Open Saber Sound Project

[xl97]

in some of my 'other' hobbies.. we have used piggy back boards to add other functionality..

so this would be a 'best of both worlds' type of deal..

however.. I still wouldnt have ANY clue on programming the chipset..

I do have abuddy that does this for a living though.. not sure if he is a hardware side more or a programmer.. I think the former though.

if there are any specific questions, I can ask Im sure.. his company does this as well..

(designing board layouts I think)

[vadeblade]

A good question to ask your buddy is which chip set should we be considering to control everything. I've done some minor coding before so once we know what chipset to look into i can start research and help develope a code for this project. keyword is 'help". I'm too dumb to be lead developer.

[xl97]

ok..should I word it like that? LOL because I'll really just be regurgitating the question back to him?

update: sent with a little summary of the wants for this project.

[tayjuswinn]

I had asked ERV a while back about selling 'kits' of parts (with the IC preprogrammed) that people could buy and assemble, similar to Build Your Own Clone (www.buildyourownclone.com). One of the issues is with the boards needing to be as small as they are to fit into the hilt, one isn't using 'regular' components, but micro-components. Very difficult to solder.

There are ways of getting pre-printed circuit boards with labels (al a BYOC) for where the components go.

As for modular, instead of soldering the boards together, how about some type of connector?

[Bleyst]

I'm in, great idea.

i understand we want this to be modular, i mean - sound board with separate power, changeable separate shock sensor, separate driver for LED with own power source?

Let's try to find suitable chip for sound - VS1011 seems to be a bit overkill, maybe something simpler?

I looked at speakers and found sometking like that:

RH-28-EMNL
Size (mm) 28
Imp.(ohms) 8
Resonant Freq. (fo) 450 Hz
Freq. Range 210-16k Hz
Sensitivity 81 dB
Power Input(watts) 0.8-1.0

I have no idea how strong it is compared to stock MR...

[vadeblade]

To make it modular we should have pin headers on the PCBs.

[Clockwork]

Well I am glad to see this has taken off. I have thought about this project a bit, and here are some thoughts on the questions that have been coming up a lot (sorry if I don't reference them by the name of the poster, Quoting that many people would be a pain in the ass, you all know who you were)

Who can help? You can, the interest generated in under 24 hours is great, if we keep it up many more people will become interested in what we are doing. I agree most of the people who would help us quickly could do it already. But there is alot of hassle in doing it for compensation, You have to design it, find somebody to build it, test it, market it...etc. Screw that, all you have to do in this project is lend some advice, maybe design a quick circuit here and there, however much you want to input.

Who can build one. This is a more difficult question, people are right, depending on our design, it may be quite large, which means micro parts. Personally I see a lot of wasted space in a saber hilt. If you run 2 li-ion cells (18650) in a MHS hilt lengthways, there is a **** load of space on top, over an inch wide and 6 inches long. Sure it isn't for every saber, but these boards are customizable, if you need a smaller one, you have to build with micro parts, but if you take into account you are going to use this board and leave room for it, I see no reason why we can't design something suitable. As for your average sabersmith with no soldering experience what so ever, they probably won't be interested in this kind of project, but if they are, many have friends who have soldering abilities (they could also get to know forum members who could help them in this dept)

Design issues

Modular boards: It sounds like people are thinking like I am with the seperate boards. Only thing is, if they are schematics, we give the builder the option of putting them on one board or multiple. All we need to do is label the different schematics with the inputs and outputs to the other module schematics and they can do the rest, including adding things like cable attachments or whatever they may choose.

Parts list: So this is what I have been thinking about for some time, and I think the design has several parts. At the center will have to be a microcontroller that controls all the modules. I have been looking at these, a good choice may be the PICAXE series, as it doesn't require costly development boards. Unfortunately it is all programmed in Basic which can be quite limiting. I like the looks of the Amtel AVR series, because they can be programmed in C, unfortunately the development boards on these chips can be quite costly. Some features I think we will need on our chips are this. A, Pulse Width Modulation (PWM), This will allow us to have flicker and clash effects on our LED driver. B, several Analogue-to-Digital-converter inputs, this will allow us to interpret the analogue output from an accelerometer to detect swings..etc. This feature may not be necessary if we use premade accelerometer boards, which already have the interpretation part done. C, a digital-to-analogue-converter. This is how Erv makes his sounds, and for size purposes I suggest we do the same. MP3 decoder chips start to make things larger with more chips. if we have our sound files as .raw files and output them in digital, the converter can turn them into analogue sound which we can pump directly into a small amplifier. This brings us to the other parts of the board that are not the microcontroller. A small amplifier curcuit, A way to store our sound files (not enough memory available on most microcontrollers), the accelerometer, a constant current supply that can be flicked by PWM. I am sure there was abit more, but for some reason my mind is drawing a blank.

Features: This is the most difficult question I think, Sound quality?, LED driving? flicker effects..etc. Hopefully I would like to see it like MHS. We could have multiple schematics, 1000mAh drivers, 750mAh drivers with flicker, 16 bit sound, 8 bit sound, hasbro 2 bit sound (sorry for the pun), superloud amplifiers, multiple LED drivers...etc. One would choose the features one wants and pretty much plugs those schematics into the microcontroller. Personally, I would like to see, ramping LED's, flicker effects, a resonable amp, 16bit sound, RGB colour mixing LED driver...etc. But I know that may be abit much to ask for starting out

[xwingband]

The real issue I'm having with this is soldering as mentioned. Breadboard has the potential to be nasty and it's not reasonable to have everyone solder SMD stuff. Making it too hard would lead to a few crazy first timers trying it and then a dedicated few (probably like me) who would end up being assemblers.

Since this is very much talking aloud stuff... personally I think an accelerometer sensor is a but much. For most they could live with something like an MR (I think we'd be a bit more picky with the sensitivity of our sensors though). For the sake of programming it'd be loads easier to be on/off as the trigger.

I'll also throw this out: http://www.replicantfx.com/ (kudos to Stevegel who showed this to me)

His board is essentially an open software platform with a closed hardware system. Though it's marketed to blasters if you read the tech specs the inputs could easily be ball bearing sensors like the MR and you could write the code to play a font. The price has prohibited be from seriously trying that, but I'm willing to bet he's charging a decent surplus.

[ThreeQuadFive]

Holy crap..I really like that FX module!
I'm so putting that on my xmas list

That would be sweet for a custom saber sound board/led driver!

[Jonitus]

I agree with xwing, there is no need for an accelerometer, and we need to think through how this board will actually be built. I shudder to think of people trying to solder this thing themselves...especially if the parts are SMD.

It's an open project, correct? Well, let's start off with something simple and get that working FIRST, then update/upgrade it to incorporate better features. It's going to take forever to turn out a prototype if we shoot for the moon right off the bat. Let's start small and work from there.

Simple clash and motion sensors will work fine for this application, provided they are sensitive enough, and provided the lag between the sensor going off and the resulting sound is very small.