Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - inio

Pages: 1 2 3 ... 9
General Hardware Talk / Re: RF BUTTONS 1 AND 2
« on: August 10, 2008, 07:56:58 PM »
Don't hook up power and ground backwards ;)

General Hardware Talk / Re: RF BUTTONS 1 AND 2
« on: July 31, 2008, 08:38:25 AM »
I would have gone for momentary instead of latch.  With latch you'll need to press one button, then another one (to un-press the first) to sync.

For frequency, neither is actually legal to use in australia as far as I could tell.  433MHz is an unlicensed band in some parts of the world, but not in Australia (ITU region 3).

General Hardware Talk / Re: Vibration motor details
« on: July 30, 2008, 08:26:08 PM »
Um ... why?

Frequency (RPM) will vary with the torque.  Torque varies with how firmly you restrict the Wii Remote.  There's also probabably significant variation between Wii Remotes.

As for a data sheet, it's almost certainly a custom part.

General Hardware Talk / Re: Nyko wireless nunchuck - Bluetooth/WiFi ?
« on: July 30, 2008, 07:52:39 PM »
Most likely it's a proprietary 900MHz (or other ISM band) protocol.  Not much you can do.

General Hardware Talk / Re: RF BUTTONS 1 AND 2
« on: July 30, 2008, 07:51:17 PM »

VT has me confused too.  I couldn't find any docs on the part - did it ship with a datasheet or anything?

General Hardware Talk / Re: RF BUTTONS 1 AND 2
« on: July 30, 2008, 08:56:55 AM »
You'd be better off using the sync button instead of the 1 and 2 buttons, since that way you only need to close one connection.

If the outputs are active-high, you want a circuit like in the attachment

The 1k resistance is a minimum.  Maximum for that is probalby about 50k.
Transistor can be any NPN transistor, 2N3904 is the most common.

If the outputs are active-low, you can go even simpler.  Just a diode between the sync sense line (anode) and RF output (cathode).

General Hardware Talk / Re: buttons 1 and 2 'remote'?
« on: July 18, 2008, 10:28:51 AM »
I noticed that the logic ground and negative battery terminal aren't directly connected.  However, their voltage stays about the same, so you can use bat- as a node to short the sync sense line to.

General Hardware Talk / Re: AC adapter for Wii remote
« on: July 18, 2008, 10:26:30 AM »
If you go to a real electronics supplier (not Radio Shack) you can get just about any regulator imaginable.  the 2.85 I'm using is:

« on: July 14, 2008, 11:27:06 PM »
My guess:

1. accelerometer (only needs 2 axes but probably just the same 3 axis part that's in the nunchuck)
2. 1-axis gyro on the plane of the base of the remote
3. 3-axis compass

Price:$20 -$30 USD (depending on if it has all 3 sensors)

This would basically turn the Wii Remote into a Motus Darwin but with an IR camera, speaker, better ergonomics, and probably lower price point (I've heard the Darwin MSRP is planned to be around $100).

General Hardware Talk / Re: buttons 1 and 2 'remote'?
« on: July 13, 2008, 11:11:14 AM »
Exactly.  Just make sure not to supply more than 3V on the power line.  The Wii Remote is not designed for source voltages above that, and can be significantly damaged by voltages as low as 3.5V.  Lower is just fine, probably down to around 2.2V.  I add a 2.85V regulator inside each Wii Remotes I've wired up like this to make sure I don't over-volt it.

General Hardware Talk / Re: Wiimote Power through Projector?
« on: July 11, 2008, 11:21:57 AM »
USB A connectors are for hosts and B are for devices.  Hosts supply power to devices, so it makes sense that the +5 pin wouldn't supply power.  What other connectors does the projector have?

General Hardware Talk / Re: buttons 1 and 2 'remote'?
« on: July 11, 2008, 11:12:49 AM »
By drilling a hole into the front face of the Wii Remote at the right spot, you can solder a wire to a specific pin of the red button.  Connecting this pin to ground (negative battery terminal) causes the remote to sync.  Which pin you have to use seems to change between hardware revisions, but in the latest ones I've worked with it was the it's the bottom-right pin, located a little up from the space between the player 1 and player 2 LEDs.

You could use speaker wire to create the extension.  Solder one wire to the pin on the switch and the other to the negative battery clip (soldering to the clips themselves can be really tricky and requires a HOT iron).  And the other end of the wire, solder a normally open button between the wires.

General Discussion / Re: 3D scanning accuracy ?
« on: June 13, 2008, 10:09:54 AM »
So you want 5mm depth accuracy.  How much spatial resolution do you need? You say it's a fairly smooth surface, so I'm guessing 2-3cm would be acceptable.  This is beginning to sound doable.

What you'd probably want to do is use three wii remotes.  One is positioned a fair distance away, looking at the entire wall, giving mostly x/y coordinates.  The other two are placed at fairly steep angles to the wall, giving mostly z coordinates.

For the LED, you'll need something bright to be seen by the distant wii remote.  You don't want to shine it at the wall (producing a large blob) but instead just point to back towards the wii remotes.  I recommend the Osram SFH4231, but many others could work too.

Once you've taken the data, you'll need to solve for the camera parameters.  If you're at a university, I'm sure someone in the computer science department can help with this.  If not, I'm sorry but I can't.  When you do go looking for help, the problem can best be described as: "given three views of a point cloud from unknown cameras, recover the 3D geometry of the cloud".  This is a solvable problem, and Hartley & Zisserman "Multiple View Geometry" says exactly to solve it.  You'll need MATLAB to easily implement it though (it requires an SVD).

I'm working on an application of Wii Remotes that does exactly what you're talking about:

If I place two wiimotes on in front of the user and one paralel to the user I could get the X, Y, and Z, right and then work with the points...

You'd only need two, since each tracks points in two dimensions.  Calibrating them could be a bit tricky though.

On the PDF you linked to, it says that you'll be using Google SketchUp, how this tool would work with real-time modeling (changing skeleton) and etc, is it easier, cheaper or what?

Cheaper, and I'm very familiar with it's scripting system.

Pages: 1 2 3 ... 9