Well, not really failed, since every result is a result. I just found a way how this does not work. Two in fact.
Okay, what’s this about?
Oona Räisänen challenged her readers to construct a whistle that could transmit data using FSK.
So, I thought about replacing the pea in a pea whistle with a magnet, and then using an electromagnet outside the whistle to slow down (or stop) the pea’s movement during whistling. Since the pea modulates the whistles sound by blocking the “air exit” of the whistle when moving past it, slowing the pea down for a couple of revolutions should frequency modulate the whistle.
Another idea, in case the magnet is too heavy to move at all, is to use an aluminium foil ball as pea, and see if a strong magnet can slow it down (think eddy current brake).
So, I printed a couple of pea whistles (as designed by Saggo) and replaced the peas during print:
This whistle has a aluminium foil ball as pea.
And this one has a strong neodymium sphere magnet.
I tried both whistles and an unmodified one. They are quite loud, but the magnet pea is too heavy too move with the airflow.
The aluminium foil one does work. However it is hard to compare the sound with/without a magnet next to it, because whistling for more then 4 seconds on these things is challenging. (Nobody has lungs that big, and if you save air the pea won’t move)
I use a magnet salvaged from a disk drive, because it is the strongest magnet I have. If it does not affect the pea, nothing sensible will. Later this magnet could be replaced by an electromagnet, that’s switched on or off to transmit a 1 or a 0. (So for a 1, slow down the pea for a couple 100 milliseconds, for a 0, do nothing.) (However it would need to be a really strong electromagnet. The few loops of wire in the first picture are not going to cut it)
Because it is impossible to whistle for more then a few seconds on these whistles I used a vacuum cleaner.
This drowns out the sound of the whistle, but by filming trough the mouthpiece I can see the pea and theoretically count how often it passes by in a set number of frames. And compare results with/without magnet. (Yes, those are my knees. I needed my hands to hold the camera and the magnet)
I have not counted frame by frame, but the magnet does not seem to have much of an effect. As there is no way that with a few windings of copper wire and some moderate current I get the same magnetic field as with the magnet I am using, so short of whistling in an MRI scanner this idea can be scrapped also.
What’s next? A two-tone whistle with solenoid-operated valves?
Leave a Reply