Hacking Audio Hardware

class thirteen - a series of unfortunate events
dec 11th

The two toys are both more promising than the reissued speak and spell. At least I can see some chips exposed on the board to start with. Therefore, I was able to hack something out of both of them. The black toy has two ic chips and the colorful one has one exposed to my disposal.

How it sounded originally

The back of the board

To start with, I found how the pushdown button works on this device. This means I can add a light signal to tell users if the device is on. I also found this connection that jumps the scale to one scale higher. Pin1 to the lower red dot. This will also change the demos too. Once it is jumped it won’t be jumping back until you connect pin1 with the higher red dot.

Unfortunately, the connection that restores it sparks when I try to jump the connection. I believe this might be shorting something internally since after a few tries, the restoration stopped working - the scale remains one note higher - the chip overheat by default. I tried restarting, wiping the board clean, and even looked up the parts' names online. It turns out the part might be specific to the model of the toy since there is no documentation I could find.

I'm referring to the bottom left pin as pin 1. When jumping pin 1 to the third pin from bottom to top, all the sounds will be one scale higher. The one pin above that jumping with pin 1 sparks but restore the sound to its original scale. The latter connection stoped working in a few trails.

Unfortunately, one of the legs broke off the chip when I tried to add my hacks to the enclosure. The video on the right is me trying to fix it by touching the itty little bit of metal left for on the chip to the next connection - a capacitor. It did not work. I also tried dropping some solder on it, but it didn't work either since the chip is plastic and the board is also non-stick to the solder.

All I have left are the documentations for the three hacks. I was planning on putting a toggle switch for the distortion, arcade buttons for the new sounds. I already added the toggle switch for the distortion, and that's what brokes the chip. The lesson is that I will think about the enclosure first before soldering on the original board.

I was able to find four hacks on this toy.

One is a distortion for existing sound. The first part of the video on the left shows that. It was made possible by jumping the output pin to the first pin on the same row.

I also found two more new sounds. One is jumping the output pin to the pin above it - the latter part of the video on the left shows that. The other one is by jumping some of the input pins together that were shown on the video on the right.

The last hack I found is to make a constant connection to play a sound constantly that’s not supposed to be played constantly. The result is shown in the video on the bottom left. The original function for this pin was to adjust the volume, and it is only supposed to be triggered once every time the button is pressed. By jumping the connections directly from VIN to volume pin, I was able to let the volume change effect play constantly.

I think I will get another one of the colorful keyboard toys for hacks. As of now, I have a better understanding of the parts themselves. The image on the left is how the inputs are gathered in this one board and sent to the mainboard via the DAOP. As you may also see on the four buttons images, I was able to find specifically where they were sent to and how they behave. I had a lot of fun hacking this toy since the results were really rewarding. Therefore, it is also extra sad when it broke towards the end.

Something I also hoped I could figure out in the future iterations is that if I could add in potentiometers instead of merely jumping the connections. All of the hacks I found on the toy are on/off connections. I tried using potentiometers, but there seems to be a threshold for the resistor values controlling the hacks on/off status instead of analog input. Also, it would really help me in the future if I pay more attention to how to solder on exiting circuits. Maybe finding the correct cable to do that (?). Part of the reason how eventually the leg broke off is that my cable is rather thick considering the small vulnerable parts that come with the original toy.

class twelve
nov 11th

Dear world,

Hope this email finds you well. I am majorly screwed.

I know I should not have trusted Amazon and modern toys.

Best,

Shuang

I also should have read some comments on Amazon sooner. Someone asked if this model is bendable on Amazon, and I simply overlooked it. I kind of knew it would not be that easy for me to get something off Amazon to bend.

I do appreciate Amazon's return policy. I was able to drop it off at Wholefood and got a refund immediately. I was also lucky this simple toy does not have a warranty that void upon opening the lid.

Apparently, the company that makes speak and spell re-issued this model after stoping the line for a really long time. The newer model "cost-effectively" "upgrades" its parts into this small board with epoxy-sealed chips.

Most of the videos I could find on Youtube explaining this newer model are complaining about this upgrade because a) the new model is barely hackable, b) it is cheap in comparison to the old one. Click to see the comparison.

Alternative plan:

I am getting more toys from Amazon. I know how ironic this is, but most ebay items (that should probably work with circuit bending) will take months to be shipped. I am not sure if these two are going to work, so I am also going into shops all over the city to hunt down electrical toys that make a sound.

class 8,9,10,11 - final planning
nov 21st

Circuit Bending is such a cool concept. After burning my backpack in class trying to power it with a 9v battery. I decided I want to bend something. I got this speak and spell from Amazon, hoping I could bend it into my own sound deck.

I am quite excited about both the circuit bending and the fabrication process. While the circuit bending part is about what I can do with the given analog circuit, the fabrication process requires some spatial thinking since the original packaging only has limited space for added parts.

I also found plenty of bent online. This guy specifically made a thorough and promising tutorial that I think I should be able to follow.

I have also asked Yony for collaboration. We are thinking that we could bend two different toys together. We are thinking some kind of plug-in system would be perfect as they do not have to be codependent on each other.

Yony is also looking into recording toys for more interesting input.

class seven - Jam Machine
nov 9th

I know this is what I want for the protoboard. So all is left is to put things together... I feel nervous and excited.

The red button and the pot control the LED bulb. The LED bulb is connected through the investors on the other side of the chip that was idle before the LED.

Dianna was so nice that she gave me one of her amazing protoboards that basically look like a breadboard.

It is very different from how I imagined soldering on the protoboard would be. The biggest thing is that it is so much harder to test things on the protoboard than it is on the breadboard.

I was lucky enough that I did not cross into large problems in the process of making this. One thing though, I think I burnt one of the photocells trying to heat-shrink the exposed wires. I was able to identify the problem and had to de-solder and change to a new one.

The result is a little bit messy if you look into the glass jar, but I am very proud because it works!

As designed in the previous post. The exposed photocell will respond to the movement outside will be influenced by the bulb. By adding the switch to the bulb, I will be able to make different types of grooves.

class seven

class six

class six
nov 2nd

Plan for the project:

By accident, the photocell from ITP floor has an interesting filtering effect on the circuit. With the same setup, the new photocell creates a staired scale. However, when adding more of the investors with photocells in our oscillator, the effect becomes not as visible.

I like the stairs. The general idea is that I want to incorporate at least this part into the final product.

two of the invertors
in 4093 ic chip

10k pot

As of enclosure:

I like the idea of an existing commercial container that is secretly an oscillator. Though the Altoids tin seems to be a well-executed idea. I am more prone to the idea of a visible contained circuit. Therefore, I am thinking about putting the final circuit into a jam jar like the photo shown on the right. I have also obtained a small amplified speaker that enabled the possibility to be put inside of the jar, but that takes more experiment for me to see how is that affects the sound.

Even more intriguingly, the range of the sound can also be extended with other artificial light sources that come with the oscillator. Because we are only using two of the investors in the ic chip, I intend to add a separate bulb into the circuit like how we did earlier. Some experiments I need to do in the following week:

How should I allow the lightbulb and the photocell to work together while not affecting a hand input (basically, without the heat shrink)?
Is it possible for me to put the speaker inside?
The tin jar cover is conductive, how to incorporate that or not let it affect my circuit?

class five
oct 26th

Patience is the key.

After going through the circuits with the class, I finally got something to work. The image below and its circuit schematics (to the right) is a 4093 chip fully in use [click on them to see the origin tutorial]. All four investors feed into one and another. The spec is also shown on the right. From the poor-quality image, we can still see how it is wired differently than the SN74HC14 that we started. The pins are mirrored and all have three ports instead of two.

As for the previous trails, though it bothers me like the deepest itch that one cannot scratch, we may never know what went wrong in them. The closest guess would go to shitty speaker. We did find some burnt-up chips in my bin, so at least changing new ones in my previous attempts was heading to victory(?).

The thing about debugging a circuit is that so many things could be wrong at the same time that leaves plenty of possibilities for one to get frustrated and give up trying. I am glad to know the four investors circuit is working now. Watch the beautiful babies sing in the following videos.

class five

class four

class four
oct 16th

Oscillator! For god's sake, I swear to god I did the same thing with the tutorial and it just didn't work. Also after taking so many bad documentations, I am finally learning my lessons. Here is a better picture for this experiment.

This is so bad since the schematics are so easy, but this still is not working. I also tried to recreate last week's project by using the cd40106be since that is what the website says we should be using. It did not work either (whether with or without the extra protection resistor coming into the speaker)... Heck, I even tried with the old chip and it's just not working.

I tried all three kinds of inverter chips:

CD4093BE
CD4016BE
SN74HC14AN

Things I checked:

The speaker itself is working
The battery is charged
The capacitor I tried with 103 and 104
I even got 2 new chips from the drawer

Hypothesis:

The speaker is too weak to be heard

After all, this is exactly how our speaker fucked us up in class lol. Everything was correct other than the speaker (it worked with other people's speakers...)

I've also found some other tutorials using different things other than the IC chips we've been using to create an oscillator. One of the examples is this:

Q1 and Q2 are two triodes
C1 is our old friend capacitor 104

This looks very promising. The two triodes are doing similar things as our inverter. I think I will find the parts and try it out.

Something I also want to try in the future is that I want to add an amplifier to the oscillator we have been making. It should be easy enough to accomplish?

class three

class three
sep 21st

Oscillator!

I had no idea what it was until this class. The video below shows my first oscillator. I crossed into some problem when trying to recreate it. I will come back at it.

New Friend: IC Chips! - Labeling habit

Turns out:

When I was trying to recreate the oscillator on the left, the voltage is not correct. I was using arduino which gives out 3.3 volt instead of 9 volt. Weird enough, from what I remembered, I did use a power converter to try sourcing out of a 9 volt directly, and it didn't work. Something else must have been wrong that I could never find out.

Luckily, I was able to recreate the in class practice by connecting the 1A and 1Y with a potentiometer. However, connecting the other A/Y pairs does not give me anything. I did make sure I also moved the resistor that was connecting the audio jack to the 1Y notch. Hypothesis: maybe the resistor need to be different? Or maybe the circuit has to go through the 1A/1Y first before it goes into the second pair of notches?

This image will bring you to the data sheet.

class two

class two
sep 14th

It just works out perfectly for me since I have just started working on circuits and soldering with physical computing. We also went over some basic physical principle and formula which will be very handy as I proceed on this journey with audio hardware and physical computing. I made myself a promise prior my entry to ITP that I will get more physical and "harder-ware" comparing to my past. I believe this class take me closer to the goal...

class one

class one
sep 7th

We are introduced to the mechanics of speakers and microphones by hooking a speaker directly to a battery with crocodile wires and other conductive materials. I tired with a ring-pull can's ring, keys, a metal file, and the wick of a pencil. I think (?) the higher the resistance is, the lower the amplitude will be.

If the speaker is hooked directly to the battery, the circuit will be able to make one simple "dub" sound. By employing the vibration of the sounds themselves (putting the pins/crocodile-heads on the cone), we could make a continuous sound that resembles the sound of flatulence. Yonny (the person sitting right next to me) found out that if we press the pin lightly into the cone, the pitch (frequency) will go up.

As we got comfortable with the simple circuit, we played together as a group of newbie percussion band players.

shuang's itp jorney