Pattern Welding

 

Steel Selection Once Again

 

Although we have already discussed steel selection for blades, now that we are looking at pattern welding we need to reexamine our steels and the criteria for which we choose them.  Combining any of our  previous steel choices into one blade will require the narrowing of our parameters yet again. 

 

  I like to compare dealing with steel, in a proper way, to handling snakes.  Your chances of getting bit are much less if you deal with simple, docile types. The more snakes you put in your grasp, the trickier things get. Some species get a little aggravated in the presence of others, resulting in a bit of a conflict right there in your hand. Now, without going into why you shouldn’t use scrap or mystery steel all over again, just imagine if some joker turns the lights out while you are handling your bundle of vipers!

 

So now we have some new considerations that we need to look at before our final steel selection. First and foremost whenever making blades should be function.  Will the steel we are considering make a good functioning knife blade?  If not, why are we watering down or muddying our pattern welding with it?  Once we dispense with the myth of hard and soft layers due to carbon content, the only reason left for this is appearance.  What is more important to us appearance, or functional quality?  If “pretty” takes precedence then we can also dispense with the headache of heat treating and just hang it on the wall.  Truly we are wasting our time heat-treating a steel that will never gain any superior qualities from it.   So a good guideline would be to never dump anything into our steel that we wouldn’t make a blade out of all by itself.

 

A second consideration would be compatibility.  There could be some issues from different rates of expansion under forging, but mostly consider the heat treat.  If you have two steels with completely different curves for hardening, which one are you going to compromise in the heat treat?  I cannot bring my self to compromise on heat-treating.  And what will the compromise be?  Will you let one form pearlite and not fully harden or will you shock and over stress the other?  This choice will often lead to distortion or even the possibility of cracking when one steel starts to move in transformation and the other does not.  I have seen shallow hardening steel lose the fight with deep hardeners and get turned into a corkscrew and even pulled literally apart in thicker layers.   

 

Third would be appearance.  It does have to play some role in this since perhaps the greatest quality that pattern welded steel has over homogeneous material is its beauty.  For this we need to determine the look we will want in our steel.  High contrast (black /silver), medium contrast (silver/gray, black/gray), subtle contrast (gray/gray, silver/silver, black/black), or whatever degree of subtleness or glitz you prefer.  For this we need to look at alloying elements more than anything.  Various elements will etch differently. Manganese=black, Nickel=silver, Chromium=light gray, Iron=darker gray and so forth.

 

 

  

Prepping the billet

 

As I have mentioned, I like to weld larger billets with thicker layers and there are good reasons for this in billet preparation as well.  I leave nothing to chance in my pattern welding so I grind all of my initial weld surfaces clean and smooth.  If I took the popular route of many thin layers to prepare, say ten or more, I would have 18 or more surfaces to prepare (the two outer surfaces are not as critical since they are not the initial weld surfaces).  By doing five thick layers, I have only to prepare eight surfaces.  I like to make my billet wide after the initial weld but for that first weld I stack my layers narrow and tall for the purpose of cutting the surface area to be cleaned down to a minimum.  This also equates to lesser surface area to be contaminated before the first weld.  My billet layers are 6x1x3/4 this equals 12 square inches per layer, multiply this by 5 and you get 60.  Subtract 24 for the two outer surfaces and I have 48 square inches of surface to prep on my billets.  Many other folks start with something like 1.5 x.25x 6 for layers this comes out to 18 inches per layer, that would be at least 144 square inches to prep if I only used the same five layers, but such a billet would only be 11.25 cubic inches compared to my 22.5 cubic inches with much less effort in the preparation and 84 fewer inches to become contaminated before the first weld.

 

Now to prep these surfaces I grind any of the oxides or pitting from hot rolling, pickling or forging from the surfaces to be joined.  I do this with a 36X belt so that the grinding scratches run 90 degrees to the length of the steel.  This creates micro channels to pull the flux in and eject it with any contaminants across the shortest distance from the center of the billet.

 

  The layers are now assembled in the following order; O1, L6, O1, L6, O1, and then a small bead from the arc welder is used to tack the layers at the four corners and a three and a 3 ½ foot handle is welded on.

 

The gas forge is adjusted to a carburizing flame before the billet is placed into it to come slowly and evenly up to heat.  I have welded in both coal and gas forges, either one will produce useable material, but gas will accomplish this with more efficiency, in less time with less chances for problems or flaws.  For forging to shape coal can have much more versatility, but for making my pattern welded steel I will only use gas.  

 

The carburizing flame is rich in fuel and uses up any free oxygen in the forge so that there is a noticeable flame coming out the door of the forge as the unburned fuel combines with the outside oxygen.  This leaves no oxygen to interfere with my welds.  

 

After some time the layers will begin to take on a dull red glow.  Almost invariably the outer layers will heat quicker than the inner ones so it is important to heat slowly and evenly.  Once again, this problem is worse with many thinner layers.  Many problems in the first welds result from uneven heat distribution between the layers. As soon as an even dull glow is noticeable the flux will stick to it so this is time to put it on in order to eliminate any remaining opportunities for oxidation.

 

  Now is time to discuss yet another reason as to why I favor thicker starting layers over thin.  Whenever heating steel, you face the constant threat of decarburization.  If this decarburization affects the steel to a certain depth in a given atmosphere, regardless of the thickness of the steel, a far greater percentage of a thinner layer is robbed of carbon in than in a thicker one before they are welded solid. Also, when they are welded solid, many more decarburized weld zones are sealed into the billet right from the start.

 

The billet is heated to the welding temperature.  Recognizing a proper welding temperature is one of the more important skills in pattern welding. It is a skill that can only be developed from experience over time. The color of the steel, the appearance of the flux and just plain “feeling” when it is right, all come into play in determining this.  Since there are days that any of these factors can be off, it is best not to rely on any single one.   I like to see the flux completely fluid and runny (more an indicator of proper atmosphere than heat) with little bubbles scattered across the surface.  At a good welding heat the bubbles will start to ”skitter” or zigzag across the surface of the billet.  I could say that the steel is a bright orange in color but that is meaningless under any lighting conditions different from my shop. 

 

  I do not go to “white” heat and I NEVER spark my steel. My experience and research has shown, and my belief is, that the lower the temperature you can successfully weld at, the better it will be for the steel.  Excessive decarburization, abnormal grain enlargement and all-out cracking or crumbling can be the result of excessive welding heats. At higher temperatures oxidation and decarburization can also interfere with welds.   I know that I weld at temperatures 200-300 degrees cooler than many other smiths I have talked to.  Without pyrometers it is hard to tell exactly what these temperatures are so for the most part we rely on old-fashioned means.  You want a good even heat with a nice bright glow.  The flux will flow freely and bubble nicely, not white hot. That is only unnecessarily hard on both your steel and your eyes. When the time is right, the steel comes from the forge as quickly and smoothly as possible and goes into the waiting power hammer dies.

 

 I prefer to use a power hammer with flat dies for my welding.  Pattern welded steel needs to be “squeezed” in order to weld nice and tight.  Good compression is the key to bringing the weld surfaces firmly and completely together.  For this a hydraulic press is probably the best, but a larger hammer with a good amount of compressive mass behind its blow will also do the job.  I find that hammers less than 100 lbs.  tend to rap and tap, knocking the layers apart almost as quickly as it sticks them together.  A good indicator that your hammer may be too small for your billet size is if when observed from the end the billet is mushroomed on the top and bottom giving it an I-beam look with concave sides.  The hammer blows are not heavy enough to deliver the force to the center of the billet and give it the necessary squeeze. A proper blow will squeeze the center out and give the sides a convex rounding like a piece of dough being pressed.  The previously described concave caused by insufficient blows could even result in expansion of just the outer layers to the extent that weld shearing occurs.

 

  This all can limit the size of the billet for hand hammering of the steel, but it is worth mentioning that the simple patterns that rely on randomization of the pattern have always been more pretty when done with a hand hammer.  For these patterns I have plates with ball bearings welded to them that I slip under the die of my hammer and beat the steel up real good with.  

 

  When doing the weld, I start at one end of the billet and take quick 1 ½ “ bites until I reach the other end. I could do 4” at a time from the side, and 6” from the front of my dies but I like to squeeze the flux, and anything in it, out as I go.  When I reach the other end, I give it a few extra hits to seal it down good, as it is this area that most easily separates when drawing out. 

 

There will be slight misalignment of the layers and unevenness down the sides. The two steels will compress at different rates and as it is flattened this results in ridges down the sides of the now solid billet.  So I now flip the steel on its side, and give one good flattening and smoothing pass, once again in 1 ½” bites.  If this isn’t done at this point, the aforementioned ridges could fold over and cause cold shuts or inclusions when it is done later. Some stop and grind these ridges off but this can be fixed with the hammer if it is done now.  Another very good purpose that this serves is to test the welds that have just been done. The sideways compression will have a tendency to separate any weak or incomplete welds.  

 

Now for safety’s sake I reflux and take another welding heat and then draw the billet out to twice it’s original length.  During this whole process there will be oxide scale collecting on the hammer dies, or anvil if done by hand.  It is very important to keep the dies or anvil clean of this scale, in order to have a smooth, pit free and clean surface for the next weld.

 

When the billet reaches twelve inches it is time to fold.  It will also have widened out to around 2 inches in width in the process.  This is good, I try to now get my billets nice and wide since I am only dealing with one weld seam and surface prep is so much easier.  The reason for my reversal to a wider weld at this point is the added strength advantage of a wider weld. Would not a glue joint that is two or more inches wide be stronger than one that is 1 inch or less?  Why make a narrow weld and then draw it out wider, longer and thinner when instead you can make a wider weld and forge it down.

 

  For folding I use a hot cut on the steel when it is at a good forging heat.  I cut 7/8ths of the way through, halfway down the length of the billet, and leave a “hinge” of steel to hold things in place for the fold.  Once the cut is made I go straight to a post vice and lock the billet in hinge side up (this will be the new weld surfaces).  Now, with a 4 ½” angle grinder, I grind the entire surface of the billet clean.  This eliminates any oxidation, pits or decarb that could interfere with the next weld.  Now the billet is immediately refluxed and placed back in the forge for a quick reheat of the “hinge” for easier folding.  Then the far end of the billet is folded back onto the end near the handle.  If the cut was not in the exact center there will now be an amount of the billet wasted by the resulting overlap that will not double and will create an unmanageable seam.  Now the piece is put back in the forge and brought up to welding heat again.  This time it is important to start the hammer blows at the hinged end and work back.

 

  After the second weld there will be ten layers and the whole process is repeated until the desired number of layers is reached.  For twist patterns I like to fold 4 times and go for the nice bold starbursts that 80 layers give.  For random pattern the five folds that gives 160 layers shows more details in the steel.  Ladder patterns look best with 160 layers or more.

 

To do these patterns many methods can be used but all are variations on what I call “bump”, “grind” or “twist”.  It sounds a whole lot more exciting than it is.  What it basically means is that in order to produce interesting patterns in the steel you need to manipulate and expose a certain amount of layers in different ways than others.  You can do this by forging parts down before grinding, “bumping”.  Grinding parts down before forging flat, obviously “grinding".  Or drawing the billet out into a rod and twisting it to create a spiraling view of alternating edges and flats.

 

Ladder patterns and pool and eye are done by either bump or grind.  You can grind grooves or drill dimples and then forge the piece flat to expose the cut layers or you can forge in grooves or dimples or raise ridges and bumps and grind them off.  Higher layer counts allow for tighter ladder patterns since more layers can be exposed in a given depth.

 

Twist patterns can be tight or lazy depending upon the amount of twist and the amount of drawing out after the twist.  If you plan on a lot of lengthening you should twist a little tighter to compensate for this.  Twists can also be stopped and reversed down the rod a bit for nice effects.  The deeper you grind into the center of a twist the more of a starburst pattern you will get, but upon approaching the center of the bar you will begin to form odd back to back horseshoe shapes instead, as you completely cut through the layers of the twist.  Proper preparation for a twist can be important.  I have seen many people have troubles when twisting the bars in square cross section.  Twisting involves some serious amounts of stress and shear forces on the steel and the corners of a square rod are great places for stress risers and cracks to form when you twist.  Even if this does not occur the sharp corners will create a heavy threaded rod look with high ridges and deep creases spiraling down the length.  When you forge on these later they can form stress points that can pop welds or fold over and trap junk in unwelded pockets of your steel.

 

 I tip my square rods on corners and form them into nearly rounds or octagonals.  But be warned that forging on the diagonal causes the most shear forces you can put on the welds, so take it in small bites.  But I like to think of this as a good test of the welds as well.  Believe me if the weld are going to separate I want them to do it now instead of when I am twisting.  I leave both ends square so that I have something to grab onto for the twist.  I twist good and hot and as soon as I feel heavy resistance to my twisting I stop and reheat.  If you try to twist when it has cooled just a little bit too much it will get very hard to twist and then it will suddenly get very easy -if you know what I mean.  After the twist I reflux and do a welding heat with which I just smooth out my surface while continually rotating the bar.  I like to have a very nice tightly welded cylinder before proceeding with the forging flat.  Once you have gotten rid of all the heavy spiral lines, hammer evenly to a square and then start to flatten.  

 

Other patterns can be developed by the direction of the folds while welding and also by cutting and re-stacking in various ways.  Pictures or words can even be written in the steel by properly pre-stacking an image with your starting alloys, but I will not go into that here as the mosaic stuff is just not my cup of tea. I have done some and found it to be too manmade and contrived in comparison to the natural flowing beauty of more traditional patterns.  I love being somewhat surprised by the patterns that have formed when I etch for the first time.  With mosiacs unexpected surprises are bad.