Saw Project: The Implementation

In this post I will go trough all the geeky technical details of how we created the torture chamber. All gadget in the “performance” were controlled from a control center in the next room. I wish I had more illustrative pictures from the construction though 😦

My Accomplices

My Accomplices

The Strobing Light

In the roof of the room there is a randomly blinking light bulb to give the right atmosphere and to make it hard to navigate the room. I found a no longer used lamp socket hanging in the roof above the sinks. I connected a wire to this and led it to the control room. It was then connected to my solid state

Control centre

Control Center

relay pack which was made for the Gas Harp project and controls 8 channels of 2A/230VAC via the printer port on the computer. I made a quick Delphi program to give the random blinking, with three buttons for different intervals plus on and off. The program also controls other stuff mentioned below.

The Telephone

This became more of a challenge then expected. The telephone was of the old type with a thumb wheel for dialing. I only needed the bell to ring and the headphone to speak. The speaker was no problem. A mini jack in on the end of the cable, straight in to the headphone outlet of my laptop. The bell was more difficult. I think the line voltage is as high as 40 – 50 volts and the bell inside needs some kind of ac power to run (there is two bells with a bistable hammer between them so that power needs to be reversed for each bell stroke). I do not know if the central delivers this AC or, if not, how the telephone creates this (there are no active components inside). Instead I disconnected the old crap, save the bell itself, and made a small circuit board with a 555 timer circuit creating 20Hz, driving a relay trough a 2N2222 transistor. The relay then converts 30VDC to 30VAC/20Hz which drives the bell.

Only the finest ingredients
Only the finest ingredients

The voice on the telephone is wav-files played trough winamp. Billy prerecorded all the lines we needed, including reminders if people lost the track and a motivational call if people gave up. I experimented with different effects on the voice and ended up with this: First a hard compressor in CoolEdit, then ring modulation in Cubase, last 4 semi tones downpitch in CoolEdit. The ring modulation gave him a cool rough sound and, most important, removed Billy’s horrible accent.

The Toilet

Just an old unconnected toilet we found in the basement. My companion Billy cleaned it THOROUGHLY an put it in the corner with a tube in from behind. The bowl was filled with strawberry drink mix and hard boiled eggs. The tube was filled with compressed air controlled from the next room, to make it bubble and squirt.

The grinder box

A wooden box with a hole in it. You have to put your hand into the hole to reach the key for the foot chain.


Please insert hand

Inside the box waits a grinding food processor, which is of course safely gaffed. Don’t want any chopped fingers, but the sound is cool. There is also a 12V drill motor with a plastic wire whipping your hand as you tumble around. The device is running on 230V from the Relay Pack. We could just give it a switch, but it was actually less work to just plug it into the relay pack. It is manually switched on when we see on the cam. that the victims put his/her arm in there.

The Moving Camera


We used a friends old video camera with IR night vision to let us monitor our victims. It also gave us some cool footage silence-of-the-lamb style, which I will mix down later. The camera was wired to a telly in our control room. It was quite a power rush to watch our tied down victims struggle in darkness trough the screen πŸ™‚

Anyway, problem was that the room was small and the zoom to narrow to reach the whole scenery. Problem solved by putting the camera on a very slow and noisy home made experimental robotic arm I made in a manic motor control experiment a few month ago. This arm is moved with the help of three drill motors which are controlled with the help of a K8055 USB card from Vellemann. Each motor is controlled with two relays to move in both directions. There is also two PWM outputs on the card which I use to speed control two of the motors by connecting it directly to the gate of the FET transistor that was originally controlling the motor inside the drill. Easy and smooth… This setup was painfully slow but it worked, and I already had the device = less work.

Why I chose Delphi instead of Max/MSP

Max/MSP is very quick and easy to use for a project like this. Unfortunately we have not found a clever way to interface the parallel port directly from Max. In other applications where we have used the relay pack with Max, I have programmed an interface program in Delphi communicating with Max trough a virtual midi port. In this case it would be a lot more work preparing the interface program, the midi port and max work together then it was to write the whole thing in pure Delphi. On a more complex project or a project including more people and/or experimenting, it could be worth the effort of connecting the programs and make the main application in Max.

The Delphi application

The home made Delphi application

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s