Confessions of a bladesmith; secrets revealed! (Finished Pictures added)

[Kevin R. Cashen]

Originally posted by Robert.B
I currently dont own a power hammer yet attempt to forge pattern welded blades but dont turn out as good as i'd like, would the ultimate fact be i'll need a power hammer...also what do power hammers go for, or what would you recommend as a starter hammer or is it possible to make one???

if you can help a hobby smithy thanks.
Robert.B

I normally refrain from using this forum for a gateway to my web site but many more deatails of the welding process, including some answers to these questions are covered in the newest section that I brough online:

Cashenblades.com-Pattern Welding

Al is right, presses are much more available and easier on the neighbors these days. I think the pure compressive forces delivered from a press could result in better welds as well. Pattern welded steel was made for centuries without power hammers but those folks also had loyal strikers to help out and they live near as long as we will have to with our crippled up old hands. I make no apologies for using power hammers for folding up steel, since without them i wouldn't be diong it at all anymore. I narrowly escaped carpal tunnel surgery and may still need it yet.

[Kevin R. Cashen]

Originally posted by Arik Estus
Hi Kevin,
Bradleys are out and so are nazels from what I can see. Are there any recomendations you would make of the newer products on the mkt? ...

As I mentioned in another post I go into my opinions about hammers for welding a bit on my website, but I would recomend the heaviest hammer you can get or afford for pattern welding. This is not for drawing out or any sort of a machismo thing. I just have come to believe that really good welds result from good sqeezing compression. For this presses are ideal, they just seem to take too long in other operations for my liking (believe me , that this is just my opinion). When I have welded on power hammers that are 25lb or 50lb I noticed that they rapped and slapped the steel around more than squeezing it together. When hand hammering forge welds, sharp rapping blows usually knock the steel apart more than squezzing it together. When I have students having problems with this I tell them to get a bigger hammer in their hand and bring the blow down and continue to push down for a half second after they make contact. Their welds usually improve because it keeps them from rapping the steel too sharply.

This I translate over to hammers. I like hammers that have a good ram weight that will continue the compression beyond initial contact. I don't just want to hit the steel, I want to squeeze it!

[Arik Estus]

Hi Kevin,
Thank you Sir for your advice. It seems your reasoning for the power hammer is the one I have come to. Old sucks.

To misquote Heinlein: "Isnt it amazing how mature wisdom resembles being to damned tired".

[Kevin R. Cashen]

Originally posted by Michael Stora
Great post Kevin. Could you elaborate on how you do the twisting?

Thanks,

Mike

I realized that I did abreviate one of the most difficult and involved parts of the process. This is mainly because I have no pictures since my hands are always entirely too full to include photography in that step.

It really is more involved than just simple twisting. To start with, for my usual twisting in single twist patterns I prefer not to leave any corners on the bars. The corners make great points for stress cracks to form and also leave the piece looking like a coarsely threaded rod so that when you forge on the ridges they mushroom over and make cold shuts and other messy things. I have noticed an increase in weld failures in the grooves when they are forged flat also. Having said all of this there a some mitigating circumstances with this particular technique that necessitates my leaving the darned things square instead of octagonal or rounding liken I would prefer.

It is quite critical to get all of the twist even from one rod to the next. I never get a perfect mirror image but if you don’t keep it as close a possible it will get quite bad. I always count my twists very carefully while doing them, not only to make sure that I did six on this one as well as on that one but also to make sure the I end on the proper face or edge.

You see with this type of pattern you can end your twist 360degrees from where you started, you can end 180degrees from where you started (if you must) but you must never end at 90 degrees from where you started. For the straight sections in the final pattern the edges of the billets are traditionally used so that you see the 10 layers on edge forming straight lines. Opposite of the edges you have the top and bottom faces that will have a VERY coarse flat patterning. If you start a twist with and edge facing you, you MUST end the twist with and edge facing you, or it will look very bad.

Twisting must be done very hot. I prefer near welding temp. I can usually get a twisted section done in two heats. If you are twisting and the steel begins to cool, don’t ever be tempted to say “oh heck I can go round one more time and have this done without the hassle of the reheat”. You will notice that it will get harder to twist and if you continue all of the sudden it will get really easy again, and you are really done, as in – start all over from scratch.

Set your wrench to the size of your bar before you are ready to twist. I don’t know how many heats I have wasted trying to adjust my wrench. I use a pair of vice grips with an extra handle welded off the front. Life is a whole lot easier if you have a handle to either side of the work in the center. With a single handle I always end up bending the work to one side more than twisting. I have tools made from pipe wrenches with a handle welded on the upper jaw, but if you know how a pipe wrench works, this arrangement will defeat the gripping mechanism of a pipe wrench, I have never understood how this could work well. I have thought that if you were to weld a handle onto the frame behind the upper jaw that then it would work because then both handles would work with the gripping action instead of opposing.

To get the sections consistent I have made up a vertical vice with jaws that are exactly 2” wide to keep things spaced, but I need to deviate from this with this design due to the distortion of the pattern in the changes of cross section.

What really stinks is when you twist counterclockwise when it should have been clockwise! Either you twist the other rod the same (or actually opposite) and tell the customer that the change in pattern had a spiritual significance or you start all over. At any given time you will find columns of letters and numbers written in soapstone on my coal forge hood. It is next to my twisting vice and I am keeping track of sections and twists.

If during all of this you find that you misplaced your wrench and you have sections that are longer than they should be you may have to get just that section very hot and quickly upset it (hammer from the very ends to make it shorter and thicker) in order to salvage the matched pattern. Quite often what this will do though is set all of the others after it off kilter slightly so they need to be adjusted by either drawing longer or upsetting also.

One way that I have found to keep the pattern even after you have the twists matched and you need to square things up with more forging (you cannot hammer one different form the other) is to forge them at the same time. My dies are wide enough from the side for me to heat both rods at once and then hold one in either hand and hit them 3-5 times on one face and then quickly turn them 90 degrees and hit them 3-5 times again. This ensures that they are forged as identically as possible.

I could go for pages on this whole twisting thing alone. It is the biggest pain in the @## in this type of pattern. The eight bar core “Sutton Hoo” style patterns will keep me just twisting for 12 hours or more.

[Michael Stora]

Thanks Kevin,

That does sound like a very involved part of the process. The part I was wondering about is how you keep the pitch (degrees of rotation per unit length) of each twist nearly equal. I assume the temperature as well as the size and heat distribution of the hot area plays a big role in the length in which the twist occurs. Also the spacing between vice and wrench . . .

Mike

Originally posted by Kevin R. Cashen


I realized that I did abreviate one of the most difficult and involved parts of the process. This is mainly because I have no pictures since my hands are always entirely too full to include photography in that step.


<snip>

[Kevin R. Cashen]

Now the initial process of stacking and welding is repeated, but this time we will not be stopping at 10 layers. Instead the process is continued through 4 more folds:


This gives us a count of 160 layers. In this process I leave nothing to chance and in between each fold the billet is ground clean on the would be weld surfaces.


When 160 layers have been achieved the billet will have to be drawn to a little over twice the length of the desired sword blank. On long swords this can be six feet or more. It can get quite trying to handle hot steel that long and narrow. I have a hole cut in the wall behind my forge so that I can accomplish this. The long edge piece will pass through the forge and through the wall, into the other part of my shop while drawing the edge. Fortunately the makers of the Bradley had the same ideas and gave me a similar hole through the center of the hammers body.



When the edge steel is to length it is then bent in half to form the continuous wrap that will encase the core bars.

The edge is then heated and fitted as tightly to the core bars as possible.



Since I like to use L6 in my steel I have found that this step requires some care. L6, and especially L6/O1, that has cooled from welding temperature without subsequent thermal treatments is quite brittle! I have been greeted by the horrible sound of breaking metal while hammering that hot edge to fit tight around cold core bars. At this point you are several days into the process and now you must go back and make another core bar and hope it matches the other, but most likely you will be making two more matched ones. The answer is to either cycle the rods down or better yet, warm them up a bit, along with the edge, for this fitting.

You want the tip to fit as tight as possible. The most critical area is where the 3 different weld seams come together at the tip. This will always want to form a little triangular void that can be difficult to close.

I am being extra cautious with the steel since I have had a couple of “bad weld days” lately. It is nothing drastic, but if I have to stop and weld a fold twice instead of once I get a little edgy like things are not as “perfect” as they could be. Perhaps it is superstition on my part or perhaps there is something to it due to air density and the way the forge works, but in the winter I prefer to do my welding on very crisp colder days. Days that are above freezing and are wet or have precipitation seem to be bad welding days for me. In the summer time it doesn’t seem to matter though, so it could just be my imagination. There is evidence that high atmosphere moisture can affect rates of decarburization, whether it has anything do to with weld quality, who knows. I do know that intuition and instinct are big factors in proper forge welding and if none of the welds “feel” good you are probably better off doing some finishing work on the other side of the shop that day. Most problems that I cannot see while the steel is hot will not reveal themselves until the polishing stage though, so all I can do is hope for the best. Most of it is just “feeling” and when you have customers waiting telling them that it will be another week because it didn’t “feel right” is not always possible.

[Kevin R. Cashen]

Originally posted by Michael Stora
Thanks Kevin,

That does sound like a very involved part of the process. The part I was wondering about is how you keep the pitch (degrees of rotation per unit length) of each twist nearly equal. I assume the temperature as well as the size and heat distribution of the hot area plays a big role in the length in which the twist occurs. Also the spacing between vice and wrench . . .

Mike

Your assumption is correct. It can be controled through eveness, or localized heating as well as placement of the wrench. This is where vice grips come in handy since twisted areas are rounded and cannot be gripped with a square jawed wrench. If you need to "adjust" in the middle of the twist you can snug down the vice grip adjustments and lock on.

[Dennis Boas]

Kevin;

Thanks for taking the time and effort to give us a look behind the curtain at how a pattern welded sword is really created. Seeing the sword at each step of its development is great.

[RogerA]

You can find more interesting reading by Kevin (and others) in SFI's Online Magazine here: http://swordforum.com/magazine/

May I suggest The Road To Damascus: http://swordforum.com/forge/roadtodamascus.html

It may be unnecessary to post this, as all of you may know of the mag, but then again maybe not.

RogerA

[S Burst]

Amazing work;

Will this sword be for sale when its done?

[AnthonyWoo]

Very good thread.

[Thomas McDonald]

Hi Steve

This sword was ordered by our Forum Administrator , Dennis Boas , and will be his ( or his wife Jane's

Mac

Originally posted by Steven Burst
Amazing work;

Will this sword be for sale when its done?

[Dennis Boas]

The sword is a present for my wife.

[Thomas McDonald]

I have a big red apple for the teacher , as I sit at my desk, front row , waiting for Prof. Cashen to arrive!

I need my daily pattern-weld fix ....

....... Kevin , help me ,please ! Mac

[Kevin R. Cashen]

Now the forge is fired up again and we get ready to weld this bundle into a solid sword shaped piece of steel. Things need to be held tightly in place so that they don’t misalign in the welding process. If misalignment occurs there are a couple of choices, you can grind the ridges down and stock remove the misalignment away, which should only be done slight misalignments since it can affect the patterning. The other option is to beat it back into line, but the drawback here is that the things are not real tight yet and it will more often than not just pop the welds back apart. This is not so much a problem since you can then realign them and weld them again. But I have noticed that it is not uncommon for a split seam in the middle not to want to re-weld as well. And the more times you try to re-weld the less likely it will take. So if you can’t get this taken care of in very short order you could be… you guessed it, starting all over from scratch!

So it is very good to keep everything in line to begin with. The best way I have found to do this is to “band” the stack. I like to use strips of 1/8”x 1” steel to make bands that wrap around the stack very tightly and hold things in place. This works well because it is flexible. The bands can be tapped into other positions as you weld and they allow expansion of the individual components. It should be obvious why tack welding could only be done at the very ends where the tang will be, since whatever crap you fuse or introduce to the steel will be in the final pattern. Tacking on the tang ends can be done but it presents certain problems. The edge will want to expand differently from the core and pull way from the tip and also cause it to bow and warp away from the core when held rigidly by the welds, and those welds will normally just pop at the first serious welding blows.

So this leaves banding as a very good option. For long swords I like to use three bands, for short swords two will work. As I weld a section, starting from the tip and working back, I just tap the bands back farther and farther until all of them drop off the tang end.
On this day I was feeling a bit cocky so I kind of held things together with a gloved hand and zipped it up. I have done enough that this works alright when I am doing a short sword and am still in practice, but not quite as well when it has been a little while and I am feeling “cocky”. It all came together but I had to work for it. Bands, definitely band it!

The first welds that I do are about 3” back from the tip in order to tack a 4” wide area down tight before bringing that stubborn tip into submission. The welding blows need to be lighter yet firm. This is the point when you find out if you have your core and edge bars left at the right thickness. Many folks look at mine and ask why they are still so big before welding. If your core bars are smaller you could end up in the unenviable position if having your stack much thinner than it is tall. Have you ever tried to hammer down sheet metal edge on? It can’t be done. There is a point when the thickness to height ratio will only result in folding or bending. I have heard people try to give exact numbers to this ratio but I have always found folks who could either defy them or not come close to meeting them. This effect is not so bad with a for bar stack like what I am doing here but with an eight bar with double layers it can get very difficult to control.

Anyhow, if the stack folds or cups under the edge compressions (much more likely in a power hammer) you are in SERIOUS trouble. Without immediate, fancy maneuverings you will be – starting all over from scratch. You see what will happen is that the inner weld surfaces will take on the slant of that new folded or cupped configuration and when you hammer the stack flat again it will not matter because those surfaces will just pull it back into a cup again, or slide off from each other and misalign instead of welding. So buckling, cupping or folding must be avoided at all costs.

After the initial side welds are done, I go to the tip. The tip should be welded from the very point back, with and action that pulls the metal back into the sword. The opposite will result in pushing the tip forward and creating a hard to close gap. My first welding blows are not done with a hemmer at all. When the tip is hot enough I pull it out and drive it tip first into the anvil (short swords) or the floor (long swords). This is followed by an angled blow to either side. This repeated a couple of times and then I get on it with a hand hammer and weld it down real tight. I will not leave the tip until it looks like a completely solid piece steel with no seems or ridges it cools. If this is not done now, it cannot be done after forging the point.

Here I am finishing off a tip at the Ashokan seminar last September:


The rest is just a series of welds up the sides of the stack until you reach the tang. In between edge welds, I will do plenty of work on the flats. As I have said, I don’t want to see the pattern, lines or ridges at this point. I have found that the pattern will not be until you grind the surface clean and let it getting air, if all is welded tightly.

Here it is halfway:


Once again this all seem fairly simple for a four bar, but with the eight bar migration era stuff it is much more 3 dimensional. You need to weld both from the side as well as the flats and be darn sure that you do not trap any voids inside, or misalign things too much.

Final welding heats:


One final note on this step, I have found that with the blank still thick that it is bad to just let it cool from welding temp down to room temp. I suggest cycling it down a few times getting cooler as you go to below critical before stopping for the night. This is very critical if you have to go back and weld a section in the middle after it has cooled, trust me.



But for now I am usually happy that it all appears to be one piece and quite tired out so the forging into a sword waits for another day.


Looking at it here it is kind of sad to think that we will be cutting it into two pieces, since it is a great lookin longword just as it is, but we are going for a slightly shorter leaf blade and perhaps a long dagger or shorter sword.

[Robert.B]

ooooo, thanks for the picci's and valuable info. hope you dont mind but i archived this info on my computer for future reference.

Robert.B

[Joe Moffitt]

Thanks for going to so much trouble for us. This is great info.

[Joel Whitmore]

This would make a great article for SFI. I loved the one Randal did a couple of years ago about the katana. I was wondering K evin, if you cut the sword in half, what becomes of the other pieces? Can you reuse them for other projects such as knives and such? beautiful sword too Kevin.

[Erin Burke]

Originally posted by Kevin R. Cashen
Looking at it here it is kind of sad to think that we will be cutting it into two pieces, since it is a great lookin longword just as it is, but we are going for a slightly shorter leaf blade and perhaps a long dagger or shorter sword.

If you cut this peice in half, won't only one of the pieces have the edge steel (160 layers) at the tip?

efb

[Manouchehr M.]

fascinating thread, thank you for sharing.

Regards

Mnaoucher

[Kevin R. Cashen]

Originally posted by Erin Burke


If you cut this peice in half, won't only one of the pieces have the edge steel (160 layers) at the tip?

efb

The tip of the cut piece can be cut into a convex vee formation and re-welded. As in the image below.

1. This is the problem that you pointed out for the next blade.
2. Is the cut that will be made into the center of the end. It is important to make the cuts mirror image convex of what the final tip shape will be. If you cut it straight it really accentuates this fix job by the distortion of the pattern.
3. The forging together of the end.
4. The final affect.

There will be a fine seem at the tip that some people balk at as non-traditional or not authentic. That is why I use it on smaller pieces and not on orders that my customers are paying top dollar for. There is absolutely no difference in structural integrity; it is just a preferential cosmetic thing. Sometimes you need to look pretty darned close to even see the difference.

The dagger pictured below was done this way


Looks fine to me, and its owner has no problem with it at all. Dennis would be interested to know that this is the little sister to Bench Eater; they came from the same billet.

[Jeff Berger]

Wow, thanks a lot for sharing this bit of bladesmithing.

this would make a great addition to the SFI articles section (Bit barren last I checked, most things in SFI university and in the magazines, so sad that there are only 4, they have so much stuff!)

You might want to consider compiling it and sending it in, and thanks again

[Dennis Boas]

Always glad to meet another member of the family. I can definitely see the family resemblance

[Robert.B]

Kevin you referred to the proper migration stuff like sutton hoo using 8 bars etc, if you've made any that you have pictures of would you be able to post them...id really love to see those from your works.

Robert.B

[Kevin R. Cashen]

Originally posted by Robert.B
Kevin you referred to the proper migration stuff like sutton hoo using 8 bars etc, if you've made any that you have pictures of would you be able to post them...id really love to see those from your works.

Robert.B

I have some old slides of one going together if I can find them I cleaned out the entire office recently and what makes it worse is that I had some help from my wife
Have you ever tried to find anything after your other half has helped you clean up?

Two or four bars were used many times in the past, indeed just about any combination of bars and twist have been observed, but during the height of pattern welding from 400-600A.D. (finally declining around 800A.D.) two rows of four with and edge wrap seemed to be all the rage and no nobleman who was anybody would have anything else

I do have this readily available:


The photo on the bottom is cut and pasted so the bars above do not correspond, nor are they to scale with the sword on the bottom. in fact if you look you will see that the sword on the bottom is a two bar core like what I am building in this thread, where as the rods are for the two layer eight bar type.