Flash on POINT of clash

[pointoforigin]

No idea if this has been tried - I searched extensively here and elsewhere on the net - but an idea I've been toying with for a while in my head has been this:

We all like Makoto style blades. Sure, we aren't all going to go to the effort of making one, but using them as a starting point...

If you had 2 strings of LED's inside the blade, such as the v3 super blade Makoto showed us - but had only one of them hooked up to whatever board or driver you're using, you could use the other string for flash.

This is where the idea gets more complicated. And perhaps, with what we have at the moment, a bit ugly. So you do the clash string in segments of 10 LED's. Fairly normal. Just before LED 1 of each segment, you attach a clash sensor. The sensor acts as a switch to turn just that bank of LED's on. So normally it would all be off. When the corresponding part of the blade is hit, the clash sensor of that part turns on the LED segment attached.

What I'm not too sure about at the moment is these two points:

How sensitive the clash sensors that are available are. The ones for our soundboards are generally designed to "feel" a clash anywhere on the saber - so even with the wiring done this way, all the segments might light up at the same time anyway. We'd need something LESS sensitive. Which I'm thinking might have to be bigger, unfortunately. I'm not even sure you can get them to be as un-sensitive as would be needed.

How long the clash sensor closes the circuit. By nature, they're a momentary thing - so even if it did make the LED's flash, would we see it? There would have to be a way to use the clash sensor to close another switch, that stayed open for x milliseconds. I'm sure there is one - but in the space of a blade? Trying to keep out of the way of the light? That I don't know.

Issues that I can see at this point are as mentioned - components other than LED's in the blade make dark spots. Everything that wasn't an LED would have to be seriously tiny, and wired up very well.

And if you wanted to run the flash from an aux button in addition - then the clash sensors would prevent that.

It gets to being a LOT of wiring, but maybe it could be done thusly -
Current goes to the (hopefully tiny) clash sensor in the blade, then back down to the hilt. The hilt has the timed switch activated by the clash sensor in it. Current then goes back into the blade to that clash sensor's LED bank. Bank lights up and current comes back down the common anode.

With the clash activated switch being in the hilt, it could be possible to wire the aux switch to bypass the clash sensor's activation. Then we'd retain our aux button clash.

Please note - the idea is that the other string in the blade, the non-clash flash one, would be driven exactly the same as you would normally set it up - so with a Makoto board, CF, MR, driver, direct... All of my thinking here is ONLY involving the wiring of the clash flash LED string - to supplement the normal blade.

If this all worked, potentially you could have a driver board controlling each of the clash flash string segments individually as well as all of this - to make a flickering blade in segments, instead of just all of it flickering in time.

It does sound like a major headache whichever way you look at it though lol!

Any thoughts on the feasability of this concept? I'm not very electronically minded (as yet) so don't really know if I'm sprinting a marathon across lava before I can crawl.

[eastern57]

I like where your thinking is taking you. It's a rather jerky and jumpy description, but I can still see the method to your madness. While it does sound feasible, the logistics and layout will need some more thought - but the original idea isn't so fantastic that it does not warrant further consideration, quite the opposite.

Long story short, the impact on the blade would be darn-near impossible to contain on a single segment. Analog clash sensors are simple enough that you could make your own for your own purpose. It's just a spring in a can that closes a circuit (a complete circuit is considered closed), and I can see myself trying this with a string of 5mm RGBs...

... But the hard part would be finding that happy medium between "how sensitive to make your flash segment"... and "how to keep the whole blade from flashing when you hit & swing it". Analog sensors are just on/off, and can't differentiate between swings, hits, and hard hits. An accelerometer might be able to tell you how hard an impact you get, but that's programming and development beyond me...

IOW, this is outside the box thinking, and the more you learn about electronics - or better yet, the more you learn to improvise on the fly - the more ideas you'll get on what to do with those skills. Keep it up... I'm gonna stew on this one for a while...

[pointoforigin]

Post edited so "open" now says "closed" lol. Don't mind me...

I have absolutely no experience whatsoever with accellerometers, but I presume that they're quite a bit bigger than the clash sensors we have in the store, which kinda rules them out for use inside the blade. After posting though I was reading through another thread where someone had forgotten to secure their clash sensor to their saber, and it wasn't being "as effective", which for this purpose is good - so maybe it'd be possible to mount it more open air? I realise the constraints of this in the space of a blade.

Being able to tell how hard it hit would be awesome though. As well as being able to tell if pressure was still being applied - sort of "auto saber lock" if you will. I'm not sure that's realisable at the moment though

If it's that easy to make a clash sensor, I'm thinking that's gonna be part of the experiment - to make one that's juuuuuuust right - goes off on clash not swing, and only that segment goes off when it is hit.

Maybe mounting them horizontally rather than vertically? I don't know yet.

Don't worry people I AM planning on experimenting with the idea Just may take some time...

The most unfortunate thing at the moment in the design in my head is the size of the connector from hilt to blade! Say 6 segments per string, common anode. 6 cathodes for the main string. 6 cathodes for the flash string. 12 wires for the clash sensors - two each. 25 total. Holy moley that's a lot of connections... It would HAVE to be a PCB based saber to keep the jungle down, wires used only when you absolutely had to.

[eastern57]

Perhaps, it came off differently - accelerometer require programming in order to operate. I only mentioned it as to express the vast chasm that still need be hurdled... not as a suggestion.

For the impact detection, it's not a matter of mounting location or orientation - it's sensitivity. How do you "tune" an analog component? That's the toughy. Lets concentrate on one idea at a time; it's still a good idea. I think the thing to focus on at the begining would be to construct one string, to test out the clash sensor theory. If you can make a single string flash on at specific segments, then I'd say it's 'go' to move on to calibrating sensitivity... from there, maybe two strings operating simutaneously... it's a lot of work, no doubt, but good ideas are worth the work.

And I wouldn't worry about the cram-fu yet... a proof of concept (prototype) will be hard enough as it is.

[Ari-Jaq Xulden]

I figured I throw this out there.I was actually thinking about this idea as well.My thoughts were to use a bi colored led with 3 legs. This way, no dark spots. There would a common leg, main led leg, and clash leg. I don't think the bi color leds are as bright though. Just throwing the idea out there.-A

[FenderBender]

This probably wont be feasible until we can use touch sensing OLED fabric blades Something that can sense where and impact is and then only light up that particular section with a white flash. I have had this idea for a while, unfortunately OLED material is out of my R&D budget right now.

[pointoforigin]

Yeah, OLED would definately be the way to go. I'll just pull out my spare million dollars and buy some to test lol! I'm not thinking so far ahead though, more what can we do with what we have?

Another option that wouldn't leave shadows in the blade would be laser sensors around the base of the blade, imbedded in the blade holder. If, again, you could get them small enough. Idea being - laser senses at which point the beam is broken, so a clash occurs, replacing the clash sensor in the blade - talks to a chip which tells the corresponding bank of LED's in the blade to light up.

I've been searching for lasers that would be small enough to mount in the walls of a blade holder and haven't found anything yet though. Could end up being mighty expensive as well. Maybe clash sensors is the (cheap) way to go with this.

Bi-colour or RGB LED's would work as well, and would probably take only the same amount of frustration and time to solder up as two single colour strings. It gets even more complex to think about though, which is why I suggested at the start for there to simply be 2 normal strings.

[xl97]

This probably wont be feasible until we can use touch sensing blades

I didnt read the whole thread.. so Im sure I missed a bunch.. (there may be more than one goal for all I know)

but you would 100% need a way to 'track' here on the blade you have hit.

I dont know if some sort of pressure switch or something can be used..

as stated using a clash or even accelerometer is like an 'all or nothing' type of detection..

maybe something like detecting how hard a hit was you could add juice/light up 1, 2, or all 3 additional led strings?

(and you can get VERY small accelerometers)

and you can 'tune' in a clash sensor for example.. at least for 'sensitivity'.. but not for 'location'..(ie: knowing which section to light up)

[Matt Thorn]

Totally off the top of my head here...

If you used the setup described by pointoforigin in his first post, you could have each clash sensor hooked up to an Arduino board, and then program the board to send the "light up" signal to the section of the first sensor to react to the impact, and only to that section.

Of course, this is far easier said than done. I can tell you from my extremely modest experience with programming Arduino that even just programming to account for the "bounce" in ordinary momentary switches (which a clash sensor is one kind of) is quite tricky. Getting it to detect down to a hundredth of a second or less the first sensor activated is certainly beyond my skills. And then there's the possibility that the first sensor to react may not be the one closest to the point of impact. It might, for example, always be the one at the tip, where acceleration is greatest, or, for that matter, at the base, where deceleration is presumably greatest.

I'm not even going to try to dedicate any more of my already-overstretched brain capacity to this problem, but if someone actually works it out, I would be very interested. Having the spot of impact light up white would be very cool indeed.

[xl97]

what you stated could easily be done..in theory (at least the Arduino portion I could do no problems).. but also, like you covered.. this is in 'no way' a good approach for detecting WHERE the blade was hit.. and could in fact trigger false postives in other areas.

detecting a clash in not so hard using an anolog sensor.. (detecting a swing with one is more difficult)..

IMHO..its reverse problems/difficulty when using an accelerometer.

I like the effect proposed..kinda like when you press your finger to an LCD screen.. but on your blade..

I just think it woul dbe very hard to accomplish, effectively and still keep a transparent/illuminated blade (and not covered in some sort of feedback sensors)