Building a Homemade Atomizer

[windblown]

I

"All in favor of reinstating smoke breaks, say aye..."

Aye, aye, Cap'n......

[UncleMidriff]

I'd like to connect that hose to an air conditioning vent leading into the room where Tobacco-Free Kids would be holding its annual convention.

"All in favor of reinstating smoke breaks, say aye..."

Funniest. Post. Ever.

[dnakr]

I can see TB now - sneaking into the air vent putting his plan in place... LOL

[Jason]

It looks like this topic has died, but I think it has enough merit to breath some life back into it.

When I started looking into this, I was thinking a ptc (positive temperature coefficient) ceramic element was the answer. Now I think that's overkill.

Here's where I'm at now, and I welcome any advice or feedback:

A small nichrome wire loop (30-35 gauge) could be heated with a standard 9v battery. Temperatures well in excess of 300F could easily be achieved depending on the current/resistance tuning, a small nichrome loop could reach that temperature using less than 15mA draw from the battery. The 9v should have enough juice (~500mAh) to power the thing for 30 hours or more.

As loosely specced above, the nichrome loop would probably be underpowered for heating a cartridge (and wouldn't survive being jammed up against a cartridge anyway) but would be ideal for heating 2-4 drops of liquid quickly and efficiently.

If done with any competence at all, this should give us an "atomizer" that's suitable for dripping and can be replaced in about 5 minutes at a cost of less than 10 cents.

Anyone interested in contributing thoughts/ideas?

Something I could use some help with is thinking up an appropriate housing for this homemade atomizer.

[Mihai]

I am very interested in this, please don't drop it.

soon, I foresee, e-cigs will be banned in most of the countries, or/until the tobacco/farma companies will get a monopoly on it. Then it won't be as easy as now to buy it or the costs will be even higher than for smoking.

now back on topic


ptcs are great but I doubt they are available/affordable in small quantities. leave them for the next gen of e-cig producers

[clavit]

A small nichrome wire loop (30-35 gauge) could be heated with a standard 9v battery. Temperatures well in excess of 300F could easily be achieved depending on the current/resistance tuning, a small nichrome loop could reach that temperature using less than 15mA draw from the battery. The 9v should have enough juice (~500mAh) to power the thing for 30 hours or more.

Interesting ideas, but...
I wonder if the math holds up. What you're trying to acheive here, is not very different to dripping in a conventional e-cig, i.e. putting a couple of drops on the atomizer and then smoking till it dries out.
Your wire loop is drawing about 140mW of power. I found some data on a dutch forum indicating that a mini e-cig will draw about 0.9 Amps from the standard 3.6V Lithium battery. That would give a ballpark figure of 3 Watts.
The mini may be wasting a lot of power when used in dripping mode, but i would think that 140mW is a bit on the low side to boil/evaporate any meaningful amount of fluid.

Thoughts/sentiments/ideas and criticism is welcomed :-)

[Mihai]

yes, the power needed should be calculated on the basis of the quantity of the liquid to be vaporised.

The current loops have about 3.5 W to get to 290 deg C

Making the wire loop to heat just enough to vaporize the liquid would be great - it will draw less power, so more life for your batt - and it won't burn the liquid.

[Jason]

Interesting ideas, but... The mini may be wasting a lot of power when used in dripping mode, but i would think that 140mW is a bit on the low side to boil/evaporate any meaningful amount of fluid...

I started in on the math, trying to figure out how many joules it takes to boil a drop of e-liquid, guesstimating at the boiling temp and volume then realized it would be tons more fun just to give it a whack. I'll order some nichrome off of ebay and see if I can't fry up some nicotine.

Heading off for vacation next week so I probably won't have anything to report for a bit.

[jigtg]

I did some watt calculations on two types of atomizers.
atomizer.zip
Type 1 has lots of problems so I'm dropping that plan. Resistance is really low and wires joining on both sides would make such a big pile that the whole thing would shortcircuit. Type 2 looks good at least on paper.

With 1 cm diameter, 3.6 volts, 33 ohm/m 0.274mm diameter wire and 32 side turns, it would spouf around 3 watts. Assuming my calculations are correct of course There are some errors in calculations but probably not very significant.
Quick calculation suggests that 46% of surface area would be covered in wire.
0.8mm atomizer would fill 58% and produce ~3.6 watts.

Finding suitable material where to assemble this is the main problem I think.
0.3-0.4 mm CNC drill bits should do.

Octave/matlab code follows:

# d = diameter(as cm)
# wire_resistance_per_meter (ohms)

# type 1

# parameters
volts = 1.5
#volts = 3.2
wire_resistance_per_meter = 50
#wire_resistance_per_meter = 33
#wire_resistance_per_meter = 4.4
side_turns = 16
d = 1.2


wire_resistance = 1/( (side_turns) * (1/(wire_resistance_per_meter/100*d)))
wire_current = volts / wire_resistance
wire_watts = volts * wire_current
wire_len = d * side_turns





# type 2

# parameters
#volts = 1.5
volts = 3.6
#wire_resistance_per_meter = 50
wire_resistance_per_meter = 33
#wire_resistance_per_meter = 4.4
side_turns = 32
d = 0.8
#d = 1.0
wire_diameter = 0.274 / 10

wire_len = 0;
r = d/2;
increment_as_mm = r/(side_turns/4)*10
# let i go from r to 0 in -r/(side_turns/4) increments
# FIXME: distances between side turn points aren't really fixed since they lie on a circle
for i = r:-r/(side_turns/4):0
    # using pythagorean theorem
    # FIXME: this doesn't apply to every second wire
    wire_len += sqrt(r^2 - i^2) * 2;
end

#only calculated half
wire_len *= 2;

wire_len
wire_resistance = wire_resistance_per_meter/100*wire_len
wire_current = volts / wire_resistance
wire_watts = volts * wire_current
surface_cover = (wire_len*wire_diameter) / (pi*r^2)

[jigtg]

freepatentsonline.com/y2006/0196518.html for your amusement. Register and login to see figures. Guess you don't have to pay these guys if you want to make an e-cig...