|
Post by fe3c on May 14, 2017 3:20:45 GMT
Hello all. I'm wondering if you guys could provide me a bit of a reality check here regarding what to expect for blade flexibility. I'm working on two type XVIIIb blades. Frustratingly I have found that they will both take a set when bent by hand past about 45 degrees (applied over the length about 2/3 from the tip). I'm pretty sure there is something wrong with my heat treat process. I can explain the HT steps I'm using if someone wants to critique that, but mostly I would be interested to know what kind of flexibility is generally expected from this type of blade.
|
|
|
Post by Timo Nieminen on May 14, 2017 3:55:58 GMT
What to expect for flexibility? It's a cut-thrust blade and needs to be fairly stiff. It shouldn't be easy to bend it past 45 degrees. IMO, if it copes OK with a 45 degree deflection, that's enough. Perhaps your heat treatment isn't optimal, but the results seem adequate. I'm sure there are originals out there that will take a set from a 45 degree bend, and they were OK in their time.
Don't have a XVIIIb at hand, but playing with a XVa, which shouldn't behave too differently, I think it's stiff enough so that 45 degrees is plenty.
Sure, having it able to survive a bigger deflection is better, but it needs to be a functional sword, not a supersword. 45 degrees is enough. If, in addition to that 45 degrees, your edge is hard enough, then what's the problem? In that case, your heat treatment is fine. If your edges are too soft, then fix what's wrong.
How thick are your blades?
XVIIIb don't look easy to make, compared with blades with lenticular/diamond/hex cross-sections.
|
|
|
Post by fe3c on May 14, 2017 7:32:31 GMT
Sword A is 0.17" at about 12" from the tip. Sword B is about 0.140" in the same place. The subtleties of taper profile are something that I'm still learning, so they are probably not optimal in that respect. They are made from 1/4" stock. Blades length is 37-1/2" and total is 51-1/2" including uncut tang. Weights are 2.27 lb. for A and 2.37 lb. for B.
|
|
|
Post by Timo Nieminen on May 14, 2017 8:23:56 GMT
Thickness near the tip is in the right ballpark, thickness at the base is on the thin side. They take a set about 1/3 of the way from base of blade to tip? Might be flexing too much in the base half of the blade; thicker stock would help.
Still, 45 degrees should be OK.
|
|
|
Post by L Driggers (fallen) on May 14, 2017 9:04:42 GMT
What type of steel and what is your heat treat.
|
|
|
Post by fe3c on May 14, 2017 21:36:02 GMT
Mainly I'm concerned about the portion of the blade near the tip. The whole blade seems a bit too easy to deform though. For example, one of the blades came out of HT warped. After tempering, I was able to bend it back with my hands against my knee. Maybe the word flexibility is misleading here. The blades actually feel reasonably stiff. What I'm worried about is the deform-ability. The steel is 1065. Here is my HT setup. The heat is from two kilns stacked on top of each other. The top one has digital control, but the bottom one has two knobs, and no indication of temperature. There is a hole cut in the lid of the top one which I hang the blade through. My process was: - Heat to 1550 F
- Soak for about 20 minutes
- Quench vertically in about 45 liters of soy oil, with about 3 seconds between heat and quench
- Temper for about 90 minutes
During tempering, I used two K type thermocouples, one in the top kiln and one in the bottom. I tried to stay between 400 F and 415 F. It got higher a few times, but I'm confident it didn't get over 450 F. My current theory is that the quench is not hardening the the 1065 deeply enough. I'm afraid I will need to either go to a faster quenchant, or else a through-hardening steel like 5160. One more thing I should probably say is that I haven't actually heat treated a sword blade successfully. I've been at this a while now, but heat treating has been the biggest hurdle. Until I finish one it would be wrong to present myself as other than a beginner.
|
|
|
Post by fe3c on May 14, 2017 21:43:20 GMT
Thickness near the tip is in the right ballpark, thickness at the base is on the thin side. The two swords have different tapers, so the thicknesses near the tip are different. They are both 1/4" at the shoulder and tang.
|
|
|
Post by Timo Nieminen on May 14, 2017 22:08:10 GMT
As you said before.
I'd expect about 8-10mm at the base of the blade. This will give you over 2 to 4 times the stiffness near the base, compared with 6mm.
But as I said, it's a stiff blade, and surviving a 45 degree bend is good enough for practical use. How hard are your edges?
|
|
|
Post by fe3c on May 14, 2017 22:20:51 GMT
As you said before. I'd expect about 8-10mm at the base of the blade. This will give you over 2 to 4 times the stiffness near the base, compared with 6mm. But as I said, it's a stiff blade, and surviving a 45 degree bend is good enough for practical use. How hard are your edges? Interesting. I know that some of these blades were historically pretty thick at the base. Quarter inch is what I can get though, so I actually ground them with a hexagonal section for a few inches to compensate. Hardness is hard to gauge. I can mark it them with my file, but the surface is definitely harder than when I was grinding and draw filing.
|
|
|
Post by fe3c on May 14, 2017 22:25:46 GMT
Maybe I will just finish them and try some heavy cutting. That should provide me some feedback.
|
|
|
Post by Timo Nieminen on May 14, 2017 22:49:56 GMT
Hex section at the base will help the stiffness.
Good plan to just finish them and try them out. They sound OK compared to historical blades. Shallow hardening was common. Blades with soft cores were common. (Achieved through laminated construction, shallow carburisation, and shallow hardening during heat treatment.) Your blades would have plenty of historical company.
You might like to get a set of hardness testing files. Fairly cheap, and will give you a fair indication of hardness (not accurate, maybe only within 5HRC, but that's a start).
|
|
|
Post by fe3c on May 14, 2017 23:26:48 GMT
Hmm... it sounds like it might not be as bad as I thought then. Thanks for the insight. I'll look at files too, although I've also been playing with the idea of building a hardness tester.
|
|
|
Post by L Driggers (fallen) on May 15, 2017 6:15:11 GMT
If you are able to bent the blade over you knee after tempting you heat treat is off.
Don't wait 3 seconds to do the quench get in as fast as possible.
Do your blade skate a file after the quench, if a file will bite in easy you are not hard enough.
|
|
|
Post by fe3c on May 15, 2017 15:58:10 GMT
Three seconds is my high estimate. The tank is as close as physically possible to the kilns, and I work as fast as I can. With practice, I think I could do it in two seconds consistently. The issue is that the tank is a small target, and I'm trying to plunge it in as vertically as possible to avoid warping.
It seems like 1550 F should allow for 75 degrees of cooling margin, right? The charts I've seen say 1475-1550.
|
|
|
Post by Lukas MG (chenessfan) on May 15, 2017 16:50:40 GMT
Depending on steel and heat treatment procedure, it is absolutely possible for blades to take a set quite easily after tempering as long as the oxides (and the decarburized surface) aren't ground off. I experienced this myself and like you believed the blade to be poorly heat treated but Peter Johnsson and some other gents over at Don Fogg's forum cleared that up. Go over the blade again on the sander and remove some material, then try again.
I am not a fan of flex testing. I tells very little valuable as far as the performance of the sword afterwards is concerned. A blade should be tested doing what it is supposed to do. Cut stuff up, hit things with it. Feeling the flex is very important to see how the blade flexes and in what parts (you want to eliminate "dead areas" and want the blade to flex in a smooth curve in the foible, without any "steps") but bending it past 45 or even 90 degrees and seeing it spring back (or not) is not a very useful test.
|
|
|
Post by Verity on May 15, 2017 19:06:40 GMT
^ This.
What matters is the nodes of percussion and harmonics to determine how the blade performs when used as a blade. Just a solid bend may not be happening at a particular node where its natural harmonics are.
I've always heard 30 is sufficient let alone 45. Past that is just unrealistic and is abusive IMO.
Do the nodes operate as expected? Do the harmonics work as expected? If yes. You're good.
|
|
|
Post by brotherbanzai on May 16, 2017 19:27:42 GMT
There are several things to consider here; the steel, the cross section, the heat treat, and the method of testing flexibility.
1065 is a tough steel, but not an exceptionally hard steel. Generally speaking, more likely to take a set as opposed to breaking or cracking under stress when compared to a high carbon steel. There isn't enough carbon in it to reach the full martensite transformation potential of the steel, so you will have some pearlite left in the final mix. Tougher = not as hard, harder = not as tough, just how steel is. I personally wouldn't prefer to use 1065 for this type of blade, but it's perfectly usable for one.
1/4 inch thick at the base is pretty thin for a type XVIIIb, especially one that's 51 1/2" long. A long, narrow, and thin blade is going to be less able to resist deformation, in addition to being subject to more leverage by virtue of it's length. I don't know where you are but steel of various thicknesses is available by mail order anywhere in the continental US. There are two options to address this potential issue; get thicker steel, or make a different type of blade.
20 minutes is too long of a soak for 1065. Taking the steel to the high end of the heat range will make it harder to cool fast enough to harden. High heat for an extended time will also promote grain growth. If you are reading the heat only from the top portion of your heat treat kiln, it's possible the bottom section is significantly cooler since heat rises. Uniform heat is fairly important for an optimal heat treat.
Also, I have personally witnessed what Lucas mentions about the steel taking a set if the post heat treat surface has not been ground off. If you have a layer of scale and a layer of decarb, you can essentially have a reverse differentially hardened blade; hard in the center and soft outside, and so easily prone to take a set. Additionally, if it was overheated and/or not cooled quickly enough during the quench, it could be soft on the outside, then hard, then soft again at the core. Harden the blade, temper as soon as possible afterward, then grind the blade to remove any scale as well as any decarb surface, then temper again to remove any retained austenite.
Also, I will disagree with the idea that 45 degrees is plenty, because this will depend entirely on how you are flexing the finished blade. If you are clamping it in a vise and throwing all your weight into it to get it to bend that far, then absolutely, 45% is plenty. At the other extreme, if you are holding the blade out in front of you with one hand at either end and are able to easily flex it to the point that it will bend past 45%, then it is almost certainly too easy to deform, unless you have Arnold Schwarzenegger like upper body strength. This type of blade should be very stiff, so it should be pretty difficult to flex. Basically, if it flexes easily to the point of bending that's bad. But it can potentially be impossible to determine if the test is done before the blade has been ground close to final shape.
One final thing; extreme flexing can very easily be damaging to a blade even if it does not take a set. Not that 45% is necessarily extreme, but it can be depending on the blade type. It's better to test the blade to be sure that it hardened properly by any method other than extreme flexing. You can test a blade to see that it will flex through the blade's normal range and return to true, but that normal range will vary with the blade type. Most Euro blades, and especially thrusting blades, should be fairly difficult to flex very far. A blade should be flexible enough to resist damage from a poorly executed cut or hitting an unexpectedly hard target, but a whippy noodle doesn't usually make for a good blade.
|
|
|
Post by Lukas MG (chenessfan) on May 16, 2017 20:03:28 GMT
Good post, agree fully. It is very important to note that flexibility can vary hugely between blade types and a general rule like "30 or 45 or xy degrees is enough/good/whatever" is impossible to give. Hence my suggestion to test the blades doing what they are supposed to do. A XVIIIb is a thrust-oriented blade, it needs to be able to deliver deeply penetrating and powerful thrusts. Set up a bag with some old clothing or such and thrust into it, for example.
|
|
|
Post by fe3c on May 17, 2017 3:20:38 GMT
Wow, thanks! These posts give me a lot to digest. I see no one has questioned my use of soy oil. I am almost tempted to try water again, but I have broken blades that way. brotherbanzai, thanks for the analysis of the steel and what is potentially happening during the HT. That makes sense. I don't know where you are but steel of various thicknesses is available by mail order anywhere in the continental US. I'm located near Seattle, WA. Steel thicker than 1/4" in 50"+ lengths has been hard to find. I am considering getting some 1/4" 5160 from Admiral though. If you are clamping it in a vise and throwing all your weight into it to get it to bend that far, then absolutely, 45% is plenty. At the other extreme, if you are holding the blade out in front of you with one hand at either end and are able to easily flex it to the point that it will bend past 45%, then it is almost certainly too easy to deform, unless you have Arnold Schwarzenegger like upper body strength. It is closer to the latter. I can put a set in the tip half of the blade with my hands only, with no assistance from a vice or bench. They don't feel whippy in motion though. Depending on steel and heat treatment procedure, it is absolutely possible for blades to take a set quite easily after tempering as long as the oxides (and the decarburized surface) aren't ground off. This is really interesting, and not something I would have guessed. I find it interesting that you guys are emphasizing the stiff, thrusting nature of the XVIIIb. When grinding, I tried hard to balance stiffness with weight. Actually, I felt like both my blades were a a bit overbuilt in the strong, but decided to go with it anyway. That is from comparing the weight with numbers I've found for antiques, and also modern longsword blades like the Hanwei Tinker. They are probably wider and more cut oriented than some. Of course, I need to actually finish and use them to tell.
|
|
|
Post by L Driggers (fallen) on May 17, 2017 4:17:54 GMT
Call Admiral Steel they have thicker than 1/4in in 5160 and wider. If you go through their main catalog you will see the sizes they have. I've ordered wider and thicker from them.
|
|