Soft-switched circuit bent Funky Furby

This Furby bend has been prompted by a workshop I’m running with SoundNetwork in Liverpool next week. The demo Furby I’ve previously taken along to these things seems to have come off the worse for wear, so I wanted a bent Furby that’s still within its protective layers. I also wanted to try something a little bit different to the ubiquitous push and toggle switches, so I experimented with conductive thread to see how that changed the experience of affecting through touch.

Since my posts documenting circuit bending a Funky Furby consistently get a lot of traffic and people seem to find my ‘research‘ useful, I thought I’d better post this video:

Funky Furby bend with conductive thread from nikkipugh on Vimeo.

I’ve simply soldered two wires to the bend points, fed them out of a convenient hole in the back of the plastic shell, made a loop in the ends of them and sewn this loop to the inside of the fur with conductive thread.

After replacing the fur (I didn’t need to detach it fully, just open it up around the base and peel it up over the back so I could open the shell) I then sewed more thread to the points where the wires were attached.

Many thanks to Kit Larks and Lynne Bruning for supplying the raw materials.

Circuit-bent furby makes an appearance

Small-child-and-circuit-bent-furby win. from nikkipugh on Vimeo.

See, it’s all about the performance for one.

Circuit-bending a Funky Furby #7: performance for one

Since I started this circuit-bent Furby project, people have been nudging me to do some sort of a performance with it.

I struggle with this idea, I think for two main reasons.

The first is that I’ve never been to a performance that even remotely indicates to me what a circuit-bent Furby gig might be like. What might a circuit-bent Furby gig be like?

The second is that, as Danny started to get close to with his questioning of how I relate to the Furby as I’m messing with its circuits, the interesting thing about this kind of object for me is how people interact with it.

Fiona asked me to bring the Furby along to her recent birthday party. Brilliant! A chance to see how other people play with it!

It was great to just sit back and watch as different people responded to the thing in different ways. Also good was how people responded when the Furby failed to respond …and the different ways in which frustration, anger and dominance were expressed! This is the stage where I start to find out what it is that I have made. Up until then, it’s just been a learning project as I try and improve my electronics skills. It starts to come alive only once I put it into the hands of other people.

So, the idea of a performance and therefore of an audience is quite an alien one for me. I’m more interested in participants; in audiences of one.

Here’s Pete figuring out Furby-Thingamagoop interaction:

Performance for Furby and Thingamagoop from nikkipugh on Vimeo.

Would you like a circuit-bent Furby?

I’m unlikely to part with this one (still needs a name, by the way), but I’d like to know if people would be interested in commissioning their very own circuit-bent Furby?

It looks like Funky Furbies (like the one above, video here) are going on ebay for about £20, so let’s say that after they’ve been circuit-bent and had extra components put in etc etc you’re looking of a starting price of around £50.

furbile from nikkipugh on Vimeo.

It might be possible to get a smaller one, like the above, on ebay for less, maybe even £5, but that depends on how the auctions play out…

Anyway, if you’re interested in commissioning your own unique little ear-wiggling bleepy thing please contact me and we can arrange a design brief and a budget to work to.

Circuit-bending a Funky Furby #6: new ears

Furby Ears from nikkipugh on Vimeo.

I replaced the LEDs from the last round of modifications with 4 white LEDs (in parallel) and a 1k ohm resistor per ear. Much better!

Circuit-bending a Funky Furby #5: switches

Working directly on the circuitboards to find glitches was very difficult, especially since most actions involve, well, action and the Furby kept moving around.

Removing the feet again allows the Furby to stand fairly steadily on the base of the battery compartment without the leg cams reaching the surface it’s on to make the body move. However, this also means you have to down tools and pick the Furby up each time you need to turn him onandoffagain. With all the exposed mechanics and electronics, finding a safe place to pick him up by is a challenge in itself.

On-off switch

The on-off switch doesn’t appear to be directly connected to the battery compartment as in simpler electronic toys I’ve worked with, instead 3 wires come out from the back of the switch and join onto the circuitboard.

I just snipped the yellow, red and green wires about an inch away from the switch (enough to leave me room to solder back onto the wires going into the switch, if needed) and then extended the wires connected to the circuitboard and soldered a SPDT toggle switch onto the other ends.

Tickle me

I also wanted to add remote switches to the belly, mouth and back of the Furby. All these plug into the circuitboard and, thanks to the coloured wires, are reasonably easy to identify.

It’s also pretty straightforward to solder a couple of extra wires onto the leads of the plugs where they come out onto the front of the circuitboard. I’m just using some momentary push-to-make switches to trigger the Furby. It’s nice to have a few actions that are not in response to voice commands.

I also tried to use a push-to-make switch to control the sound glitch, but found I was killing a lot of switches in the process. At first I thought I was over-heating them at the soldering stage and messing up some of the internal contacts, but I eventually noticed (with the help of my multimeter) that it was after a few pushes that the switch stopped working. I thought that the current might be too high for the switches and so added in a 10ohm resistor, but was still getting sketchy results.

Increasing the resistance had the effect of speeding up the sound (not particularly desirable) so rather than increase the resistance I eventually abandoned the push-to-make switches in favour of a chunkier toggle switch (also with 10 ohm resistor).

At one stage I tried a large latching push-to-make switch, but I struggled with this because it was never clear whether it was on or off. If you’re going to go down this route, I suggest maybe using one with an indicator LED…

All-in-all there’s now a significant number of wires coming off the Furby and the solder points are quite delicate – particularly those ones that connect directly to the circuitboard. In order to add some protection against accidental pulls, I tied a piece of nylon cord to some of the plastic structure of the Furby and added a few polymorph figure-of-eights. I then looped each wire through the figure-of-eights before then taking it to the switch box. The nylon is slightly shorter than the length of the wires and tied off inside the switch box so it should hopefully take the brunt of any excess force.

Flashy lights

I’ve also attached a few LEDs to various points on the circuitboard (again, the solder points on the other side to where plugs attach). These were initially soldered straight onto the board, but have now been added onto wires in series with a 1k ohm resistor.

As I write this, the LEDs have made it to the ends of his ears and mohican spikey thing, but I’m not quite satisfied with the result. More tinkering required – perhaps either to add more LEDs down the length of the ears, or to add some sort of translucent cover.

I need a name from nikkipugh on Vimeo.

Circuit-bending a Funky Furby #4: audio jack

As I mentioned in the last post, I found some sound bends on the Funky Furby I’m circuit-bending.

The next step was to add an audio jack so I could play the Furby’s sound through a mini amp.

This is only the second project I’ve ever added an audio output to, so I’m still kind of fumbling my way through.

To cut a long story short, through a combination of peering at Alfonso‘s innards and section 7 of this circuit bending faq, I arrived at the following:

I’m using a 3.5mm enclosed chassis mount jack socket from RS, (part number 106-874), which has quite an open structure, so it’s pretty easy to see which bit is which.

Basically the two bottom connectors are the signals, and the one coming off the side is the ground.

Here’s a diagram of the wiring and how it connects up to the circuitboard and the speaker:

It’s ever so slightly different from what’s described on the oscillateur faq page, since I’m connecting c to b, rather than c to the ground of the speaker. I did try that, but got no sound.

The other thing I’m finding is that I still get sound from the Furby’s speaker when I plug the amp in. This doesn’t happen when the resistor is omitted, but the sound quality isn’t so nice, so I think I’ll put up with that.

I’m still not really convinced I’m doing this right, but I’ve not yet found any other combination of wiring that works. Maybe I’ve got my positive and negative terminals on the speaker mixed up?

If anyone can point me in the direction of a solution then I’d be most grateful, because this is something that’s going to keep coming up and I’d quite like to learn how to do it properly!

Update!
I had wrongly identified which were the positive and negative terminals on the speaker. After changing that starting point, the audio jack now works as expected with the following wiring:

Circuit-bending a Funky Furby #3: sound chip

Today I found a stable bend point on the Funky Furby I’m trying to circuit-bend!

On a whim I started exploring the small chip behind his left ear and I found this:

[audio:http://www.npugh.co.uk/media/funkyfurbybend1.mp3]

Connecting a few different pins (with a wire, to short circuit them) gives the same effect. I’ve marked them in the photo below…

These bends don’t affect movement at all. Though I’m happy to have found these ones – and think they’ll sound great when amplified – it’d be nice to have some movement loops too…

Update: Pete Ashton has set the above sound file to music

[audio:http://posterous.com/getfile/files.posterous.com/peteashton/gFvUu6BzNhLcIgc6yZ0oUaGaZYSsCNXRybyzLu6k9kzDCak5Ej94uMZyQfrZ/Hey_Furby.mp3]

Circuit-bending a Funky Furby #2: looking for bends

I’ve not really got organised properly with this yet, but did stumble across a short-circuit yesterday that locked some of the motor stuff into a loop.

Motor glitch from nikkipugh on Vimeo.

Another furby bend point. Possibly. from nikkipugh on Vimeo.

Talking to the furby or pressing the belly or back switches doesn’t break the loop, so at the moment it looks like the only way to interrrupt the cycle is to turnitonandoffagain.

Circuit-bending a Funky Furby #1: de-skinning

Whilst there’s quite a few resources available online to help out with any planned circuit-bending of a regular furby, I’ve been unsuccessful in my searches for information relating to circuit-bending their larger relative, the Funky Furby.

This is probably one of the reasons why the purple and green critter I picked up at a carboot sale over the summer has been sat on my hack-shelf for so long, without even so much as having had fresh batteries put in it.

It is also the reason why I’m going to document as much of the process here as I can, starting with today’s instalment: de-skinning.

The first task is to remove the feet. This is likely to be the biggest hurdle due to the need for a triangular-shaped screwdriver. If you’ve got that (and a normal, small-sized cross-head screwdriver), then the rest is really quite straightforward.

After removing the two screws in the base of each foot, you can lift off the fleshy toe bit to reveal a second, black mount. This is where the feet connect to the main body, so you’ll need to remove these too. Once you have removed these (the screws are quite long) put all the foot parts and screws to one side – you’ll need these again later.

Removing the black mounts reveals a panel of plain material (rather than fur) with a small hole in it. If you poke around through this hole, back in the direction towards the battery compartment, you should be able to hook a finger (or maybe a screwdriver) under the black plastic strip that is sewn onto the white cloth along the corner edge where the battery compartment sticks up. Pull hard to undo the clip and free the fabric.

Some of the tabs on my furby came out ok, but one or two snapped. Your mileage may vary…

After that, it’s a case of gradually peeling the fur up over the top of the furby.

It’s easiest to work up over the back and bring the pelt over the front of the face. There are a few more of those tab things, and a few other places where you need to unhook bits and pieces, but you don’t need to snip any of the stitching or anything around the face plate – the aim is just to clear the fur layer from the back half of the shell and any of the articulated bits coming out from the inside of the shell.

Once you have done this, you can unscrew the two halves of the shell and the fur, still attached by the face plate to the front half, can just stay how it is. The microphone is attached to the front half of the shell – unscrew this so you can remove the furby innards completely.

From here you can reattach the feet to give the little critter some stability and then you can do whatever you’re going to do with the electronics…

As I said, I haven’t found any references for bend points, so there’s a high chance the furby won’t make it through the next stage where I noodle around on the circuitboard looking for glitches. I made sure to get some video of him doing his stuff before I went any further:

Funky Furby from nikkipugh on Vimeo.

…after which I did actually Read The Manual and found out that you have to play music at it for it to dance!

I need to rig up some sort of jig or stand to hold the furby off the ground for the next stage – triggering sounds also means triggering movement and that makes it really difficult to work on the circuitboard to try and find the bends.

Hopefully the furby will survive long enough to have something to put in the next of what is planned to be a series of posts documenting the circuitbending process…