FM-1500 Fog machine heat exchanger melted

Today I was given the smoke machine that we tried to use at a gig a few months ago where it tried to burn down the hall. Thankfully it only burnt a hole in the floor a bit.

So I am going to try and repair it as the only problem seems to be with the heat exchanger(Al structure fully melted and in a puddle in the bottom of the case) and somewhere in either a relay that controls the heating element or the temperature probe. My gut feeling is that the temperature probe is to blame and hopefully I can find a way to test it(perhaps with the oxy-acetylene torch from school). When checked with a multimeter the temperature probe gave a reading of between 0 and 3 ohms at room temp (about 30 degrees C) and a voltage reading of 0V(2000mV scale). If it were a thermistor I would expect the resistance to be somewhat higher and if it were a thermocouple I would expect a higher voltage(from a little reading on Wikipedia).

Here is a good overview where we can see the piping that was inside the heat exchanger and the Aluminium that made it up all spread out.

A view with the heat exchanger unit removed from the case.

The nozzle(I did mangle it pretty badly when trying to remove it and appear to have shorn through the silver solder(?) that held it onto the end of the smoke pipe.

The heat exchanger module. Here you can see the coils of pipe to carry the fog juice, the horizontal heating element and the puddle of solidified aluminium from when it overheated.

Just a quick picture to remind me which way to connect the temperature probe. It appears to be polarised which makes me think it's a thermocouple. If it is a thermocouple and a wire has broken from repeated heating/cooling cycles then that may explain the lack of voltage and the lack of voltage would explain the amount of heat applied to the heat exchanger when it overheated.

AVRs on the lighting desk

I've been looking at Amtel's AVR microprocessor chips and they look like the next step for my search for a simple cheap lighting desk...

According to this wikipedia page the smallest amount of RAM that an ATtiny has is 32 bytes... now I have 15 8bit fader channels and 12 boolean channels... by my maths that makes 132 Bits... According to the Wikipedia entry on bytes it says that 1 byte ~ 8 bits on most systems... using that conversion I need 16.5 bytes of RAM for my variables, having about half of the RAM in the system means I don't have to worry too much about having to make my code too compact even on the smallest of microcontrollers... I hope...

As to the problem of needing more ADC channels then are provided I can use the I2C bus or 2 wire bus to attach extra ADC chips to the AVR.

That's all for now... exams are almost over and I have a job coming up shortly that should give me some money to spend on this stuff so I can finally get this desk working properly...

Cheers,
Rex