Serpent Spear part II

 Continued from Part I here

Quite some time ago, as I now realize, I forged a spear head in search of possible historical techniques to achieve this pattern sometimes found within sword blades. The thought experiment and discussions relating to this topic and, due in part to the lack of information relating to the original artefacts, I have come to no great conclusions (yet).

In the interim of when this project began and now, I was able to at long last get my belt sander back up and running and with the excess of both time and old projects these last few months I figured it was time to go back to this one. While effective in light surface removal, the scraper I forged to work the faces of the spear was nowhere near fast enough to get through the thickness of this thing for pattern development reasons, and outside of taking several consecutive days of just scraping, I ripped through it on the 24grit belt instead.

Almost all of the seams and corners from the welding are gone with exception of where the socket meets spear tang, and those are only small aesthetic defects. If I were to tack the socket in place with a welder or even leave everything just slightly more oversized and neck down, it would be gone entirely.

Overall, I am not totally convinced that the pattern is as even and sinuous as intended, there being significant variation for how small an overall length (in relation to sword length) the spear head ultimately ended being. I'd be curious to try this again with a different approach, but first want to look into what compliment of tools were proven to exist at the time of the first emerging serpent cores. The way the forcing down of the straight core by flattening the undulated edge bars obviously had the desired effect, just not to the extent it needs. I believe that there is an upper limit to how severe a counter bend can be achieved in this method, and any additional depth to the valleys of the edge bars will simply translate to elongation rather than undulation of the core.

I have been thinking about this project a good deal lately, and have revisited some of the initial planning concerns with how to achieve the pattern. In theory, the same pieces could be forged, but the outer bars flipped so the undulations are on the core rather than the edge. I initially discarded that idea, and as I think more about it, I think I was right to do so. Based on how the serpent pattern of a core is relatively well known (not unusual, that is) and how clean the geometry and welds seem to be in originals, I do not think starting with a pre bent twist bar would work. To fit the serpent to one of the edge bar undulations would be simple enough, but getting a perfect match for the second opposing side would be far more difficult. Sort of in the examination of wolf's tooth patterning I did with Emiliano and Luke a few years back, there is a fine margin for getting everything to seat perfectly before welding.

One of the methods of reexamination I am considering is forging the individual bars closer to the final thickness prior to welding the billet together. Although this one wasn't overly thick, there was enough residual deformation from thinning that some of the tightness of wavelength may have been lost. If I had thought about it at the time, I would have measured the overall length pre and post shaping. With thinner bars, it not only reduces the lengthwise pattern deformation but I think more importantly will make the 'upsetting' of the undulating outer bars into a rectangle less taxing with regard to stretching.

Although it may be an unrealistic approach, the inclusion of wrought in smaller pieces (thinking back to the wolf's tooth pattern with several individual teeth plus a thin bar across the entire surface) may be able to eliminate the need to upset and forge back down to straight lines. Welding several smaller 'teeth' in the valleys of the serpent with another edge layer outside that may achieve similar results. I am not convinced of the practicality of that approach, however, and the asymmetry of the process is dissuading. Worded another way, the wavy edge bar would be used as a form while cold for a hot twist core forged down into it to match, then the second edge would take the 'teeth' one valley at a time in order to reduce the required precision of perfectly matching a mating frequency of peaks and valleys. Those individual parts for the second half could be easily rough forged, then finish forged into the cold, already bent serpent prior to welding for a much tighter fit. That way the trialing ends of the 'teeth' could be smoothed over and possibly lap welded over one another or if close enough to the same height, just adding another thin bar of wrought over that to match thickness to the first side.

I should note that wrought is not necessarily a choice material, having made swords with wrought in the core before and experiencing the ease at which it takes a heavy set regardless of the other steel, but other material could as easily be substituted there. For some reason in my head wrought iron makes sense simply for the subsequent etching and revealing of the process in its grain.

Anyway, I digress. Whether or not that method is either viable or plausible doesn't matter much, but in thinking out loud through as many different ways to do this with minimal tooling I begin to develop a list of details to look for in originals and how the processes either exclude certain techniques or demand them.

Until I am able to find better documentation of original patterns and preferably high resolution imagery of the steel, here are the final results of the first attempt at recreating this pattern.