The Bud Toaster - (currently: Model 14, version 3)

Circuit Diagram for PWM controlled heater. Jan 10, 2010

Software for PIC 12F683 and current Bud Toaster photos posted on my Windows Live/SkyDrive:

Bud Toaster project folder

Schematics

PDFs of assembly language program for 12F683 - 6 files

photos of Lady Bud Toaster - Model 13 Version 5

photos of Perfect Cube - Model 14 Version 2


Note: All images have been restored - there had been a problem with Image Shack, but that has been fixed.



Any DC voltage from 5v to 20v is acceptable. This prototype uses two A123Systems LiFePo batteries for power -- about 6.3 volts. The heater coil is 0.5 ohms, so the maximum current to the heater coil is 12 amps. The batteries can provide 70 amps continuously, 120 amps for a brief surge. This gives about 15F per second heating rate. With no current, the coil cools about 2F per second -- very asymetric, eh?

There is a control panel for user input/output. Three buttons are configured in a resister network so that each button feeds a different voltage to one Analog-to-Digital Converter channel on the PIC. Button 1 increases the setpoint temperature by 20F, Button 2 makes the current temperature the new setpoint, Button 3 decreases the setpoint temperature by 20F.

And here is the first pre-production prototype using this custom PWM heater control:

There are two LEDs on the control panel .. the top one lights when the temp is within 2F of setpoint, and the bottom one lights when power is on.



There is a red LED underneath the glass tube that lights when the heater coil is getting power.

Kinda looks like a bomb, eh?

I wouldnt have the nutz to take it on a plane hehe.

Kinda looks like a beta of a flux capacitor to me.

how are the clouds?

It actually made me lol when I saw that last pic wow what a rig...me likes

Whoa! What the heck is it supposed to be? A vaporizer, I suppose, but that thing is all kinds of crazy looking. The Bud Bomb, well at least the name suits it.

sick.

where does the medicine go? Which end do the clouds come out of? Im equally impressed and confused! Off topic briefly, hey Josh i see by your sig that you lost your vapor virginity on valentines.. Nice one!

I'd hit it.

When I look at it I picture some tesla coil action going up the center as you hit it which would be amazing

Uh, it's a vaporizer. The bud goes into the "bud vial", which is a flat-bottom vial (borosilicate glass, of course) that is 12mm in diameter (~ 0.5 in) by 1 3/8" long:

The vial is inserted into the "oven tube" that is wrapped with the nichrome wire coil.

This is looking into the top of the Bud Bomb when running:



The red LED is under the oven tube. It gets power from the PWM pin, so when the coil is on, the red LED is on. The PWM is configured to have 10 bits of resolution, so the red LED dims and brightens as the PWM is operating to maintain the setpoint temperature.

The pictures are of the 5v version. i decided to go entirely with surface mount devices (except for the PIC, which is still in an 8-pin DIP socket for ease of programming). This required a change to 3.3v which is what the circuit diagram shows. i'm in the process of assembling the 3.3v version.

Here is a schematic of the airflow through the vaporizer: down the oven tube between the walls of the oven tube and the outside of the vial ... up into the bottom of the vial ... through the bud ... into the glass whip ... into the lungs:



The whip actually sits on top (and in contact with) the top of the vial ... it is shown above the vial to more clearly show the pieces and their relative positions.

This is what the bottom of the whip tube looks like after a dozen or so sessions:



The holes are 0.75mm in diameter and act as the "screen" to keep bud bits out of the airflow. The same hole pattern is drilled in the bottom of the bud vial. The oven tube has no holes in the bottom.

looks good hippie. and beta . but mostly good. you are much farther along than I am, that's for sure. hooray for open source!

thanks for sharing,
-skippy

i've been working on this for a while ... the first Bud Toaster was created in 2002. The design was based on the Eterra/Aromazap -- fixed voltage, fixed resistance, depends on Ohms law to set the temperature. It worked okay -- took 15 minutes to get to vape temp, then got heat saturated and slowly drifted up in temp. It was a 2" cube. It was mainly an exercise to learn about working with borosilicate glass -- cutting and drilling. And to get experience with DC current.

i was like "who needs a stinkin' temperature control?".

Then i discovered the A123Systems batteries. And started playing with a couple of different PWMs and for the first time experienced near instant vape temp (20 seconds) and long-term, rock-solid temperature control ... it was manual control, but i was hooked.

i started working with the PIC processor in July 2008, and got the code running in Feb 2009. i don't do much hardware design, but fortunately the MAX6675 and PIC 12F683 can be treated like black-boxes and they just friggin' work the way a software head would expect them to work.

i've been using the Bud Bomb for at least 3 months now.

The next version will be pretty ... or at least prettier.

how much?

I like the idea of how it works...I could see it in my lineup one day. Does it give large thick hits?



p.s. nice vape

Thank-you.

i'm still thinking about the price ... i will be making the first 1000 myself so there will be an early adopter premium to the price. (i don't work for slave wages.)

Yes, i get very nice thick hits. And the bud turns a rich dark brown.

units parts cost is about $20, but in 1000s, the cost would be less than $10.

i'm going crazy trying to come up with a decent design to make it fast and easy to assemble. Currently there is a lot of fit and trim to put it together as i work out the details of the enclosure.

so, did you use some type of control theory to get the temperature to stabilize? Is it proportional control or something more complicated? I was looking around on this page:

http://en.wikipedia.org/wiki/PIC_micro

and it looks like Microchip actually just released a C compiler for several of the larger microcontrollers in their lineup. I think they said the 18f, 24f, and 33f lines. You are using a 12f, but it would make it much more practical for me if I could use C code. Maybe I could adapt your code to C. What about Atmel microcontrollers? Have you ever used one of them? According to wikipedia, they are a competing brand and have been supported by GNU Linux C compilers forever. Again, thanks for posting all this stuff.

On a side note, this really puts into perspective what these heater manufacturers are telling me about needing 100's of watts. wth are they talking about... I'm starting to think I should stop listening to the "experts" at heater companies. Maybe I just gave them bad information to work from?

Not really using control theory ... but i am doing a quadratic control algorithm. The algorithm calculates the setpoint error, the change in the error, and whether the error is accelerating or decelerating (i.e. the change in the change in the error). Right now i'm just calculating the duty cycle value (for the PWM) but a table lookup would be much more accurate. But the vape is working so fucking well that i'm pushing ahead with converting to the surface mount pcb (actually four small ones) and the all-in-one enclosure. i'll refine the algorithm after this next version is running.

i kind of agree that 100s of watts would be better - i'm at 72 watts and if i had more current flow i could reach vape temp faster than 20 seconds. The thing i have come to realize, is that by using the PWM, the heater can be any size. So, by extracting all the trichomes from the plant material, the bud bowl could be much, much smaller ... the current flow could hit 50 to 70 amps and the heating would be virtually instantaneous - fast as flicking a BIC. Have you priced a 50amp power supply? That's where these batteries win big.

i think the experts have a very different application in mind. the big guys are ignoring the marijuana vaporizer market ... so far.

About using C: it will never be as efficient as hand coding and there are only 2k words of program space. There is a C compiler available for the 12F683 but it costs extra.

Havent' looked at the Atmel. The PIC is the first micro i've played with since my Apple ][ days. i just love the 12F683 ... 35 instructions and peripherals that in the past would fill an equipment rack all on the chip. For $1.35 in single quantity. Pretty cool.

man I love the all glass vapor path and the high current batteries. Just from the design it looks like it works great. I have used A123 systems batteries before and they rock and don't explode. How do you control the batteries from draining past their no recovery point?. This is a funny video of the killacycle powered by A123 batteries 0-60 mph in under a second! http://www.youtube.com/watch?v=WZ1eMYdLKrs

put me down as a prototype customer. excellent work.

thank-you, thank-you.

the voltage regulator has a dropout voltage that is above the no recovery point -- evidently, as i've got a lot of recharges on two sets of batteries. one charge lasts one day of vaping - usually 5 or 6 sesssions.

The PIC timer shuts off power after 7 minutes - i find one vial of bud can be quite nicely toasted in 7 minutes -- and so can i.

Eye candy ...

Here is the enclosure for the new Bud Toaster ...

top ...


bottom - showing the recharge plugs ...


the battery compartment ...


loading the batteries ...




new printed circuit boards for surface mount devices ...


the new heater: oven tube and nichrome coil ...


user input - a three button array ...






user output - two LED panel ...


power board ...




control board ...




all the pieces ...

Just awesome.

But not a good idea to try to decipher those schematics after vaping

Sorry to say though, the "bud bomb" name is taken. I've got a metal pipe that looks like a little torpedo called the bud bomb.

thanks. and thanks for that info about the name. i was just calling it that for this thread. The real name is The Bud Toaster. (Maybe). i'm awful at naming stuff. it took GF and me a week to name our kid.

Very Cool.

So, does all of the inhaled air pass through these few small holes; and does this twice? If so, how hard it is to draw?