Jerrod Miller

You may want to reach out to specific colleagues of Prof. Verhoeven, rather than the school itself. I would imagine Christoph Beckermann worked with Verhoeven and thus may have a personal email address or something that could help. Last year when I spoke with Beckermann, he said he was going to be retiring soon (I think this is his last year), so he may even have some insight into contacting retired professors, too.

It shouldn't need to be air tight. A tiny hole will allow pressure relief, but it isn't going to really suck the CO out very quickly. There will be a little bit of drive to mix the gas, but it isn't like you will be blowing air through the canister. So, a pinhole is a source of lost efficiency, but not a big one.

Yes.

I'd recommend 2 small changes. 1) I would go a bit cooler than non-magnetic for the normalization cycles. That is about 1414F, and I'd go down to at least 900F. 2) I would temper a second time after the first. Cool down to room temp between temper cycles (still air, water, whatever - doesn't matter).

I would think it would help get it extra hot and soak it for a bit first. There are primary M7C3 carbides in there (meaning they form during solidification - right from the liquid), which are pretty big (comparatively speaking); they're long and skinny. You have to soak at pretty high temp to dissolve them, then the steel will move easier under the hammer and you will get smaller carbides forming at the end. The steel may have already done this from the mill, but perhaps not. ASM Heat Treater's guide says the forging range is 1905-2150 F, and I would say you should be closer to the top end. When done forging, do a soak at about 1200F to ensure carbides re-precipitate. Then HT as normal. Of course, you should absolutely test the results of this procedure to ensure it ends up doing what you want it to do for you. Note: Kevin Cashen recommends never dissolving carbides to a point you get more than 0.8% C in your matrix, but that is qualified with having small and dispersed carbides to begin with. If your carbides are big to begin with, then you need to dissolved them and create new smaller ones. It is certainly easier to not mess with the carbides if you don't need to.

It depends on the application, often it is not best. Carbon generally reduces toughness. A case hardened piece will have been quenched. You can carburize the surface without quenching, too. The point of case hardening is to change the outer layer (case) to have a chemistry that will harden readily while the inner chemistry remains less hardenable. If one only heats/quenches the surface and not the center, that is surface hardening, which is also useful. Surface hardening can be done with flame or induction pretty readily. I suppose it could be done well with molten salt baths and other fun things, too. No. It will be brittle if you quench it without tempering, but if you carburize and slow cool it will be as soft as it is ever going to be. It will just have pearlite (and possibly ferrite, depending on chemistry) if slow cooled, just martensite if quenched. No. Carburizing is one step in case hardening. It is just the adding of carbon to the surface layer. You still have to quench from appropriate temperature (either straight from the carburizing cycle or a secondary heat treat cycle). Through hardening is not done with case hardening, and is in fact counter to the goals of case hardening. Differential hardening can be achieved via chemistry, like case hardening, or by temperature control, like surface hardening. This is also something to think about with differentially hardened blades. Hamons may look cool (beauty is in the eye of the beholder for things like this), but they definitely mean there is less hardened steel than a through hardened piece and you will eventually use up all the hardened steel before you would on a through hardened piece. This is why I personally do not like to see hamons anywhere near the edge. That is too much of a sacrifice of functionality for my tastes, but it isn't a big deal for others.

Yes, that is correct.

@Rean Lubbe Have you seen this thread:

That is a bad assumption, because it doesn't work like that. If something doesn't melt into the bath when making steel, it stays looking like a rock. When we add elements to steel, it looks like this: Ferromanganese https://en.wikipedia.org/wiki/Ferromanganese#/media/File:Ferromanganèse_métal.jpg Ferrochromium https://en.wikipedia.org/wiki/Ferrochrome#/media/File:Ferrochrome.JPG If that doesn't melt down, you will have a very bad and obvious inclusion, not a little shiny section. Anyone making anywhere near decent steel (and even placing making quite bad steel) will not have this happen. The entire bath is generally taken a few hundred degrees above melting point, and it is often held at hot enough temperatures while chemistries are verified and such that this just doesn't happen. What may happen is alloy segregation, where alloying elements are not evenly distributed in a sample. This can be caused by segregation during solidification, or from things like dissolving carbides (either primary carbides from solidification, or those present in powder metallurgy) and not giving sufficient time and temp to even out via diffusion. The elements are still dissolved in that they are part of the matrix. They just aren't evenly dispersed. Polishing alone isn't likely to tell you what you want to know. Etching is needed.

I was just referring to these sections: Not this type of splotches: Exactly. It sounds to me like you desperately need to start polishing (and preferably etching) your samples and looking at them on the metallurgical microscope to learn what you are wanting to learn. Those are only intended to be used with polished samples anyway.

Nope. This is just light reflecting back into the lens off the shiny parts that just so happen to have a bunch of small areas clos together at the same (or close enough to it) angle to all reflect together.

You would still need a gating system, but it wouldn't need to look the same. You have to get the metal into the mold somehow, and that is a gating system. Generally speaking, the gating system is to be deigned to get the metal into the mold as quickly as possible while exposing the liquid metal to air (mainly oxygen) as little as possible. There are people that do the whole melt and pour in a fairly hard vacuum to reduce oxidization (mainly for things like difficult Ti parts). These set-ups can have gating systems that would generally be considered bad because they have removed the consideration of oxidization. So, depending on your set-up (gravity, pressurized, vacuum, etc.), you design your gating to get the metal in the cavity as fast as possible with the least amount of exposure to air, and with the ingate contacts located such that removing them and cleaning up the contacts is doable. In some situations you will also want to consider the heat of the metal coming in through the gates for flowing into thin sections as well as heat distribution for shrink/feeding purposes.

Gating is the system/path that gets the metal into the actual part cavity of the mold. This would include the pour cup, sprue, runners, and ingates (AKA - gates) at a minimum, but one could also include vents. Typically the cavity sections of a mold will have gating, risers/feeders, and the part. Flow-offs, slag traps, pouring basins, and other flow management features would fall under the general "gating" category, but more specifically are typically considered part of the runners.

Impact failure is drastically different compared to the slower loading of tensile testing. When you bend material you get an inside radius and outside radius of the bend. The inside radius is in compression, the outside radius is in tension. The outside radius is where the steel will fail. Impact testing is only really useful for situations like swords and axes. One could make an argument for chisels as they are struck by hammers/mallets, but they typically do not experience impact failure as tested by a Charpy Impact Test. Geometry factors are going to be huge players in such a test and will likely obscure any alloy/heat treat variables. No. Geometry is the big factor on force required for bending. Tensile strength directly relates to material failure. For example, a chisel that is 50% thicker than another will require a lot more force to bend, even if they have the exact same tensile strength. It would even take a lot more force to bend than the thinner one, even if the tensile strength was significantly lower than the thinner one (up to a point, of course).

Charpy testing is not applicable to these. Tensile testing is.

This was my initial thought as well. Not only would the data be hard to come by, it would be pretty useless due to the spec ranges. Joël is correct. Si is completely rejected by the carbides, and these levels do not produce significant carbides (though there will certainly be Cr in the iron carbide that is present). These tests that we refer to with foot pounds are Charpy Impact Tests, not something a torque wrench could do. There is essentially a calibrated pendulum that swings to break the sample and the energy absorbed in the breaking of that sample (as measured by the reduction of swing in the pendulum) gives the impact toughness. Notice that the 26C3 range is about 5 foot pounds, or just over 40% of the average value. I wouldn't read into this result very much. I am not sure how important impact toughness really is in any of this. I think tensile strength would be much more important for both knives and chisels.

I don't have a TIG welder, so I cannot, but in general it can be done.

This is a good illustration of why going by color for tempering isn't a good idea for optimal results. Clearly when you moved the heat source away, it wasn't getting hotter there, yet all of a sudden it started changing color. That is because the conditions changed and the oxide layer started forming to get those colors. And the colors kept changing as the part continued to cool off; so clearly we can see that color isn't locked into a specific temperature instantly.

You are at least 1000F too cold at 500F. You didn't get hot enough to really burn everything out. You should have a dedicated gating system, like the example I showed in your last thread. You want the metal to enter the mold smoothly but that never really happens, so you want a sprue and gate combo that smooths things out before entering the part cavity. And always fill the part from the bottom whenever possible. You cannot just pour more molten brass on a void and have success. While it is possible to get the part pre-heated just right then add more, it is very difficult to do right. TIG welding would be the easiest way to accomplish this type of repair. The vents you have look fine. The too low pre-heat and bad (non-existent) gating are your bigger problems.

The first video is pretty good, but I have a couple things I would like to point out. They dramatically undersell dislocations and other such defects. There are a TON of them. And dislocations can move pretty easily; it doesn't require heavy, or even noticeable deformation. These dislocations, especially when there is a localized high density of them, are the basis for new grains to form during grain refinement heat treat cycles, like normalizations. Secondary (and more) phases don't just form when there are elements that cannot fully dissolve into the lattice. Sometimes you get phases that happen to solidify at a higher temperature than others, so as a molten sample solidifies you'll get more than one phase (carbides do this a lot). Sometimes you'll have elements that are dissolved in the lattice that later precipitate out. They talk about this a little, but I don't think they were overly clear about it. I really like the expansion into the iron-carbon phase diagram. Too bad it was cut short here in favor of putting behind the pay wall. I love the older videos like the second one. I have been trying to convince some people to make some modern versions for steel castings. The impact test they are using during the grain refinement section is an Izod impact test. This is rarely used anymore, with the Charpy being more common. The very last sentence is incorrect. You won't end up exactly as you started (from a normalized state). You will have some work hardening (from the rapid cooling and thermal contraction) that wasn't there in the normalized state, and you should have a reduced grain size.

Those SEM pictures were taken with the department's brand new $5M FEG SEM (20 years ago, and had lots of bells and whistles). I used others that were so janky that we had to hand position magnets on the outside of the column to focus the beam. We'd help each other out and it was reminiscent of old CRT TVs in the 80s with the rabbit ears. One person would sit at the monitor and 1 or 2 others would stand around the unit moving magnets, sometimes having to hold them in place for the images to get taken. The pictures were not nearly as crisp, but there was rarely a line to use that SEM. Overall, I am pretty happy my days of SEM use are behind me. They are finicky on a good day, and there is a lot more to work on before getting to those nitty and gritty details.

Those are all optical. The vast majority of my SEM work was with polycyristalline silicon and friction stir-welded titanium (alloy 5111, it was my senior thesis project). Here are some SEM images of "high carbon steel" fractures (same piece/fracture surface) showing primarily cleavage (top) and a mix of cleavage and dimple rupture (bottom). My friends took these images (Richard and Laura, respectively); my role in the group was to get other images (Al, AlN, and a paperclip). Carbides are generally on grain boundaries in these types of alloys, but occasionally pop up in the middle of grains. Grain sizes can indeed get very small, but that isn't always a good thing. It can reduce depth of hardening, for one thing. Martensite is a diffusionless phase change that happens at the speed of sound for the material. So things happening within a grain go pretty quick and easy, and grain boundaries can get in the way a bit (which is generally why they are a good thing when it comes to the material failing).

If a thermometer can verify the consistent temp high enough from a heat gun, that (and some insulation) would be a pretty cool way to go.

All the light spots in the middle picture appear to me to be just reflected light. Those sections just happen to be angled such that light is reflected back at the camera to see that. If you want to get a good idea of grain size, you'll need to polish to about 0.05 micron or finer, then etched. I prefer about 3-5% nital for etching, swabbed and checked in about 10 second intervals. With that treatment you can see grains and thus measure/calculate grain size in un-hardened steel. For martensitic structures you can also see prior austenite grain boundaries, but we don't really talk about actual martensite grain sizes. As an example, below are micrographs of 4140 I took many years ago. You can clearly see the needle-like structure of the tempered martensite, but if you look closely you can also see a pattern to it all that follows/shows the prior austenite grain boundaries. There is a little proeutectoid ferrite in there as well. I'd prefer to show blade alloys, but these are what I have readily at hand and they still show what I am talking about. For fun, and it relates a bit to your usage as well, I found a micrograph I have from a failed chisel, though the alloy is unknown. This was also a bit soft at about 53 HRC (test at about 560 Vickers).

Looks like I joined March 10th, 2003. Definitely didn't know back then that this place would be so big and full of such great info.