Japanese Shellmaking

[AdmiralDonSnider]

This thread tries to open a discussion about characteristics peculiar to Japanese round shells (Chrysanthemum shells particularly). One of them - at least concerning Japanse high end shells produced for the domestic market - is the fact that they are based on newsprint hemispheres.

 

For example, this skilled craftsman does use newsprint hemis for both the outer shell and the petals:

(see 2:56 min)

 

"It is said that the newspaper hemi-spheres are the most suitable for shell manufacture. (At present they are not used for export, but only for the shells displayed in Japan.) The reason is that they have a moderate tensile strength and they split more uniformly than the straw board spheres." (Shimizu FAST p.172)

Given these sayings it´d be quite worthwhile trying to reproduce these hemis and shells at home, as well as to learn the techniques and nomenclature involved. Trying to do so, one will soon begin asking for starting values concerning things like I.D., O.D. and desired wall thickness. Everyone wondering why not to simply copy the specs of strawboard hemis must note that Japanese newsprint hemis deviate from the former in both diameter and thickness. This is connected with the fact that newsprint based shells show a different setup than shells based on strawboard hemispheres, in order to optimize the break and to account for the different tensile strenght of the hemi material.

This will quickly become apparent when we take a look at Shimizus table in FAST p.173:

http://img444.imageshack.us/img444/1575/shimizuhemis.th.gif

Focussing on chrysanthemum shells, we have to look at the specifications listed under "small" on both sides. The right side lists the common strawboard hemi-specs, while the left side gives specs for newsprint based shells.

We quickly notice that the grade number on the left can´t refer to inches, because the O.D. of an unpasted 3 "inch" hemi can never be

80mm, as this would result in a shell unusable in a common 3" mortar. Instead the grade number gives a Japanese measure called "Gou", with one Gou being 3 centimeters. Thus the correct hemi for a 3" shell would be the one listed under grade number 2,5 - being 65mm in O.D. and showing 2mm newsprint wall.

 

Another table I found lists popular specs for japanese domestic shells:

 

http://img687.imageshack.us/img687/6545/pro016.th.jpg

 

 

Also here a 2,5 Gou shell is shot from a 3" mortar, and e.g. a 5 gou shell from a 6" mortar. The table gives nominal mortar diameter as well as shell O.D. (pasted), weight, lift (4FA), effect height and diameter. (by the way: it - as well as Shimizu - shows that the Japs are using rather uncommon mortars such as 90mm, 120mm etc. for some of their display shells)

 

The shell O.D. is particularly interesting for us if we combine it with the values given by Shimizu. Our example, the 2,5 Gou or 3" shell has an O.D. of 72mm when finished, but according to Shimizu, its hemi O.D. is only 65mm. This gives a pasting thickness of 7mm, or 3,5mm on each side, which would result in a (calculated) amount of about 30+ (!) layers of 40lb kraft paper.

 

I keep having my troubles trying to believe this result; on the other hand a smaller hemi necessarily has to receive more layers of pasting to reach strawboard based shells in diameter and to become shootable in a common mortar. Many people have already told me that jap newsprint shells receive much more pasting than strawboard ones (the result is that the pasted wall becomes thicker in relation to the the inner hemis, which results in a shell with thousands of possible failure points. In contrast, ordinary strawboard shells have hemis accounting for about half their wall thickness, with this half showing one preferred failure point: the equator.)

 

Can someone verify my findings, or give some more insight on japanese newsprint shells?

 

 

Edited May 30, 2010 by AdmiralDonSnider

[FREAKYDUTCHMEN]

My two cents;

 

For example, I paste my 3" shells with 3 layers, so why 30? Doesn't sound logical to me. 30 layers with only BP as break would probably blow your stars blind, and will sound more like a salute I think.

You'd better work on the practical side than reading so much theory. The chinese paper hemi's will do just fine, and is compatible with the commercial fiberglas mortar tubes.

In my beginning period I've tried to make newspaper hemis but it really sucks. They shrink and get extremely oval.

 

FAST is quite old literature, which isn't really correct these days with new insights and new methods.

 

We all know the big Japanese shells, (and someone who's seeing this topic, is referring to these movies right away) but well made Spanish shells, are just as good. These Spanish shells are made in a total different way than the Japanese shells are. And it's more than just a paper hemi that influences the chrys, like type of burst and way of pasting for example.

[smartie66]

I'm using newspaper hemis only and get very good results:

 

3'' ODcore=65mm (tennis ball), ODhemi=~67,5mm, OD after pasting ~70mm (mortar=75mm)

4'' ODcore=90mm (wooden ball), ODhemi=~93mm, OD after pasting ~96mm (mortar=102mm)

 

For 5'' and 6'' I have also two wooden cores, made exactly by a modeller especially for my purpose.

 

The 3'' hemis are made by 28 pieces of 18x5cm newspaper strips, with wallpaper paste and additionally 7-10% Dextrin and 7-10% wheat paste. (Made some glue tests, this composition works great for me)

After 2 days drying time I separate it with a cutter.

 

It needs some practice and time, but the outcome is very very good.

[AdmiralDonSnider]

FAST is quite old literature, which isn't really correct these days with new insights and new methods.

 

 

Shimizus FAST is old, but if you could point me towards a better author treating Eastern techniques, that would be great...

 

 

You'd better work on the practical side than reading so much theory. The chinese paper hemi's will do just fine (...)

 

 

Without any desire to sound rude I must say that I dislike the way you dispatched my statements. They tried to draw people´s attention to techniques not necessarily ubiquitous.

Other than Dave Blesers crude description of inner petal shells, there is no source (known to me) that deals with the use of newsprint hemis in practical pyro.

 

That´s a strange outlook on theory you seem to be committed to. I thought we were reading theory to make better practice, a behaviour that would

do a world of good to those whose masterful practice amounts to nothing more than just dumping hulls and flash into a hemi. Far be it from me to prescribe what

has to "do just fine" for them; I hope they´ll also leave it up to me.

Edited May 30, 2010 by AdmiralDonSnider

[AdmiralDonSnider]

Don,

I have made many many newsprint ball shells. I have even pasted them

in with newsprint, which is not worth the time. Here's the deal. The

objective is that the hemis in a high quality shell should be as thin

as possible and still hold their shape while assembling the shell.

That allows for more layers of multi directional pasting which both

gives a more uniform container and also minimizes the effect of the

seam. Strawboard hemis are thick so the Chinese don't have to paste as

many layers. Japanese hemis are thinner and a little smaller so they

can paste more layers on. (Quality seems to be a little more important

to the Japanese.) Competition grade shells frequently use both the

thinnest hemis they can as well as the thinnest pasting paper

available. This effectively creates thousands of small failure points

and a very, very uniform container.

 

So, yes, you do paste more layers and you also use a slower burst

charge.

 

 

Russ Birch on rec.pyro

Edited May 30, 2010 by AdmiralDonSnider

[FREAKYDUTCHMEN]

Shimizus FAST is old, but if you could point me towards a better author treating Eastern techniques, that would be great...

 

 

 

 

 

Without any desire to sound rude I must say that I dislike the way you dispatched my statements. They tried to draw people´s attention to techniques not necessarily ubiquitous.

Other than Dave Blesers crude description of inner petal shells, there is no source (known to me) that deals with the use of newsprint hemis in practical pyro.

 

That´s a strange outlook on theory you seem to be committed to. I thought we were reading theory to make better practice, a behaviour that would

do a world of good to those whose masterful practice amounts to nothing more than just dumping hulls and flash into a hemi. Far be it from me to prescribe what

has to "do just fine" for them; I hope they´ll also leave it up to me.

 

Don't get me wrong, and I agree with you to make better shells by reading more theory, but I think you might learn a lot from making mistakes in practice than from reading only.

 

Look at my uploaded 8" it is pasted with only 8 layers if I remember it correctly, so it clarifies the story you showed from rec.pyro.

Edited May 30, 2010 by FREAKYDUTCHMEN

[Peret]

You don't have to be very long in this field before you realize there's almost no published professional literature, and even less recent. There's Weingart, Shimizu, Shidlovskiy and Lancaster, all the rest being amateur or anecdotal. Even Lancaster admits in his book that the pros, himself included, withhold some of the details as trade secrets. In many ways it's remarkable that a book like Shimizu exists at all, given how competitive the Japanese pyrotechnists are between themselves. As for the rest of the literature, the great mass of collected magazine articles, it doesn't take very long to see that it's mostly the blind leading the blind, with a few pyro celebrities occasionally chipping in with a safety lecture or dropping a few hints that give little away.

 

This forum, and others like it, is the new wave of pyro literature. It's to these online archives that people will be turning in ten or twenty years ("Best of APC Forums, volume 1" anyone?) when they're looking for advice on a chlorate cell or making their own hemispheres. If Shimizu is "quite old literature, which isn't really correct these days with new insights and new methods", surely that makes the Admiral's efforts to extend the science even more worthwhile. It's not like there's a thriving industry left in western countries to keep the knowledge alive and carry it forward. If we don't do it, there's nowhere else to go.

[AdmiralDonSnider]

Thanks Peret!

 

Don't get me wrong, and I agree with you to make better shells by reading more theory, but I think you might learn a lot from making mistakes in practice than from reading only.

 

 

True; but it´s always nice to avoid a trial and error process, at least in such a labour-intensive field. That´s the reason for making this thread...

 

 

Look at my uploaded 8" it is pasted with only 8 layers if I remember it correctly, so it clarifies the story you showed from rec.pyro.

 

Not really. Bear in mind that the 3"-3layer/8"-8layer system is as off-standards as the use of newsprint hemis with heavy pasting. I´m glad this system also works, as it shows people another viable way to make shells, but such shells are not Chrysanthemums conceptually (no judging intended).

 

In my first response I simply wanted to point out that invoking things that "work just fine" is not a rewarding concept in a thread trying to discover new techniques and ways to make things, as it cuts off any further discussion. This is what I´d really love to see now!

[Jerf]

I agree with the Admiral that there seems to be some great techniques that may not be the norm today, but not because they didn't produce a quality product. Just in simple comparisons between Japanese and Chinese shells, from what I've seen the Japanese seem to be keen on producing high quality shells where the Chinese seem to be more concerned about producing high quantities of shells. I know I've dissected some Chinese "festival" shells (as I'm sure we all have at some point or another) to find that it was just a mix of inconsistently sized stars, meal coated rice hulls and some burst charge.

 

It also seems that the evolution of shell casing has been fueled more by ease of production than improving on quality. With the final installment being plastic shell casing, easy to produce and assemble, but not really designed to have a symmetrical break being that all point are equal except the equator.

 

Thinking outside the "box" of tradition (past or present), and towards a new and improved shell concept I'm sure there is something out there that would be a better fit. Maybe something like a sawdust/glue mix pressed into a thin hemi mold and dried. providing hemi's of a thin composite board like material (like they make cheap furniture with, only thinner.) Using wood glue to join the half's and it might not need too many layers. Because of the nature of the material it would have a multitude of failure points to encourage a symmetrical break.

 

I'm sure there are some very good reasons that will soon be posted as to why this would not be the ideal material to make shells from, but I imagine there has to be something "different" that would be better.

 

Please feel free to jump in and tell me how bad an idea this is now....

[Peret]

from what I've seen the Japanese seem to be keen on producing high quality shells where the Chinese seem to be more concerned about producing high quantities of shells.

The Japanese have craftsmen and run national firework competitions for their own amusement. The Chinese are all business and cutting costs and exporting for the lowest possible price. So there's a big difference between their philosophies as we see them, though I don't doubt there are Chinese craftsmen too, and probably cheap Japanese goods.

 

I know what the Admiral is getting at, though. In some of the videos, a shell that appears to be made out of newspaper gives an almost ceramic "clink" when they tap it with a stick to settle the contents. No paper shell I've ever seen is that hard. The casing also looks well over an eighth inch thick before pasting, and very rigid. Maybe they paste them up with sodium silicate. Whatever, it looks like a trade secret worth knowing.

[Pretty green flame]

I know what the Admiral is getting at, tough. In some of the videos, a shell that appears to be made out of newspaper gives an almost ceramic "clink" when they tap it with a stick to settle the contents. No paper shell I've ever seen is that hard. The casing also looks well over an eighth inch thick before pasting, and very rigid. Maybe they paste them up with sodium silicate. Whatever, it looks like a trade secret worth knowing.h

 

I would say that the ceramic "clink" comes from the large ammount of glue used. Compare newspaper and rice board, the newspaper is very thin and you need a lot of it to build a wall 3-4mm thick, with rice board you only need 3-4layers. Many layers of paper also mean lots of glue. And lots of glue will give you that clink when you tap the casing, it also gives a more brittle casing, which in theory, should give you a more symetrical break.

[AdmiralDonSnider]

Many layers of paper also mean lots of glue. And lots of glue will give you that clink when you tap the casing, it also gives a more brittle casing, which in theory, should give you a more symetrical break.

 

 

That´s right. This is one of the reasons why people recommend using thin pasting paper; another one is that overlaps are minimized in thickness.

 

My main question concerning newspaper based shells was if it is conceptually intended that they receive a lot (!) more pasting layers than strawboard based ones. This is based on the observation that you can´t just use the "standard" 3-4 layers per inch of shell diameter, because if you do a newsprint shell will be hardly shootable in a standard mortar. It´s simply too small, because of the smaller hemi O.D. you start from.

I´d leave this assumption open for discussion!

[Pretty green flame]

My main question concerning newspaper based shells was if it is conceptually intended that they receive a lot (!) more pasting layers than strawboard based ones. This is based on the observation that you can´t just use the "standard" 3-4 layers per inch of shell diameter, because if you do a newsprint shell will be hardly shootable in a standard mortar. It´s simply too small, because of the smaller hemi O.D. you start from.

I´d leave this assumption open for discussion!

 

Wait, so the shells are pasted with newspaper? I doubt that. In all videos I have seen (on japanes shell making), pasting is still done with kraft paper, sure, thickness of kraft may vary but pasting shells with newspaper would be counterproductive, newspaper has no fibre orientation, it is weak, sure, it can be used for casings but you have to remember that a shell casing provides no confinement, the pasted layers of paper do.

[AdmiralDonSnider]

Wait, so the shells are pasted with newspaper? I doubt that.

 

Nope. They´re pasted with kraft, but still require a lot more layers to gain suitable size ( at least that´s my assumption). Try pasting a newsprint core with the ordinary amount of layers (for a strawboard core): you´ll find out that the shell will not come out large enough.

 

Shimizu states that Warimono and Poka shells have to be pasted to the same final diameter since they are both used in the same mortars. This also applies to newsprint hemis compared to strawboard ones.

[Peret]

Of course, it's quite possible that the video showing a wise Japanese master sitting cross legged on the floor in an idyllic woodland setting, filling a newsprint hemi and surrounded by young female apprentices, was staged as psychological warfare on his competitors.

[AdmiralDonSnider]

Just bought myself a 65 mm steel sphere to try out hemi-making. Just tired paying 50 cents for a set and getting 30% deformed stuff...

 

If R.Birch´s theory applies, a 1mm newsprint wall should suffice with the usual amount of pasting. This will bring them up to the common 67mm OD, adding up to about 70mm with the pasting.

 

I will report the results. Could take a while though...

[50AE]

AdmiralDonSnider, why don't you try vacuum formed hemispheres?

They are very weak, maybe as newspaper hemis or even weaker and I obtain awesome breaks with them.

The molds are very easy and cheap to make. All you need is fiberglass and epoxy, but I also use polyester resin with fiberglass with great sucess. You also need a round mold, which is just a rubber ball for kids. Theyre are balls with the size of 4", 5" 7" and more. To get the ball to the desizer diameter, just pump it with air.

 

Then, some PVC tubing and a vacuum comes in part. The later is the most problematic, I hope your wife/mom has one \

When the setup is done, you will see how fast they are to produce. I recently made 10 sets of 5" hemis for 30 minutes.

I'm sorry for showing an old video, but here's a shell of mine with vacuum made hemis I really love. Notice the break.

 

http://www.youtube.com/watch?v=HR9K3T2EA8w

[ExplosiveCoek]

What kind of blue comp did you use there 50AE? Looks quite nice on vid.

[Ralph]

That´s right. This is one of the reasons why people recommend using thin pasting paper; another one is that overlaps are minimized in thickness.

 

My main question concerning newspaper based shells was if it is conceptually intended that they receive a lot (!) more pasting layers than strawboard based ones. This is based on the observation that you can´t just use the "standard" 3-4 layers per inch of shell diameter, because if you do a newsprint shell will be hardly shootable in a standard mortar. It´s simply too small, because of the smaller hemi O.D. you start from.

I´d leave this assumption open for discussion!

 

The thing to remember is that you chose the final OD of your hemis by the size of the former you use so you could (would result in a flowerpot ) fire a newspaper shell with only a single layer of print and have a rather snug fit

 

I mean absolutely no malaise when I say this but pyrotechnics is very much an art and the science is really rather optional the questions you ask make it sound as if your conducting a very careful experiment in the lab and alittle trial and error never hurt anyone (its one of the best ways to learn) burst weights are given by shimazu but they are really just a guide you fill the void with the appropriate amount of burst just as coating ratios are given but they are a guide your bp undoubtedly varies from that used in Japan which surely varies from mine and probably varies enough from the pyro that lives closest to you and it is upto you to determine a ratio which gives you the best break and star ignition and is within your milling constraints

 

live on the wild side take an idea and expand on it to work for you than try it and enjoy the results even if they are foliar our whole hobby is based on a failed attempt to discover the elixir of life or so the chinese think ask any pyro and he will tell you that it certainly gives meaning to life

[50AE]

What kind of blue comp did you use there 50AE? Looks quite nice on vid.

 

It's my own comp I use

 

67 KClO3

13 Cu oxy

8 PVC

5 Shellac

5 Dextrin

[ExplosiveCoek]

Hm that sucks, I really don't want to use Cu oxy in blue.. That stuff is way to expensive.

[Ralph]

Hm that sucks, I really don't want to use Cu oxy in blue.. That stuff is way to expensive.

 

not here

Im spending sometime working on a good blue with it because I can get it for around $3 per pound compared to $15 per pound on CuO

[AdmiralDonSnider]

The thing to remember is that you chose the final OD of your hemis by the size of the former you use so you could (would result in a flowerpot ) fire a newspaper shell with only a single layer of print and have a rather snug fit

 

I mean absolutely no malaise when I say this but pyrotechnics is very much an art and the science is really rather optional the questions you ask make it sound as if your conducting a very careful experiment in the lab and alittle trial and error never hurt anyone (its one of the best ways to learn) burst weights are given by shimazu but they are really just a guide you fill the void with the appropriate amount of burst just as coating ratios are given but they are a guide your bp undoubtedly varies from that used in Japan which surely varies from mine and probably varies enough from the pyro that lives closest to you and it is upto you to determine a ratio which gives you the best break and star ignition and is within your milling constraints

 

live on the wild side take an idea and expand on it to work for you than try it and enjoy the results even if they are foliar our whole hobby is based on a failed attempt to discover the elixir of life or so the chinese think ask any pyro and he will tell you that it certainly gives meaning to life

 

Ralph, the description of me you gave is certainly appropriate - I am too much on the theoretical side of things. This may be based on the fact that I´m stuck in a city most of the year, with no means or opportunities to "live on the wild side".

 

However, I mainly pose the questions I pose because I wonder about things. E.g. when I see Shimizus tables about hemi sizes I wonder why the newsprint ones are made smaller and thinner than their strawboard counterparts. The resulting question is not a scholastic one (the burst charge spec question was). I wonder: do this specs have a reason based in practical pyro, allowing us to produce better outcomes if we stick to them?

 

Trial and error is a viable way, but it´s good to grasp the principles and move (and modify) from there. You made some good points though!

[allrocketspsl]

It's my own comp I use

 

67 KClO3

13 Cu oxy

8 PVC

5 Shellac

5 Dextrin

 

yeah I like the blue of your comp gotta give it a try in the rocket contest