View Full Version : Ivy's Soul Calibur Sword, Soul Calibur

03-24-2011, 10:17 AM
Okay, so yes, I am trying to make Ivy's "Soul Calibur" sword from SCIV. I've had some trouble finding a reference picture online, but from the image of the regular sword, Soul Calibur, it should be fine because I'm just wondering what to make my sword out of, not necessarily how to make the shape I want. (No, I am not making it come apart and go back together, I am making two swords, one completely together (sword form), and one in whip form).
http://soulcalibur.wikia.com/wiki/Soul_Calibur (Soul Calibur ref pic)
I've had a few ideas and I've been brainstorming with a few friends but, since the sword looks like it's made of crystals I wanted to make it translucent, oh and I want to make it light up, insanity I know. So at first we thought, resin with a clear rod in the middle with a strip of clear LEDs around it. Then thought that a fully resin sword would be really heavy and thought about making some sort of foam center that would be blue and make the resin clear (with some glitter to perhaps look like frost) and wrap the foam center in the clear LED strip. Then my question became, can you immerse a full LED strip in resin? and then would the resin eat the foam?
Then I have some more technical questions about if you can fade colors/glitter intensity from the center of the sword to the edges? Could this look be achieved with a two part mold? Does anyone have suggestions about what type of mold I should use? A two part mold, or a 3D mold with a core?
Any suggestions/ideas would help.

03-24-2011, 10:57 AM
Wanting your sword to light up is not nearly as insane as wanting an extendable/retractable Ivy sword, so you're good.

Yes, you can immerse LEDs in resin. However, you usually want to avoid it. It increases the chances for airbubbles, it doesn't look very realistic most of the time, and if there are any wiring failures, you have to remake the entire prop.

Polyester resin (for example, Castin' Craft) will eat polystyrene foams. Polyurethane resins won't.

Getting a diffusion of colors could be achieved by rotocasting or brushing in multiple batches of resin, but I suspect there are better and easier ways to achieve what you want. One good example would be to use automotive candy coats built up in multiple layers. On the cheaper side, you can use translucent acrylic paints.

Ideally, yes, you want a 2 part mold. If you can afford it, the best type of mold would be a 2-part box-type platinum silicone mold. You can potentially save money by using a matrix mold, or switching to polyurethane rubber. You probably do not want a 3-part mold, since with most sword designs, it will be difficult or impossible to pry out the core. If you want a hollow part, you may need to cast the sword in multiple pieces that would be glued together.

Airbubbles are a major issue when casting translucent/transparent resin. Professionally, this is handled by casting the piece inside a pressurized container. This may be unfeasible for casting a large piece, such as a sword blade.

My major suggestion for lighting effects would be to look into the custom lightsaber building community. Granted, they use prefabricated tubing, and you'll instead need to cast your own sword, but they know how to integrate LEDs and special films so that it looks good.

03-24-2011, 11:27 AM
Thank you! That helps a lot.
This sounds like a perfect project-- I love trial and error :D

03-25-2011, 10:17 AM
A bit off topic, but I'm actually going to be making Ivy's Valentine for a friend's Ivy costume soon and had actually thought of a relatively easy way to allow the sword to extend/retract.
It would involve making the sword, cutting it into each blade segment and hilt, drilling a hole through the center of each piece and the hilt except for the tip. The tip of the blade would have a "rope" anchored in it which would then be fed through each segment of the blade and through the end of the hilt. When there was no tension on the "rope" the sword would be in its neutral extended state, when pulled taught it would pull all the pieces together to return the blade to the sword state.

Of course that is a very simplified explanation, but it gets the idea across and might get someone thinking if they had ever wanted to make such a prop.

03-25-2011, 12:47 PM
(Preface: Dangit, now you went and got me started. I didn't mean to type this much on a thread hijack, but oh well.)You've got the general solution, but as you say it's simplified. There are two major issues:

One is setting up stops so that each segment is an equal distance away from it's neighbor segment. Without stops, all the segments would just hang together on the end of the cord. Probably the easiest way to achieve this would be to use 2 sets of rope, one connected to the tip, and the other connected to each segment the desired length apart. It works the same way venetian blinds work. The blade segments would need to be sufficiently hollowed to store the gathered rope when in the retracted state. The problem is, this would cause the segments to gather in the opposite order seen in the animations. First the tip will join with the second to last segment, then the 3rd to last, until finally the full blade joins to the hilt. this causes gravity to pull the assembled blade down such that it would require a herculean force to pull the blade up to the proper angle. The result is that you can only retract the blade by pointing the hilt straight down.

One alternative to allow the blade to gather in the proper order would be to use a separate cord for each segment, and then gather them together using a clutching mechanism in the hilt, operating similar to a fishing reel. On retraction, when the first segment meets the hilt, that spool would just idly spin as the remaining spools continue to gather. This of course would be a complicated mechanism that would be difficult to hide, and you'd have all these extra cords looking funny, and potentially tangling. Perhaps they could be concealed within a woven sleeve, but that would require testing.

The second problem is that the blade pieces tend to twist while extended and during retraction. This would not be a problem with a cylindrical "blade" since orientation wouldn't matter. But since the blade is flat, a twist of even a single degree could cause many designs (such as the simple 'V' shape visible in the game) not to lock together on retraction. You need to specially design each segment such that the retraction force causes each segment to reorient itself. Personally, I see this as a conical design that acts as a quarter turn (maybe less) screw. When I first considered this, years ago, the conclusion I reached was that I wouldn't want to try to design it without access to decent 3D CAD software.

I also considered using a squared extension-spring as the main cord. You can see this being used on the Dremel flex-shaft. This either requires either a much more complicated retraction mechanism, or using two holes, one square pipe mounted in the dead center of each segment, and one auxiliary hole for all the other cords. Plus, I have no idea where one gets a six foot long square spring.

So yeah, in conclusion, I've put way too much thought into this.

03-26-2011, 12:42 PM
Yup, the equi-distant stops for the individual segments is really the only reason that I haven't started on the project yet.

Originally I had thought of incorporating some type of device similar to a "Chinese finger trap" inside of each piece so that as the segments slid down the rope in the extended state they would eventually stop on their own, but that seems difficult to keep aligned as the sword is retracted and extended several times.

The dual rope theory sounds promising, but then you'd have to cover that in some type of wire cover lest you see two separate ropes tethering the segments together which could lead to problems if any tangling occurred.

I wasn't especially concerned whether or not the sword could be retracted quickly and smoothly, and had planned to deal with that by installing small pegs and slots into the segments that would connect together. It might take a bit of aligning by hand, but it would also make it more secure when in sword state.
Your ideas on retraction sound solid, but as you said, very intricate and either difficult to obtain parts for or difficult to conceal.

Over thinking things like this are part of what makes the hobby fun. :)

03-28-2011, 12:39 AM
Could it be constructed by using a point and hilt along with a series of segments each consisting internally of two drums wound in opposite directions and a toothed gear mounted on a common axle? A wire connecting the tip to the last segment would be of some length, lets say 1 foot, and would end on one drum with the other drum mounted within that segment having a wire of 2 feet in length, and the next segment would be connected to the 2' wire and a 3' wire and so on until the winding mechanism in the hilt would have a foot of wire per section (sum of tip and segments). Each axle would be locked in place with the sword extended, and as it retracted a mechanism would free the axle within each segment after it locked into place. As it descended, the gear could drive a toothed pin which would release the locking mechanism and fix the axle once the toothed gear had rotated as many times as would be needed for the drum within that segment to fully extend its lower wire.

Fully extended, here is what happens (not sure if this is how it works in game, though): The hilt winding mechanism begins to retract, raising the segment closest to the hilt. At this point the #1 segment's gearwork is frozen. Upon clicking into place, the drums inside segment #1 are free, and the hilt continues to wind in wire, now pulling wire from one drum inside segment #1, and causing the other drum in segment #1 to rotate and take in wire. The gear in segment #1 also begins to retract the geared shaft, moving it away from the unlocking position. Segment #2 raises and locks into place, freeing its gears to raise segment #3 and begin retracting the unlocking pin. And so on, until finally the point section is pulled up and held in place by tension. Whipsword is now sword, with all drums free and all sections locked together.

To descend, the winding mechanism is reversed and the tip drops. As the tip drops, the toothed gears all begin to drive their shafts/pins into the unlocking position, though each pin would be calibrated differently. As soon as the point reaches its lowest position, the pin in the segment by the point frees itself, also causing its axle to lock. At this point, the pin in the next segment up is 1/2 its travel distance toward unlocking, and the segment above that 1/3 and so on. This process continues until the segment nearest the hilt unlocks and is lowered into position and the winding mechanism deactivated.

Permanent magnets might be useful for helping to align the pieces, too. Even if they aren't necessary, seeing segments start to twist into position before their geometry interacts could be well worth the additional effort.