these experiment's. You should look at an annealing furnace for metal or plastic experiments, you can do some pretty awesome stuff with one of those bad boys. I don't think I have any of the pictures around but I got to use an electron microscope to look at my phase transition borders once, it was pretty amazing. Keep pushing forward though, you're doing great! Also make sure you record all your experiments, this youtube channel is a great idea, it will help
It’s important to remember that this happens with pure iron too. Just at a slightly elevated temperature compared to steel. And I never have lite iron in the shop anyway so it’s a bit difficult to experiment with it
I have a set of them but I don’t think they would be much use here since I was dealing with a non sphere which I couldn’t accurately predict how evenly it would expand. Cheers
Since everyone keeps bringing it up, steel being an alloy has nothing to do with this phenomenon. It is not an unfair comparison. Pure iron will form austenite just at a slightly different temperature compared to steel and the same shrinkage will occur
Kinda shocks me that the graphs you showed with a literal "CARBON % OF STEEL" didnt clue people into that.
You did this demonstration the hard way. When I was in college, I attended a chemistry magic show. A steel wire 4 feet long was snugly strung between two insulating poles. The wire was connected to a variable voltage power source. The power was slowly turned up, causing the wire to heat. As the wire heated up, it expanded and sagged down until it reached that critical point, when it then rose back up. This could be repeated as many times as you like.
That's pretty ingenious.
@@mikemhz exactly what I was about to write
Thays exactly the way it was shown to us. Its a done method and I wanted to show something a bit different.....
Sometimes there is value and beauty in doing things the hard or "wrong" way.
An interesting not-related bit: you probably know that ice expands, but as it cools down water actually already begins to expand at 4 degrees celcius, so before it turns to ice.
So if you want the maximum hydration per gulp, you should drink 4°C water.
@@Tvngsten if you put water in a solid shell without any gas gap to stop any expansion and take the temperature well below freezing it will not freeze. On opening the shell the water will stay liquid untill the surface tension is disturbed where upon it will quickly freeze!
@@waynethomas3638while you're right, it's not the whole truth. Minute physics made a video about that (/watch?v=_bcfxty39Cw). If or how much of the water freezes depends on the temperature. Water in an infinitely strong container would still freeze entirely if the temperature is low enough.
And you don't need pressure on the water to supercool the liquid. Any perfectly smooth and clean container works (/_9N-Y2CyYhM?si=BS8M5CbByTlhBz88)
I fell for that bait by the science teacher in grade school. 😂
water is ice, but just in a different state
I love how every time I hear ~730°C in the context of steel I have to remember my material science teacher hammering 723°C into our brains. That number stuck.
That still is better than when they start to ask what letters mean.
Latex rubber shrinks on heating as well!
It's the basis for a sun-tracker.
Matched elastic bands pull evenly, platform doesn't rotate.
As the Sun moves in the sky, one band moves into shadow, where it cools and expands a bit, causing rotation.
The platform rotates until both bands are equally illuminated.
Now that's really interesting. I never thought there could be a sun tracker without electric motors.
wouldnt that just break quickly? every piece of rubber I've seen exposed to the sun and just cracks and breaks after a few months.
@@vx-iiduit’s a low cost solution though. I doubt maintaining electric motors and putting down electrical wires for something out in the middle of nowhere is very economical. Replacing a rubber band is probably just cheaper.
This "Shrinks when heated" thing happens with water, too. When water goes from a liquid to a solid, it expands. Water just uses a different mechanism for its' volume-alteration.
As a industrial blacksmith I always wondered why when you heat a piece of steel with a oxy acetylene torch the steel bends away from the flame but bends back when past 900deg c
Thanks
Its because it gets uncomfortable and tries to pull away but then it gets used to it and it thinks “this isnt so bad” and leans back into the flame.
As a novice knife maker I am trying to get my head around hardening and tempering steel. Your explanation as to why steel expands and contracts provided me with a better understanding of the hardening process including the affects of altering the carbon content.
BONUS!
Thanks
"Ball turner" This is going to be some this old tony type gag right? Wrong, well, that sure looks like a bona fide ball turner to me.
No, it's a kink thing.
@@jhonbus 🥵🥵🥵
@@jhonbusAs if that wouldn't be the ToT joke.
Needs an inclusion of an Iron-carbon phase diagram! What's being addressed is the AC1 line.
Also a TTT diagram!!! Time, temperature, transformation. It's the recipe book for heat treating different alloys.
translation to fellow members of "we leardned dat shit on wallpers" gang- ttt=c-shaped diagram. IMO c-diagram sounds nicer, but I probably just spend too much time in TTT lobbys back in time.
What a great explanation. Just enough detail to understand what's really going on and a physical demonstration that I've never actually seen done. Awesome job man.
This is an incredibly nerdy video and I love it.
Excellent information. The FCC and BCC cracking is super interesting. I had no idea.
I've never salt tempered, but i have oil tempered small parts before to minimize warp and cracking.
I wasn't too technical about it, just used one of the cheap counter top electric deep fryer and set it to 375 F for the quench and then raised the temp (with the part still submerged) to 420 F for the temper.
Fun fact: Steel shrinks as it cools down to a smaller size than it was before you did the whole heat/cool process. The amount that it shrinks is visible with the naked eye.
Woah... I need to look back through your videos to find "I turned a boring head into a ball turner". :)
You're burning the carbon out of the steel. If you could check the carbon content of the steel, you would find that the percentage has changed. If you were going to try to heat treat the steel you would have poor results. That's why when heat treating tool steel we use some sort of controlled atmosphere like stainless steel tool wrap. If we didn't do that, all that effort that we put into making the precision part would be ruined.
Decarbonising is a thing but the rate is pretty much the inverse of carburising so you’d have to soak it for a considerable amount of time for a considerable amount of carbon to diffuse out. Cheers
That would mean this expansion curve isn't consistent and repeatable which it is.
The preview picture was hammering into my head: "HOT STEEL STINKS". A few times, before my brain started doing what lf is supposed to do -> "HOT STEEL SHRINKS".
😂
So does water below 4 degrees Celsius. Water is at its densest at about 4 degrees. If you heat it above 4 degrees, it expands, but it also expands if you cool it below 4 degrees. So if you have 1 degree cold water and you heat it up to 3 degrees, it will actually shrink in volume.
There are two states of hot steel. Face centre cubic and body centre cubic. It goes from one state to the other and then back again, thats why it shrinks.
Yep. Thats the video
@@artisanmakes Sorry, but I Didn't watch it. Just remember from my apprenticeship days, many years ago, that this was the case. Was just putting it out therei n case people didn't know. 👍
Thats alright but you were able to figure it out
that dip is extreme with cf reinforced polymers (just at lower temps). its why even though they're marketed as "engineering grade", they are rarely used in end products for real-world use. its a much more extreme dip too. some (like cf reinforced pekk-teflon alloy) dip so hard that it contracts to be smaller than room temp briefly at the 150c range.
it 's because at this temperatur the grain structure of the steel change and making it contract instead of expending during that change.
At first all kind of thoughts of why, what variables, and what's been proven currently. I think you answered 99% of it. You certainly could go a few depths deeper into the exact meteorology but even I would be lost in the woods. Let's just say that you definitely hit all of the high notes. Thanks🤙🤙🤙🤙
As QA/QC student, mentioning those particular temperatures immediately ringed the bell in my head. I remember what happens to steel there
I was really doubtful of the headline as I’ve used an acetylene torch and liquid nitrogen to fit steel parts but that little window of expansion is interesting and I didn’t know about it. My laser maxes at 1900F which really messed my last bronze casting up
Thanks for the hustle to demonstrate this!
Also, we now know you have 2 balls of steel 😂
a shadow line from a light source cast onto a measuring surface would probably be a much easier way to do this considering the act of manipulating the metal though the hole while it's hot with that much leverage would very possibly reform the metal and cool it down at the same time.
Good thing you had to make a second one, because now we can comment on your balls of steel!
As far as heat expanding the hole, if it was a thin ring, yes it would expand the whole thing, increasing the size of the hole, but with this setup, large plate and heat is exclusively being delivered into the plate in the bore of the hole, the metal around the hole is being held by all that cold steel around it so the expanding metal will ever so slightly stretch the rest of plate but mostly expand into the mostly free directions inward and up/down making the plate thicker in that area, so the hole should get smaller. Obv the expanding metal is running into itself in a constrained ring so i dont think its even close to linear like normal expansion of basic objects. No clue how to calculate.
Shooting compressed air between aluminum plates has to be my favorite "quench". Forces thing things flatter. Harder to set up though and you've got to be quick and have enough air.
Salt bath for quenching looks fun! Never seen that before. Wonder how hard it'd be to make a simple heat chamber from ceramic bircks and nichrome to hold a pot for melting salt to a precise temp! :)
Not hard in theory, build the physical furnace or chamber thingy, make your elements, hook em up to a pid controller and power supply. Pull your hair out trying to figure out PID tuning lol
idk if pid is actually hard to get going and tuned in that kind of application, I just know I had a hard time following along on a pid drone control video.
@@miles11we It's not too hard to PID tune, PyKiln algorythm isn't bad and can be autotuned, and any semi-decent commercial controller these days have fairly reliable PID tuning built in, cheaper ones not so good lol There's a few other open source projects that have PID algorythms of decent quality too :)
I've built my own kilns and furnaces, using commercial and diy solutions, it's a lot easier than it sounds
Pid tuning a drone is completely different to what is basically a simple thermostat.
@@matfan81 Yes, but as far as I'm aware the same open source resources for PID tuning drones automatically exist, that's literally what caused the explosion of cheaply available commercial drones in recent years, the ability to quickly and reliably do the tuning using easy software that's available to anyone
The lattice changes from body centered cubic 9 atoms to face centered cubic 14 atoms, martensitic then quench, very hard no ductility until tempered.
Thank you for going to so much trouble to explain your point. I particularly liked the dynamic graphics and would love to know what program you used.
bless your resilient heart. you made it happen, no matter the stupidity...
i'd say, the shrinkage is due to temporary loss of magnetism...you lose the valence first. and then the next ring of electrons. so it collapses at a given temperature, meaning, some of the electrons vacate...allowing the atoms to squeeze together...maybe.
This is a very well understood phenomenon and it is all down to a shift in the arrangement of the atoms in the crystal lattice. The loss of magnetism is down to the formation of austentie, which is not magnetic. Ferrite is.
Interesting! I never thought in these terms about brittleness of materials.
I'll put my guess here before watching the explanation:
It probably has to do with the cementite component of the steel dissolving, allowing the crystal structure to become much more compact
It's be much more work than an electronic measurement tool, but you should also be able to estimate temp by cooling the ball in a known volume of room temperature water based on the amount of water lost to evaporation and the temperature of the remainining liquid.
Makes me think of the videos of hot rivets in old skyscraper and airplane construction. I can envision putting in a rivet and flattening the end could make for a tighter fit.
Such a random sighting of Mick West. I guess it makes sense that he'd be interested in this kind of topic
This is very cool and fun! I wish my physics professor had included this in one of his materials science lectures.
Stumbled across this the other day. Just wanted to say that your analysis is spot on. However, I did want to make one minor correction. Carbon steels like 1045 are not alloy steels. AISI 1045 is just a plain carbon steel with 45 points carbon. Same for 1018, 1010, etc. Even steels like 12L14 or 11L55 are not alloy steels. They're just leaded steels to allow them to be free-machining. (Easier on the tooling.)
Alloy steels have elements like nickel, chromium, molybdenum, vanadium, etc., in them--that are above a certain percentage.
And be careful with 316L. That L doesn't mean lead. Just means low carbon. 🙂🙂
For the most part yes but carbon steels are never just carbon and iron. There is always other stuff alloyed in, just in small quantities
I don't work with steel much. On the motorways sometimes in the morning we have to remove a length of crash barrier for access.at the end of day after sun's been out the section of barrier won't fit cos the long runs each side have expanded in the English sun . That's only a couple of degrees c but can be miles long.
Result is 10 to 15 mm each side.
Sorry its long but I was interested first time I came across it
Wow Thanks I had no idea this was a thing, makes me wonder if you got it to shrink temp and quinched a knife blade in a vacuum or an extream press when cooled ,would it retain its dense properties for edge retention ? Any how thanks I did not know.
It’s because steel has carbon in it too. The other metals on that graph are all pure, but steel is a combination of iron and carbon. This means that the crystal structure of the steel can take multiple different forms. I believe that the contraction is because of the shift between martensite and austenite.
He says that in the video
Austenite forms without the presence of carbon
@@artisanmakes my inorganic chemistry is a bit rusty
in fairness its a really obscure part of chemistry
Very nice to see you explaining some metallurgy! :D
Am I right in thinking that the cooling you get from quenching comes from the energy being used in the phase transition from water to steam? Oil boils much higher than water, is that why it's slower? Or does it have to do with the conductivity of the liquids and heat capacity
I feel like making the hole progressively larger would have been way easier and more accurate than sanding the ball. (And by "feel like" I mean "experience tells me." )
It's because the steel is shifting heat gears
Another question: There is a swimming pool with water in it. There is a canoe floating on the water in the swimming pool. Now, which will raise the water level in the pool the most, tossing a penny in the water or the canoe, and why?
Tossing in a penny because the canoe is already in the pool.
that's a fun little physics question, & like a perfect primer on buoyancy.
Hmm, I was going to say that each would displace the same amount of water, but that's not correct. Volume and mass BOTH matter. So there's math involved to find the answer.
Or, just try it both ways and measure the result, because I hate math.
@@TehButterflyEffect no math; u just need to know if zinc is denser than water, which it is.
penny in pool will lift water level base on penny's volume. penny in boat will exert downward force equal to its weight that has to be countered by buoyant force, which would require weight of water equal to weight of penny. water is less dense than zinc, so more water has to be displaced to create buoyant force to offset weight of penny than the penny would directly in pool.
So what happens if you pop the hot ball in the hole and cool it? Will it deform the hole? Will it have a dent around the center where it tried to expand but couldn't? This is assuming you keep the steel "ring" cool so it can't expand as it heats up.
At a risk of sounding a bit preachy (as both mechanical engineer and a high-school teacher), might I offer an explanation I give to my students? The basic sctructure of Fe-C alloy (namely steel) is BCC. Steel at equillibrium at room temperature is 1 atom of carbon per 8 atoms of iron, the rest of carbon lies between crystals of BCC steel. Heating the steel causes the Fe atoms to drift further apart and at some temperature (usualy best gauged by magnet, since BCC crystalline structure is magnetic and FCC, ie austenite, is non-magnetic*) Fe atoms drift so far apart that 5 more atoms of carbon can squueze in the space to create FCC structure. Rapid cooling will 'freeze' the crystalline structure of the material (slow cooling would allow Fe atoms to 'squueze out' those 5 carbon atoms back out), but after cooling Fe atoms will be at a greater distance apart then they should naturally be (due to those pesky nuclear and gravitational forces which are at an equillibrium in BCC structure), and the force attracting Fe atoms to each other will cause internal stresses in the material causing the brittleness (once the carbon atom is squeezed out of FCC, but not other 4 the ballance is diminished and the crystalls start to break). I explain the tempering as a process of allowing Fe atoms to slighlty drift further apart, allowing enough time to squeeze some of the carbon out of FCC crystalls but allowing them enough time and space to revert to BCC, while majority remains in FCC, and expelled carbon atoms lie between those structures- and we name different FCC, BCC, free carbon structures as bainite, martensite, &c). Or am I getting something wrong?
EDITED to add that the dip in the graph (increased density) IMHO occurs as all the BCC crystalls are converted to FCC, since I assume that BCC crystall +5 atoms of C will occupy more space than a single FCC crystall- as carbon gets absorbed in FCC steel crystalls instead of lying between those BCC crystalls the steel will shrink
This is largely incorrect. Carbon only occupies interstitial positions, not principal lattice positions. The BCC and FCC iron crystals only have iron atoms on those principal lattice positions, the carbon occupies special positions in between. So when you look at a BCC crystal and see 8 corner and 1 central atom, those are all iron, for FCC you have 8 corner and 6 face atoms which are only iron. The transformation from BCC to FCC upon heating is related to a free energy difference between the two crystal configurations and is a diffusion controlled transformation. Also technically BCC iron is only magnetic below the Curie temperature, there is a small temperature range above the Curie point where iron is still fully BCC but no longer magnetic. This change is not readily detectable as a structural difference by most methods but is apparently observable by neutron diffraction, if you look at really old Fe-C phase diagrams you may notice a phase called beta-iron which is no longer referred to. Beta-iron is the non-magnetic BCC phase that exists under the austenite range.
Tempering is performed to improve toughness of martensite and/or bainite which is BCC or BCT (depending on carbon content) not FCC.
Could have connected your steel ball to the tip of an electrically controlled soldering iron for accurate temperature control.
How hot do soldering irons get? I didn’t think that they got red hot
You know, bacon also shrinks when heated. So, when making bacon, you can feel good that you can justify it as science 😁
if you used a molecular pump to evacuate a champer with cooling steel couldn't you get an even stronger vaccume when it shrinks?
Would water cooling the block not have been pretty easy to make sure it didnt expand when you were fit testing? Like you could have just put it into a bath and done this sideways for minimum effort.
Great explanation, Ray.
Isn't nickel lost every single time the alloy is heated up to melting temp?
Also remember that iron is the key ingredient for the formation of blackholes. 😊
Thanks, man!
Stay safe there with your family! 🖖😊
So? Water expands when it freezes. Thats not normal either but water doesnt care whats normal. Apparently neither does steel.
Get yourself some Temp Stick markers, that how I check my temps when heat treating knives or for shrink fits. 👍👍
They are very expensive, but they work great.
You hear that steel your not normal mate
Might want to see some one about that.
Very cool, had no idea - thanks for that, friendo.
And with that, I am now the smartest person I know.
I was totally not expecting to encounter Crystallography on your channel! A surprise!
(Shoutout to the axis label “tempreature”)
Hey if I had to sit (sleep) through materials class I think everyone else should too :)
We do annealing / decarbonisation of electric steel for making fan motor. We have been doing it for so long. The size of the steel doesn’t change and is same after slow cooling. But suddenly this one odd time the size of steel sheets shrunk. The Dia went from 127.3 to 127.1 mm which is a big failure as we work in the microns. Please help how did this happen? We heat the steel till 750 deg C in H2 and N2 environment. What was the deviation or failure that caused this? I can’t figure it out and it is imporant
Holy F… how did I never know that kind of ball attachment existed for a lathe?
Aren't rail sleepers the wooden or concrete pieces which go across underneath the rails?
Isn't it the metal rails which expand and contract?
hm..that is why the blades i try to make with hight carbon steel used to crack a lot.
Very good explanation.
isn't it an unusual comparison, comparison pure Al, Cu, Ti against a Fe-C3 alloy? there might be alloys of the other elements as well which have a lattice change with temperature
I don't think so. Pure iron will do this exact same thing, just at slightly higher temperatures than carbon steel. Most metals don't form these allotropes like steel does, so they wont observe these lattice shifts. I know that cobalt does, but instead of shrinking, the lattice expands further with the phase shift
I was really hoping he would heat up the ball to the point where it shrinks then let it cool down in the hole
could you make a video about casting steel and iron? i have done a lot of research and there is basically no hobbyist steelcasting on the internet, but it could be done. i think you could get a LOT of views making cast crucible steel tools and billets, and it is achievable with coal coke and preheated forced air.
Mechanical engineering explain. Thank for this
Would quenching in boiling water bring any noticeable reduction in cracking/warping? Since the energy required for the phase change from liquid to gas is a lot more than just heating up liquid water, I'd imagine it would cool a bit slower.
From what I have read. Boiling water causes an instant steam blanket around the part and insulates it from rapid cooling, so it ends up slow cooling and not hardening.
I was just asking my self this yesterday. As my part got smaller after heat treating in water.
What’s he’s describing here is shrinking while it’s hot. What you encountered is what’s called grain size reduction. The individual grains that make up metal restructure themselves during the hardening process. They get smaller and tighter together, and you end up with a minutely smaller part in many cases depending on the metal you are working with!
when water turns into a solid it also expands
Great video. A tapered cylinder would have helped you.
Kinda like ice. It expnads when u might expect it to condense.
Ok you just blew my last 2 brain cells
A man told me that a long time ago and I didn’t believe him. Turns out he was right.
The 'Black Art' of high temp steels and casting...nice! Talk to pattern makers about this to get a real insight to the issue.
I spent a few years working in an engineering shop that had castings/forgings made regularly and got to spend a lot of time with pattern makers.
For castings pattern makers have different sets of rules (that are have scales greater than 1mm = 1mm) to 'build in' the shrinkage of molten metals back to the 'design size'
Also, depending on what way long, irregular items (like "Banbury mixing shafts") are poured and cooled, different shrinkage occurs along the length of the item (Don't ask how many mixing shafts were scraped before we worked out what the F@#% was going on....LOL)
Does it shrink or is it the Oxidation layers peeling off?
im falling asleep.... and I'm playing this in 2x
You don’t have to watch the video.
Can you in oil at 200deg? Or does the piece have to come right down to room temp?
hot oil is fine you'll just produce a piece that is not as hard as it could be, assuming we are talking about carbon steel
Now I have to figure out some practical application.🤔😄
Outgassing and oxidation makes any reactive metal shrink. This is normal.
This is not a product of oxidation, this is a change in allotrope which very few metals have and iron is almost unique in that it shrinks when it changes
I may have missed it, but another equally bizarre thing is that at some point in the upper temps, it stops being MAGNETIC! THAT is strange!
No. Everything when it gets hot loses is magnetic ability
@@tylermcnally8232 ....Good job missing the point. Troll.
Is this repeatable in a vacuum environment?
Thanks for that mate
And if you were to try this with type 304 stainless steel, 316 stainless steel, or actually any other 300 series stainless steels, the thermal expansion would not have this kink. These are austenitic (FCC structured) steels, even at room temperature, and do not experience a phase transition as they heat up.
Depends on how hot you get them. If you get them hotter than a certain temperature (I honestly can't remember, but I think it's around 5,000°), then it cooks the stainless and changes its properties.
@@TehButterflyEffect Uhh. 5000 °F (2760 °C) is not the melting point of steel, it is the _boiling point_ of steel. (Give or take some, since that depends on the chemical composition of the steel.)
So while technically it's correct that by _bringing steel to the boil_ you'd change the chemical composition, because chromium would evaporate and burn out of the melt, and it could become non-stainless, it's a moot point, because it is something different than a straightforward crystal structure change. It's an academic scenario and something that'd be really challenging to put into practice, outside of a lab with very specialized equipment.
Thermal gun?
Who's gonna tell em about water💧
Water expands when chilled under 32 degrees
Water is an outlier too
Cool, thanks!
I feel like you are trying to make this simple for people but this is a very deep subject and I am confused about some of your findings. With the thermal expansion diagram you show at 1:00 you are comparing an alloy against pure elements, steel isn't behaving weird, it's behaving like an alloy. The downward bump you see is the iron and carbon rearranging themselves into a more stable geometry, it's steels first phase change, that change uses up some energy and that is the dip. When dealing with phase diagrams you need to make sure that you are using the diagram with the alloy you have. You'll notice that each type of steel will have 5-6 or more different configurations that will form at certain temperatures. In your experiment it looked like as soon as the steel looked the color red you wanted you immediately tried your experiment, if you redo this experiment I would recommend getting the steel up to temperature and leaving it at that temperature for 10-15 minutes, that will make sure the heat is the same throughout the steel sphere. If you just get the outside up only the very outside of the sphere will undergo the phase change while the majority of the steel stays the same.
Sometimes you want a certain crystal structure so once you get the steel to the temperature needed and give it some time for it to rearrange itself, you'll need to cool it quickly so the steel wont spend time in any of the other phases that require a lower temperature. This might work for your experiment, get your steel sample up to the temperature you would expect it to shrink at and quench it in water once you are sure the sphere is the same temperature throughout, this should lock the crystals in that configuration. There are different quenches; you mentioned water, oil, salt, those will cool at a different rate because of their thermal conductivity and their boiling point. I believe salt melts at around 600 degrees, so if you quench in that it will help pull all the energy until the salt hits 600 degrees and melts, that means once the metal hits 600 and the salt hits 600 it will start to cool slowly again which can help remove some of the stress. Water is 212 degrees, so like a car hitting a brick wall, oil is hitting a brick wall with a seat belt, and salt is like hitting a brick wall with seat belts and an air bag. You can also use an annealing furnace to control a lot of this, you can effectively control the temperature up and down to chose different crystal structures you wouldn't normally see in nature. I don't know if you'd be able to notice any size change at that first phase shift, we are talking about atoms rearranging themselves, so if there is a change I don't know how you would even be able to measure it, i'm guessing there would never be an application where that tolerance is needed unless you're dealing with huge distances or large amounts of material. I would think though that the thermal expansion up to that phase change would be way more of an issue.
Your experiment is an interesting idea, it's really awesome that you designed your own experiment and then proceeded to test it. It really shows that you understand the tools and processes you need to be a great machinist if you go in that direction. I spent years in school studying metallurgy and it's absolutely fascinating with tons of job opportunities and pretty decent pay. I'd say keep working on these experiment's. You should look at an annealing furnace for metal or plastic experiments, you can do some pretty awesome stuff with one of those bad boys. I don't think I have any of the pictures around but I got to use an electron microscope to look at my phase transition borders once, it was pretty amazing. Keep pushing forward though, you're doing great!
Also make sure you record all your experiments, this youtube channel is a great idea, it will help
It’s important to remember that this happens with pure iron too. Just at a slightly elevated temperature compared to steel. And I never have lite iron in the shop anyway so it’s a bit difficult to experiment with it
I know they cost a lot, but a micro meter wouldnt hurt😅
I have a set of them but I don’t think they would be much use here since I was dealing with a non sphere which I couldn’t accurately predict how evenly it would expand. Cheers
It's probably because steel by definition contains carbon, which shrinks when heated.
Even without carbon, pure ferrite will form austenite and shrink. The amount of carbon will effect the temperature that is occurs.
So you're telling us steel ice floats over liquid steel?
Did you watch the video? That is not remotely what I am talking about
interesting topic !!!! ;)
Eh... it kind of _is_ normal though. Same as water under 4°C. A change in microstructure makes all the difference.
Water is an outlier too…..
And it’s not a change in microstructure. It is a formation of one
Interesting.
2:23 it felt painful just looking at it.