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

[Egzoset]

Hi, i'm far from having all of the answers but i'll try to make sense. Thank you for the interest anyway!



The heater's resistance provides a path to the positive rail. Does that voltage level get close enough to 0 Volts when the transistor shunts it to the ground? I don't know, but if it does then you have the means to control a multi-colour LED!

Lets assume GP0 can source/sink 20mA and the J1 jumper is positioned to connect that multi-colour LED to your MOSFET's drain where the heater coil is attached, if you synchronize the GP0 output signal with GP2 (the PWM duty-cycle) then i believe it should become possible to control one half of that multi-colour LED at a time depending on wether GP0 is complementary of GP2 or not. In other words, you'd have the means to make the LED flash while its colour changes from red to yellow to green as the heater receives more energy. Remember, we agreed that the thermocouple's purpose is to measure temperature while energy is a function of the PWM duty-cycle... Now, allow me to refer you to this previous post:



I thought we also agreed that temperature and energy are not equivalent. If that is so then it's one thing to power up a heater coil until it reaches some preset temperature but it's quite another to monitor energy absorbtion in order to determine when vaporization is complete.

My suggestion at this point is to forget about the red LED, a multi-colour LED can fullfil the same role and more. Of course, there's no need to remove the copper traces leading to the red LED, no more than it would be necessary to make the J1 selection permanent through soldering. My point is this could be included merely for the simple purpose of exploring future options when the time is right... Put clearly, in your shoes i might want to include pads for J1, just in case. Maybe J1 will prove to be useful or maybe not! J1 doesn't even have to installed, actually.

 

[Hippie Dickie]

Does that voltage level get close enough to 0 Volts when the transistor shunts it to the ground?

it shouldn't with these batteries. The coil resistance is 0.5 ohm, so the current is about 12 amps. Rather than drop the voltage, these batteries can deliver 70 amps (120 amp burst), so the voltage will only drop as the battery loses its charge. So that's not a feasible approach.

And, again, unless i'm measuring the battery voltage (or, rather, current flow), i can't tell whether energy is being absorbed or the batteries are discharging.

From my perspective, i think you're over-engineering and i'm trying to keep it simple.

 

[Egzoset]

It seems i may have failed to explain myself clearly.

Suppose the MOSFET transistor (Q2) is also equivalent to 0.5? when conducting (while heating) then the voltage at the junction of the heater coil (the "load") and the transistor's drain pin should drop around Vcc/2, even closer to 0 Volts if the MOSFET behaves like a better conductor. That voltage can be measured easily using any decent voltmeter.

Though i concur entirely with your opinion that power supply voltage variations due to weakening batteries will cause the PWM duty-cycle data to be contaminated with other data not relative to energy absorbtion in the blend, i still think the amount of heater energy can be displayed in a continuous succession of colour shades ranging from red to yellow to green, or at least in a somewhat similar fashion to circulation lights at major street corners:

Wait
Prepare to inhale
Go Johnny, go!

 

[Hippie Dickie]

The MOSFET used is the Fairchild FDB8832. The rDS (resistance Drain to Source when on) is 0.0015 ohm (1.5 mOhm) -- very cool device. So it should have no effect on reducing the Vcc -- and it wouldn't anyway, since the MOSFET is not across the heater coil -- the MOSFET is merely providing a ground to the coil whose other end is directly connected to +Battery.

It might be more useful to use the color-coding to show when the bud is cashed out --- maybe in a future model.

Status update ...

i located the Deans plug on the vape body:

and i finally cut the copper clips that will hold the heater coil:

and with the MOSFET soldered to the pcb:

i've had a bear of a time getting the forstner centered when drilling the bottom disc. It finally occurred to me to use a 1/16" bit to mark the center, then switch to the 1.25" forstner to make the cut. Duh.

And another bigggg Doh! i finally realized that during one of my code tidy-up efforts, i screwed up the interrupt handler code by trashing the W register before i saved it. Gosh what a noobie fuckup! How i noticed it to fix it i'll never know. Anyway, all better now, and some of the anomolies are finally gone. Corrected code has been uploaded to my SkyDrive folder.

 

[Egzoset]

Good morning!

Well, it seems we don't understand each other so i've put it in graphic animated form:

Wait
Get ready
Go!

N.B.: Yellow light is produced by mixing red and green in the multi-colour LED (50 % PWM duty-cycle).

 

[Hippie Dickie]

Nice animation, but sorry, i don't understand it. Maybe you could explain from the perspective of what the processor is doing.

Here's what's available:

1) temperature from the heater coil
2) output (pin) to turn on green LED
3) output (pin) to turn on the MOSFET

Additional factors:
a) Battery voltage, starts at 6.6v, decreases to 5.0v over 9 sessions

b) Processor/LED voltage = 4.5v

c) the voltage to the coil is battery voltage, and not fixed at 5 volts. My design allows any DC power source from 6v to 20v, e.g. 12v car battery, 18v DeWalt power pack.

d) maximum current sourced/sunk by GPIO is 90 mA, with 25mA per any I/O pin.

So, i think plugging the battery into the PIC (top of the animation) would fry the PIC. Or that 220 ohm resistor at the top of the anime would have to dissipate 70 watts of power -- it would be bigger than the vape itself.

i just don't see the three color LED working in this design. But, also, i don't see a need for it -- this design is working quite well in daily operation. my green LED signals "Go" and that's all i need to know.

What's the difference between "Wait" and "Get Ready"? -- both are non-toking states ... that seems redundant. Once the bToaster makes vapor available to the toker, it doesn't return to "wait" or "get ready".

What am i missing?

 

[Egzoset]

Lets see if this one works better for you:

I may be wrong but i'd say this multi-colour LED can have at least seven easily distinguishable states:

No Glowing Light at all (dark)
Solid Red, Solid Yellow, Solid Green (maybe more)
Flashing Red, Flashing Yellow, Flashing Green (maybe more)

I'm only submitting my opinion that minor modifications to your copper trace layout would render this PCB assembly versatile enough to allow the eventual addition of such features, using nothing more than the proper alternative components and software. Anyway, in the end YOU are the man who can decide what these could be good for! I hope you enjoy!

 

[Hippie Dickie]

Okay, if i understand, you are offering an idea to provide multi-color/multi-frequency signals to the user. From a design standpoint, this many subtly different indications is too complicated - it would require an instruction manual. Anything beyond green/no-green should use a text screen -- or maybe an iPod Nano.

Again, i see no need for this in the current design. The red LED shows when the heater is active and the green LED shows when the bud it hot enough for long enough to be producing vapor. That's all the user needs to know. That and the temperature, which is provided by an outboard instrument.

One problem with your scheme is that the MOSFET must run asynchronously to the LED operation - trying to show RED by turning off the MOSFET is not always available. Also, i suspect trying to run a PWM on the GP0 pin is getting beyond what this PIC can do -- it only has one builtin PWM so the GP0 line would be totally software -- different interrupt clock -- this is getting too messy.

What i really want for toker output is a LCD panel with a serial controller -- can't find one with correct physical size. But even that would require another pin to select between MAX6675 and LCD on the i/o pins (GP3, GP4, GP5).

 

[Egzoset]

I already wrote (see my post on top of this page) that this modified design depends on the ability to synchronize the GP0 output signal with GP2. If that requirement can be met then the flashing feature would be available at all times while the multi-colour LED changes from red to yellow to green as temperature rises. My suggested scenario results in a reduction of your components count by 1: the R3 resistor and associated red LED both disapear, the green LED gets replaced by a multi-colour one and a zener diode is added. During assembly the very same PCB would offer two possible configurations, the new one being described as follows: the LED will glow (red) while heating and turn green when ready, with optional red/yellow/green flashing at any time for any purpose you might find desirable in the future. In my opinion this can greatly simplify the life of your eventual customers, with the exception of those who don't see colours (which is where the ability to flash might become a most valuable feature)...

Think of it, no confusion due to multiple (3) LEDs! Though i'm not saying it wouldn't require some effort to implement the necessary code.

 

[Hippie Dickie]

this modified design depends on the ability to synchronize the GP0 output signal with GP2

which can't (easily) be done, since they are asynchronous actions.

Thanks for your comments, maybe in a future model.

 

[Egzoset]

You're most welcome but i sense you missed my point relatively to making the PCB versatile enough NOT to need future models, actually!

 

[Hippie Dickie]

no, i didn't ... it just doesn't fit into my design ethic to modify the design for something that is not going to be implemented.

Which is not to say i don't have a number of different signals to the user, which are based on flash sequences of the green LED: 3 flashes on 2 second interval = ready to start; on solid = i have heard your command and obey; plus several different flash sequences to indicate different error conditions.

So there is, in fact, a user manual to describe the flashing green LED. And tips on usage.

And i am quite in love with the way the 3 LEDs operate now. i'm satisfied with this design as a commercial product. Iff i can actually fabricate it.

 

[Egzoset]

Oh, in that case it's only natural not to explore the matter any further! I was ready to submit another scenario where the GP0 line is not synchronized with GP2 but this doesn't appear to be suitable at the moment.

I still think it would benefit the vaporists community if you could carry on with the Bud Toaster project, as a result i wanted to mention the "Vape Pen" to you the moment i heard about it, just in case:

http://420tainment.com/2010/05/vape-pen-discreet-vaporizer/

I wonder... By any chance, whas it a source of inspiration for your vaporizer or are there things about it which you might find stimulating, eventually, even though it probably won't lead you to a practical transposition in a near future?

 

[Hippie Dickie]

Thanks for the link ... that is not strictly speaking a vaporizer, but rather, an atomizer. The first Bud Toaster predates that type of device (there are many versions of the e-cig) by about 8 years.

 

[Egzoset]

Is this using high-voltage instead of heat?

If so, have you tried similar "atomizing" methods before choosing a simpler heating coil configuration for your own project?

 

[Hippie Dickie]

The atomizer requires a special THC fluid (propylene glycol? polyethylene glycol? -- there's a thread on FC about it) that i don't know how to make. Perhaps it uses ultrasonics to atomize the fluid. So that whole concept never crossed my radar screen. Plus, that's a level of fabrication that is way beyond my basement workshop.

Really doesn't appeal to me.

 

[Egzoset]

I didn't realize such designs depended on custom liquid cartridge refills, this completely kills my interest about the vape pen.

 

[Egzoset]

In your opinion, would it be suitable to make the Bud Toaster compatible with this product?

 

[Hippie Dickie]

it already is, sort of, except for that nasty metal screen -- ewww. Just wrap the heater coil around the outside of that glass tube and you've got the Bud Toaster.

Wait ... is that a bong? does that have water in it? Then no, the water would boil off before reaching 400F

Okay, after looking at the website, there's no water used. So, okay, just wrap the heater coil around the tip and you're good to go. Just don't touch the end.

But, what's the point? 50 Euros for a borosilicate test tube? i'm getting mine for $0.10 USD.

And ... just say NO to butane.

 

[stinkmeaner]

The atomizer requires a special THC fluid (propylene glycol? polyethylene glycol? -- there's a thread on FC about it) that i don't know how to make. Perhaps it uses ultrasonics to atomize the fluid. So that whole concept never crossed my radar screen. Plus, that's a level of fabrication that is way beyond my basement workshop.

Really doesn't appeal to me.

You'd be surprised that design is actually more complicated than an E-cig atomizer besides the housing, batteries, etc which take factory level fabrication. The actual function of the atomizer is pretty crude, basically just a heated wick soaked in E-Cig liquid, the wick is wrapped in a hair thin NiChrome wire element.

+2 on that silly Vapbong being overpriced, I could see if it was $10 at a gas station.

 

[Egzoset]

Hi,

At least four coaxial tube designs similar to yours can be found on the internet:

[ins]* [/ins][ins]Ubie[/ins][ins]*[/ins]
[ins]* [/ins][ins]InaVap VaporPipe [/ins][ins]*[/ins]
[ins]* [/ins][ins]VripTech Vapor Bat [/ins][ins]*[/ins]

...and

[ins]* [/ins][ins]VapBong[/ins][ins]*[/ins]

The later product is the only one to have a bottom screen in order to prevent direct contact with the herb, yet none of them comes with a regulated heat source... I suppose a silicon carbide (ceramic) filter would be preferable to some brass screen but lets agree that a metal screen is much easier to shape! Also, i'd say a screen helps to keep things clean and vertical, making the VapBong compatible (*) with your Bud Toaster project (if dimensions can be ignored, that is).

As always, you're in charge. If you decide such things are trivial then so be it!

 

[Hippie Dickie]

??? ... i don't know what you're asking.

Bud Toaster is not trying to be any of those other products you list.

If there was a vape that does what the Bud Toaster does, i would buy it and use it. i built this kitty because it doesn't exist anywhere for any amount of money, and it is exactly what i want to use to vape weed.

i think, therefore it is.

 

[Egzoset]

Asking?



I'm merely suggesting that the Bud Toaster could take advantage of some potential compatibility with existing products, and maybe even share common features. I like the idea of a rubber piece holding the glass tubes together while sealing the air paths, for example. Rest assured i'm not asking for anything other than your opinion.

 

[Hippie Dickie]

i guess i just don't see the advantages you think might be there.

For me, rubber is right up there with metal as an ewwwww factor ... just say NO to rubber, plastic, silicone getting anywhere near the air path. Sealing the air path in my design is not a problem and i don't want to hold the glass together.

As much as i'd like to eventually sell a million units, the Bud Toaster isn't for everyone ... i think the $500 price (complete and with free shipping) will put a lot of people off. That's why the design is open source.

 

[budballer]

wow $500 each?? how much would just the unit without batteries or any other accessories cost? just one block with heating element and temp reader plus a whip and bowl?