PC Fan Generator

[jadaro2600]

I've obtained some diodes and assembled a small
test coil with one of them, and so far no results.

The diode is a Radioshack IN4005 with the following
specifications -

Foward Voltage Drop: 1.6V
Max. Surge Current:  30A
Foward Current: 1A
Reverse Current: 10uA

I've checked for resistance with my meter on the
assembled coil-diode circuit and its ok, so my guess
is the diode is of the wrong type.

My meter was set to the 10V range, and recorded no
movement at all no matter how close I held the coil.

You need to use a Schottky Diode or the radioshack switching diodes - 1n4148 ( 0.563V drop ) the small glass diodes.  Your current diode has an exceedingly high voltage drop.

They're cheap and have an acceptable voltage drop. 4-5 dollars for a pack of 100 if you're radioshack is well stocked.

[geotron]

That's as I thought - sadly my Radioshack didn't
have any 1n4148 diodes though, so I've been looking around
online and found a Toshiba 1SS389 at Mouser.

Mouser.com - Toshiba 1SS389 diode

It's got a sister diode that is a bit more expensive,
but has a 30V peak-reverse-voltage versus the 10V of the
previous one. I'm not certain which to get, or if both would
work equally well....   ?

Mouser.com - Toshiba 1SS416CT diode

Both have a .23V foward voltage, and are in Solder Pad form.

[jadaro2600]

this diode has great properties, but it's ability to withstand reverse voltages is low, since you have powerful magnets, the reverse voltage should be pretty high with forward drop pretty low you may be looking at an expensive diode.

I would settle for a 1 volt drop and a 300 to 400 volt rating... you may tr the 1n4001 or 4003 and see what the voltage drop is, you could measure the voltage drop yourself ..just connect it in series with a 2k resistor and measure the voltage across the diode terminals.  Use a AA source.

some ammeters have a diode tester tells what the voltage drop is.

[Tito L. Oracion]

hmmmm.....
This is a good idea 

How about a big magnet on top of those magnets then a lot of coils then big magnet again. its like sandwiching the coil with the magnets, of course magnetic field should be moving when fan is turning, well i think it can make excess energy . 

actually we can also put a setup from the top and place other at the bottom of the fan, and that is two birds in one bullet. 

Double sandwich in one fan. 

i think we can change the fan blade by a plain materials then glue the magnets there, well this is just a wild idea. 

[geotron]

I went ahead and ordered 50 Toshiba 1SS369 diodes with
the following specs -

Reverse Voltage: 40V (max reverse 45V)
Average Forward Current: 100mA (max peak current: 300mA)
Forward Voltage: 0.28V at 1mA (typ), 0.36V at 10mA and 0.54V at 100mA
Total Capacitance: 18pF (typ.)
Operating Temperature: -40C to +100C.
Toshiba 1-1F1A (similar to SC-70 or SOT-323) Package

jadaro2600 - I hadn't read your post about the reverse
voltage before I bought these, so - I guess if they aren't
good enough I can eventually use them for something huh?

If 40-45V isn't enough, could I somehow wire multiple diodes
together to increase this value?

I went to Radioshack previously and got two varieties of
diode; one labeled IN4005(1.5V) and the other without a label
but having a 1.1V drop.  I tried the 4005's and they didn't
do anything, but the other ones I haven't tested yet... perhaps
this is something I ought to experiment with.

Just for absolute clarity, this is what I am using
for my testing procedure -

Coil -> (+) goes through diode -> lead connects to meter
     -> (-) connects to meter

---------------------------------------------

mscoffman - I've found a much larger 120mm fan, and by
your suggestion have begun positioning magnets onto it
in an alternating pattern. A bottle cap is glued onto
the center position, holding a used DVD as the surface
onto which the magnets sit.

Setting up my 50mm fan with dual coils for the N-S top
bottom arrangement... I'm thinking I should wire each
one with a diode VS wiring them together and using a
single diode.