Nihonto drawing..

[xxlotus8xx]

Because I am an Engineer who spends way to much times with "a mind full of swords' instead of working I did a kind of CAD drawing of my blade. All the dimensions are accurate and I thought I would post the drawing for commentary. Mainly for measurements or descriptions I may have missed. I know there are some.

Let's see if this will actually work.

I would do these drawings for other folks if interested. All I need is a blade pic next to a ruler.

:Removed the drawing due to a error in measurement.

New one to be added shortly.

[huntershooter]

Nice work Ray.

I would include a "centerline" dimension that indicates the torizori point.

 

Regards.

[outlier48]

Ray,

 

Nicely done!

 

 

Charlie Brashear

[xxlotus8xx]

Minor edit so I reposted.

[Gabriel L]

Hi Ray, looks like fun. I'm sure the dimensions you used are accurate, but what about the kissaki shape? Looks like there's something funky going on with the ko-shinogi. Other than that, I like it; you always look more carefully at a subject when you draw it (or build it in CAD or whatever).

 

See attached for an old image of mine that focuses on the kissaki region.

[huntershooter]

That's impressive Gabriel.

[Gabriel L]

Thanks Todd. PS, Ray, if you did a number of these you could probably derive truly average shapes for given time periods / schools (of course, we're talking about a LOT, haha!). I always thought that it could be neat to see a more quantitative approach to kantei.

[xxlotus8xx]

  Gabriel Lebec said:

Thanks Todd. PS, Ray, if you did a number of these you could probably derive truly average shapes for given time periods / schools (of course, we're talking about a LOT, haha!). I always thought that it could be neat to see a more quantitative approach to kantei.

 

Yes that was kind of my thought to become familiar with the intricacies of school/period by doing these. My original plan was to play "guess the school/period" with the drawing and compare with my notes and research but I thought that might offend some people.

And that Kissaki diagram is nice! This is the stuff I need to refine...

Hmmm..note to self.

More discipline when working from home..

[xxlotus8xx]

  huntershooter said:

Nice work Ray.

I would include a "centerline" dimension that indicates the torizori point.

 

Regards.

 

Can you point me to an example? Nevermind I found one..

http://www.nihontomessageboard.com/nmb/ ... 402e5e3e23

[Carlo Giuseppe Tacchini]

  Gabriel Lebec said:

See attached for an old image of mine that focuses on the kissaki region.

 

Finally this image pops up again.

 

Brian, isn't it worth to be linked in the articles ?

[Brian]

Yep, if Gabriel is ok with it, I would love to upload it to the articles section.

I have it saved myself from long ago.

 

Brian

[Gabriel L]

Brian, I don't mind (in fact I'm flattered), but perhaps you can wait until I send you a slightly updated copy? I've noticed at least one typo and I hate the typography... :lol:

 

Enough off topic from me, I'll PM you later.

[Ken-Hawaii]

A couple of questions, Ray, from another long-time engineer. I assume you used AutoCAD, right? And did you generate the outline shape from an inserted raster image? I've digitized a couple of my blades the same way, but I'm still looking for a better method to get all the details of the hamon & boshi without doing a laborious & tedious manual trace. I have raster-to-vector conversion software, but am wondering if you've found an easier way.

[xxlotus8xx]

  Ken-Hawaii said:

A couple of questions, Ray, from another long-time engineer. I assume you used AutoCAD, right? And did you generate the outline shape from an inserted raster image? I've digitized a couple of my blades the same way, but I'm still looking for a better method to get all the details of the hamon & boshi without doing a laborious & tedious manual trace. I have raster-to-vector conversion software, but am wondering if you've found an easier way.

 

I actually used Visio (what I use for Network Engineering physical and logical drawings)

What I did was took a picture of the blade next to a ruler and created a ruler page and imported the picture and made the picture about 40% transparent. I then used geometric dimensioning stencils for the outline and occasionally QA'd againist the picture. When the outline was perfect I used the dimensions engineering stencil for the measurements and walaa. I also QA'd against the blade for sanity check.

As for the hamon I drew in a freeform line and adjusted against the blade pic.

 

With this I have quick reference for shape and other sue characteristics and of course more questions than answers are coming up!

[O koumori]

Anyone have a CMM? It would be very cool to digitize & draw that way - and much more precise!

D. Kedzie

[Ken-Hawaii]

Thanks for that info, Ray. I happen to be one of the team that created the original AutoCAD, so of course I choose that over Visio (which I also use). That's an interesting way to grab a raster image in Visio, by the way. It's a lot easier & more precise to do that in AutoCAD because all I have to do is measure the exact length of the blade, & then tell AutoCAD to scale up the inserted raster image to that length. Tracing the outline of the blade & hamon is done using what are called polylines, & I can then smooth those several different ways to get an almost perfect fit. And of course the best reason to create a vector drawing in either AutoCAD or Visio is that it can be zoomed into (magnified) as far as needed without losing any accuracy.

 

Huntershooter, using AutoCAD or Visio to find the centerline for sori determination isn't quite a slam-dunk because the curve of the mune isn't exactly smooth by mathematical standards, so there's still a subjective factor in determining the point of maximum deflection. It's easy to locate that point via parallel lines, but when I draw the centerline based on that, it turns out that the sori doesn't match in many cases. For example, a known Koto Bizen I have shows the centerline well into the torii-zori area, while examining it visually it's obvious that it's really koshi-zori. :?

 

Gabriel, that really is a very useful analysis & drawing of the kissaki. Did you use a vector program to create those cross-sections? I like your idea about developing some quantitative measurements for kantei, & am more than willing to help. My wife & I only have about 15 Nihonto in our collection, half of which are wakizashi, so I would need a bunch of really good photos of blades from known eras & schools along with precise lengths of each blade to begin developing a database. I would then create an AutoCAD drawing in which everything can be accurately measured. About the only thing I won't do again is try to digitze the mei - that's just way too much work!

 

Anyway, let me know if all of you think this is a good idea, & we can brainstorm the best way to share the data.

[xxlotus8xx]

Got your points and due to the imperfections of the curves I actually adjusted by using more than one arc and fastening the arcs and adjusting the curve until it was perfect. It was nice to have the picture as reference because I did must of the line outlines at no less than 300% of the normal picture size. So ya I more or less traced a picture as opposed to make a blade based on measurements :lol: But it worked! I guess the real test would be to try again on a blade that already has been measured and see if the Visio Dimensional Engineering stencils hold true.

So your one of the original Auto-cad developers? NICE! I do know that it is a powerful tool used across many engineering fields. Being that my field is Network Engineering I pretty much use Visio for logical and physical topologies. I have also seen this done in CAD as well as rack elevation diagrams.

[O koumori]

Ken, I think it's a great idea. I think many would be willing to contribute data.

 

You might want to look into finding a copy of Thomas T. Hoopes dissertation "A study of Japanese Sword Blades," - 1931, New York University, in which he provides exhaustive dimensional analyses of 103 nihonto blades. He not only tabulated all the primary dimensions, but also measurements and derived quantities like center of gravity, center of percussion, and period of oscillation. Unfortunately, no illustrations of the blades themselves...

 

Dan K.

[Ken-Hawaii]

I'd never heard of Hoope's dissertation, Dan, & an advanced Google search didn't find it, either. With no blade illustrations, I'm not sure how it would help the database, so if I run across it, great; otherwise I'll have to start from scratch.

 

A couple of questions to define what we're looking for:

 

(1) Should we stick with katanas only, or include wakizashis?

 

(2) As the overall objective is to assist with kantei (right??), what information should be included? I've Attached a PDF file that I use to kantei my own blades, but because I won't have the actual blade in hand, I won't have any of the 3D parameters (thickness, mune type, hi depth, etc.). Which of these should be included in the database, & how will they be used for kantei?

 

(3) Do we want a simple Excel spreadsheet with images, or a real SQL database? Which parameters will we be calling up to help with kantei?

 

Answers to those will at least get me started.

Final Sword Documentation.pdfFetching info...

[Lee Bray]

Can Autocad compensate for several hundred years of damage, polishing and minor changes by the smith?

[xxlotus8xx]

  Ken-Hawaii said:

I like your idea about developing some quantitative measurements for kantei, & am more than willing to help. My wife & I only have about 15 Nihonto in our collection, half of which are wakizashi, so I would need a bunch of really good photos of blades from known eras & schools along with precise lengths of each blade to begin developing a database. I would then create an AutoCAD drawing in which everything can be accurately measured. About the only thing I won't do again is try to digitze the mei - that's just way too much work!

 

Anyway, let me know if all of you think this is a good idea, & we can brainstorm the best way to share the data.

 

HEY THAT WAS MY IDEA!

Anyways I want to play also. It would be nice to compare the CAD to Visio and see if the measurements add up. Granted I am not even close to a mechanical engineer but I am pretty creative when it comes to solutions.

[Ken-Hawaii]

Lee, of course AutoCAD can't do anything that isn't present in the photograph of the blade. But I doubt that the overall shape of the blade will change all that much, short of suriage, machi-okuri, or something drastic like a reshaped boshi. But one of the criteria is that the owner KNOWS who made the blade, the approximate year, & the school. Otherwise, I'd be trying to do kantei "on the fly," which isn't what I want to be doing for this project.

 

Ray, I absolutely agree that this post is your idea - it's just using AutoCAD is a darn sight easier than using Visio for this type of work. I can accurately digitize the outline of a blade is about 5-10 minutes, although a complex hamon takes quite a bit longer. Linking arcs together, smoothing them, etc., is just too time-consuming! But if you want to digitize a bunch of blades & send them to me in a vector format (probably DXF), that's less work for me.

[xxlotus8xx]

I would love to see one of your drawings for comparison.The Visio I drew was a first pass at anything resembling a mechanical drawing and a real one would be a great template.And I fully get it that creating a drawing based on dimensions is the way to go. But how do you account for the variable of aging, etc...

So please share!

[Lee Bray]

  Ken-Hawaii said:

But one of the criteria is that the owner KNOWS who made the blade, the approximate year, & the school. Otherwise, I'd be trying to do kantei "on the fly," which isn't what I want to be doing for this project.

 

That makes sense.

It could make for an interesting database if you could work this from photos supplied by board members but I imagine the time and work involved would be massive, not to mention the variables thrown in by our photography skills.

 

An interesting point of note in the new Nobuo Nakahara book is zori measurement.

He states the measurement should be taken from the nakago jiri to the kissaki as opposed to the common mune machi to kissaki measurement.

I'm not entirely sure on his reasoning yet and whether it will become common practice but perhaps you could include it in your measurements.

[Ken-Hawaii]

Ray, I have no problem posting my AutoCAD drawing...but in which format? As you know, once a graphic departs from vector format, it loses all of the information that made it lossless. I frankly don't know what to do about that, so the database project comes to a screeching halt until we can figure out an answer.

 

I really hate it when reality messes up a great idea!!!

[xxlotus8xx]

Hi Ken,

You can email to me. I think I can open it with Visio?

Email address is xxlotus8xx@yahoo.com

[Ken-Hawaii]

Umm, no can do, Ray. I just tried opening AutoCAD 2010 .dwg & .dxf files with my Visio 2007, & get nothing but garbage. I also saved them in AutoCAD 2008 formats, & get the same thing. Let me contact Microsoft about that.

[xxlotus8xx]

You might have to import the file as opposed to open it.

I recall some open source app than can do open CAD in FreeBSD. I'll check it out. And actually I think I have CAD in some software dir at work. Also if you can include a picture of the blade that you documented I would like to check the accuracy of the templates in Visio.

[O koumori]

Lee wrote: It could make for an interesting database if you could work this from photos supplied by board members but I imagine the time and work involved would be massive, not to mention the variables thrown in by our photography skills.

 

One way to mitigate this variation might be to use a scanner, with a ruler next to the blade for dimensioning, regardless of the image's apparent scale.

 

Dan K.

[bone]

A simple way to create an outline of the blade is to trace it on white paper scan it at fairly high resolution, convert and trace it. A hairline trace and deleting the outer image should leave you with a nice clean vector image. The hamon could be done by the owner on plastic or acetate, backed with white paper and scanned afterward. You'll only lose a small fraction of integrity joining the two images. The kanji? You're on your own! One caveat. Tracing the sharp edge of the blade is hazardous to your sharpie.