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

[yayvape]

@CentiZen
heh, I had considered that LCD. It is still too large for my liking. Maybe something out of an old mobile. The other issue is the difficulty in building a case around it.
I'm using an Attiny84 MCU for my version, I'm only using 5 GPIO at the moment, 3 for the thermocouple chip, 1 for the FET and 1 to sense the battery voltage (to cut off the power when the battery falls below its minimum voltage)
This leaves me with 7 GPIO for other stuff, so I could actually use a parallel LCD like that one (in 4-bit mode) and still have a spare pin (with ADC) for button input

@Hippie Dickie
Thanks for the extra glass cutting tips. hopefully I can apply them.

i definitely don't want it to automatically start when plugged in.

Why not?

I'm still tossing up to go all out with an LCD and UI, or go super simplistic. At the moment, my vape has zero input from the user and it's perfectly operational. plug and vape..

 

[Hippie Dickie]

Why not? (autostart)

to protect a possibly impaired user.

in fact, if the start button isn't pressed within 5 minutes the cube shuts off. the previous version had 3 buttons, but i can't figure out what to use the extra button for -- maybe select temperature profiles. so i removed the 3rd button and simplified the circuit board layout.

i depend on the voltage regulator dropout to protect the batteries. the Vreg stops at 5v input, which is well within the range for these power cells (2.5v each).

i'm going with the super simplistic for this iteration, but still allow some user control.

i highly recommend a BBQ thermometer display -- there is no other way to know the cube is functioning as expected.

 

[Alan]

I tried cutting glass tubing with one of those wheels one time (not much luck) and then I discovered a slick way to cut them. Spin the glass tubing in a metal turning lathe (an electric drill would work) and carefully cut a deep groove into it using a triangle diamond file and a drop of water. A gentle pull and flex gives a perfect cut every time. No need to heat the glass. The rough ends can be shaped using the diamond file and a drop of water. It is important to use water to lubricate and reduce the heat/friction on the glass. Add drops of water as necessary. The water also keeps harmful glass dust from getting into the air. The end can be flame polished after you clean off the glass dust. It will look like it was done by a professional. Just be careful not to touch the file anywhere else on the glass or you will have some ugly scratches. Happy cutting and be safe.

 

[yayvape]

@Hippy
re: turn on, I reckon simply plugging the deans plug in is enough of a hurdle for the impared user. but each to their own.
re: voltage cutoff, yeah you are taking advantage of the A123's great tolerance for discharging to a low voltage. Because I am using lithium-polymer cells, I have to be more careful. li-po can't be discharged below 2.8V/cell. Its really not a big deal though, only required a couple of resistors, a spare ADC pin and a line of code to protect the cells.
re: BBQ thermometer. I was thinking about it, it is a nice miniature display.. but I'd prefer something integrated into the main circuit

there is no other way to know the cube is functioning as expected.

well maybe, if you don't have any spare pins to output a temperature display
but I happen to have plenty of pins. Here is the result of an afternoons work, getting an LED display going:

@TV
good suggestion. My next attempt was going to be with a dremel diamond cutoff wheel, with the tube in a bench drill. After I get some more tubes..
I am starting to think my glass cutter is at fault too. out of 15 tubes I tried to cut, zero broke perfectly along the score. whenever I tried heating (over various flames for various periods of time) then cooling with damp rag, the tube would simply shatter around the score, or not pull apart at all. very shit.

 

[Hippie Dickie]

@yayvape - that LED display is bad ass! Nice work! i hope you get some herb to do further testing -- i'm anxious to hear what you think of the vapor production.

@TV - thanks for the input on glass cutting.

 

[yayvape]

Thanks, I'm glad you like it. Testing begins tomorrow

I did the LED's using SMD parts freeformed and glued to the wood. the holes are filled with epoxy to act as light guides.
I think I'll include the led bar in my next pcb iteration.

 

[Hippie Dickie]

Do you get any outgas odor when your heater is running?

i have never had a non-herb odor when using my cube (and not much escaping vapor, either) -- much love for that borosilicate glass.

 

[yayvape]

Nope, only when first assembled, I think it was the oils from my fingers during assembly burning off.
I have the tube covered in kapton tape to hold the wire to the tube, won't have to do that next time as I bought some 19AWG nichrome

The vape seems to be doing ok mid session

 

[yayvape]

Ok. After extensive testing.. I think its pretty awesome.
Once you get the draw method that is.
ie breath out, draw slow (the hard bit), then inhale a lung full of air on top of that. pretty obvious but a lot of people don't realise.
The airflow is pretty restricted by the holes in the draw tube getting blocked after a session (I should clean it more often and drill more holes), but slow draw seems to work more effectively anyway. I think, as we were discussing before, a longer heating chamber would help keep the convection going at vaping temperature while allowing faster draw.
It does well when sharing between two people, not sure how it would cope in a larger group. half a vial seems to do the job with one person. a full vial pretty much knocks me out, in a good way I guess but its a fair bit of weed.

Efficiency, I'd say is excellent with smaller amounts but with a full vial it is about on par with other smoking methods (bong/joint). Of course a lot easier on the lungs and a more.. "relaxed" high. I suppose I don't need to talk about the difference between vaporising and smoking around here!
I think a longer heating tube might help with efficiency of larger amounts, or simply make the session time longer and stir. This is why I tend to go with half or third full vials.

I don't really have a huge experience in vaping to draw a comparison from. The only vaping method I had used was a hot air rework gun, which is pretty much a super accurate heatgun. set that to 185C and used a glass cone on a bong which fitted perfectly around the rim of the rework gun. From memory, this setup was a significantly more efficient than equivalent amounts smoked. I should reconstruct this setup for a direct comparison to the bud toaster.

Vapour isn't as thick as yours, probably bud quality or maybe your draw tube is longer than mine (125mm) or it was just the lighting in your video.

I just want to say again, thanks for sharing your design and making it pretty hacker friendly. I really wish you all the best with getting this thing onto the market. It would be fantastic to see them mass produced and sold for <$100.. maybe some day.

 

[Hippie Dickie]

Vapour isn't as thick as yours, probably bud quality or maybe your draw tube is longer than mine (125mm) or it was just the lighting in your video.

i only have access to "mids". my draw tube is 16mm x 150mm -- about the size of a cigar. i think it is optimal. Seven holes in the end, which do become plugged if the herb is too finely ground or after being stirred ... a wooden toothpick is perfect for clearing the holes. sometimes i can just blow them clear.

i never grind the herb -- i break it open if the bud is tight/compressed -- and loosely pack in the vial. my typical pack is 1/8 g.

i don't think i have any restrictive air flow, but i don't have anything to compare it to.

also, the vial has that 1/8" gap at the bottom above the oven tube, and notches around the edge to allow airflow to the bottom of the vial.

And yours performs like mine in that a very slow inhale is best. This gives more time for the hot air to caress the trichomes on the way past the herb. Like a sigh.

In fact, when the Bud Toaster finally goes retail, it will be called the Vapor Ψ Psi to help emphasis that breathing technique.

 

[CentiZen]

I really want to make my own Bud Toaster clone. This will probably be the biggest project I've tried yet; I mostly stay in the realm of digital with my work with only a bit of hardware design and development in the past. Either way I think it will be a good project to spend time on and even if it doesn't work out I'm sure I'll learn lots.

I'm not sure which is the better option; to use your code with a PIC controller (which I have no experience with) or to write my own controller using Arduino on a Teensy 2.0 which I have many of. How long were you working on your temperature control algorithm and other related code? I'm leaning towards the teensy since it's what I have more knowledge of.

My circuit is probably going to differ as well. I like the A123 batteries but I think I want a plug in design for it. Teensy will control a MOSFET which will control the power going into the coil through PWM. Two more quick questions, in some of the pictures your using loosely coiled nichrome and in others it's tightly coiled. Which works better? I'd assume slightly loose would be better than tight but that too loose would be too cold. As well, is there are particularly good source of element grade Nichrome you'd recommend?

Thanks for being awesome HiDi, not many people would develop a vaporizer and then release the source for all to see. You rock.

 

[Hippie Dickie]

Thanks -- open source is kind of a new concept for me, but after reading Stallman's credo and having used his EMACS (in the PC version called Brief) and realizing he's a genius, i thought it was worthwhile to test the concept with this project.

i am also mainly a digital head, but once i found the MAX6675 to read the temperature probe, i just took a black box approach to the design and plugged everything together and it just worked as i hoped it would.

Except for the MOSFET. i had some emails from a guy who was horrified by what i had done on the gate, explained to me why the MOSFET got hot enough to fry an egg, and helped me understand the spec sheet and get the gate voltage up to 4.5v where it needs to be for efficiency. (Open source paid off for sure) That was a problem i recognized i had, but i had no clue how to solve it, and it magically got resolved. The vape still worked, just not as efficiently as it does now.

Even after 3 years i'm still diddling with the PID algorithm to control the temperature. It took about a month to get the inital code up and running. I didn't use PID for the first year (or so) and the temperature would not really settle down. PID is better for sure, but there is a gross overshoot (trying to get to vape temp super fast) and then an undershoot, and THEN it settles (for the most part).

PID is still a work in progress. i think the main problem is the gains are still too large in some region of the error curve. By knowing the response of the coil, it should be possible to calculate the exact adjustment required (as function of error) to maintain the SETPOINT -- at least i think so.

The 12F683 PIC is cheap (about $1) and only 35 instructions, so easy to learn, but perhaps go with what you know and feel comfortable with.

The loose looking nichrome coil must be an optical illusion with the photo. It is always wound tight to the oven tube. i wind the wire tight on a 1/2" dowel and then slide it onto the 5/8" diameter oven tube.

smallparts.com has nichrome 80 - the medical grade. i used nichrome 60 in the beginning (10 years ago), but that shit definitely corrodes over time. The nichrome 80 always looks brand new.

 

[CentiZen]

Well it would be great to be able to see the PIC code; though I have a limited knowledge of c it will give me enough of an idea of the degree of work involved with getting it. Now I am leaning more towards towards using a PIC because I own a programmer from an earlier project and at that price/having the work done for me it really can't be beaten. I wanted to make it with a Arduino though because with the community around it; it makes development alot easier when you run into a snag. Plus it gives me the chance to show of my "Arduino powered vaporizer" to all of the other Arduino nerds I know. I'll probably end up doing that for a 2.0 version. Unfortunately I get a "Document deleted or moved" page when I try to access your PIC code.

Also, what in particular were you doing with the MOSFET wrong? Or is it somewhere in the thread I may have missed? I thought they were fairly straightforward

Thanks for all the other tips and I'll keep you guys posted on how I do.

 

[Hippie Dickie]

See if this gets you to my SkyDrive public folder:
https://skydrive.live.com/?sc=docum...c#cid=02B3F5D74F36685C&id=2B3F5D74F36685C!114

And here is the code folder:
https://skydrive.live.com/?sc=docum...c#cid=02B3F5D74F36685C&id=2B3F5D74F36685C!229

Code is all in assembler with a large portion of macros.

i had the red LED on the drive to the MOSFET gate, so the voltage to the gate was around 2.5v -- the MOSFET doesn't get efficient at current transfer until the voltage is >= 4.5v, so it was getting wicked hot. i moved the red LED to the ground side of the MOSFET and that cured the problem.

 

[CentiZen]

Ah, my old nemesis; Assembly Language. Good to see you again. I haven't delved into this stuff since learning the 8080 in high school. Still, it's good to know I have access to it. For simplicities sake I will try with a PIC but since my design is different and the fact that I would like to add things like a screen and other mods to it means that I should probably end up developing my own in something I am more familiar in working with. Otherwise I just feel like I am sticking a square into a circle shaped hole.

That does make sense with the MOSFET though; I hadn't thought of it that way. And now looking at everything the way I am, I realize that the Inhalater portable is essentially a Bud Toaster in a well designed package. Don't know if they got the idea from here or developed it in parallel but it works by PWM'ing a nicrome coil which is wrapped around a outside chamber with another being inserted into it. The only difference is that it uses polyimide instead of glass.

 

[Hippie Dickie]

yeah, polyimide is very cool stuff. no melting point. i use it for insulation against the nichrome heater coil. Not as slippery as teflon, so must order larger inside diameter.

i basically use macros to treat assembler as a slightly higher order language -- macros help abstract the coding just a bit, and make it a little more readable.

 

[CentiZen]

Ah, now I see what you mean. I never got into using anything like that when I was working with the 8080's so I thought you were talking about using keyboard macro's to speed up the act of programming.

I went and checked out smallparts and unfortunately I could only find the 60 grade nichrome wire. Places that I have found that advertise 80 grade nichrome want large purchase volumes or are off shady Chinese trade market sites like Alibaba. Is the 60 grade still decent enough to use without issue?

 

[Hippie Dickie]

Yeah, i love macros. i once wrote a program that did a data format conversion, where the program was laid out like the two different data formats -- input format and output format -- and macros did all the actual conversion. (amazing to me what crap sticks in my memory)

re: nichrome 80 --- you had me worried there for a minute:
http://www.amazonsupply.com/dp/B000FMW8X0/ref=sp_dp_g2c_asin

i see that "anti-rust liquid applied" which may explain why this stuff does not corrode like the 60% does.


Product Description

Nickel-Chromium (Chromel A (80-20) is one of standard alloys for electrical resistance wire for heaters and electrical appliances. Various trade names and compositions exist for these alloys. The composition is 78.4 percent Nickel, 20 percent Chromium, 1 percent Silicon and 0.05 percent Iron. Chromel A has a continuous service temperature of 2150 degrees F (1177 degrees C), which is higher than the more common Iron rich alloys on the market. Chromel A also performs better in high temperature corrosive environments and is used in the medical device industry. The smaller gauge wires are used to make inoculating needles and loops that are used in microbiology labs. Cauterizing instruments can also be made utilizing this alloy. Other applications include resistors, rheostats, resistance thermometers and potentiometers. Typical uses: hot-wire cutting, heater wire, inoculating needles, cauterizing instruments, AND BUD TOASTERS.


Nichrome 60 functions just fine, the resistance is slightly higher (5% maybe), but with a PWM that really doesn't matter -- changes the PID gains a bit.

 

[Hippie Dickie]

well, this is still a pretty preliminary prototype, but at least i've made a committment to a design decision, viz. using printed circuit boards to electrically connect the two power cells together.

This comes on the heels of a final approach for the power cell itself. i had this revelation about what i was trying to accomplish for safety of the power cell, and stopped being hung up on the shrink tubing. And, wow, the use of shrink tubing was a difficult idea to discard.

once i realized that all the heat shrink tubing was doing was to attach the end caps onto the battery, i realized that a good industrial tape -- metal duct tape for this prototype implementation -- would work just as well physically, and far better over all since the cell is not damaged from overheating with the fucking super hot air gun needed to shrink the tubing! Some ideas are like children and are damn hard to let go of.

and along with this, a new approach to polarize the cell so they can't be plugged in backwards. So, here's the new power cell:

the plus end has two wire nubs spaced 3/8" apart:

while the negative end has two wire nubs spaced 5/8" apart:

Also, notice that the plus terminal is in the center of the end, while the negative terminal is offset 1/8" from the center. The openings in the plastic red end caps are 1/4" diameter.


Now, the two circuit boards use copper rivets as the connectors -- currently 3/32" dia. shaft, 1/4" long, head is about 5/32" diameter:

these will be single sided pcb, lots of copper (2.5 oz), and simple to make and assemble:

so here is the assembled power pack:

the two wire nubs on each end of the power cell, going through the pcb, provides a very stable configuration. now i just need a replace the rubber band with a latching mechanism to hold the two pcbs against the ends of the power cells.

and, yes, the female plug on the power module is positioned to plug directly into the cube:

i've tried this configuration a few times, but it is too heavy and unwieldy to feel comfortable while toking. so, while i'm still waiting for the ultimate battery technology to appear, the standard use will be with a 3' cable.

so i've been testing this design for a couple of days. i was unsure if the copper rivets would make adequate contact with the ends of the battery (through the red cap) and happily, it works just fine.

now i just need to work on the aesthetics a bit - a nice wood trim piece on each end of the power pack. maybe a leather wrap-around cover for the two cells.

 

[LooneyTune]

wow I'm so glad to be back to this awesome thread after so long and still find you active as always. I'm proud to say that I had read the whole thread before posting some questions a long time ago, of course I've also read the posts I missed between then and now

I'm planning to build a prototype, but I'm taking it too easy cause I have some many other projects as well, however I can honestly say that I really understand now how challenging it is to build a proper all-glass-interface vaporizer. The mechanical construction is challenging itself, and you also have to measure and monitor what's going on with the airstream; it's not just turn it on, hey I see some vapor, cool I'm done, inhale it. So yeah, thank you for you continuing efforts in pushing the vaporizer frontier!

ps the new forum design is awesome too

 

[Hippie Dickie]

@LooneyTune - thanks for checking back into the thread.

i'm closing in on a final latching arrangement for the power pack. i need to try a few mechanical operations to see if the mental image is anything near a physical possibility.

i had the camera out at an opportune time yesterday and took a before/after herb shot:

This is using a temperature setting of 375°F, after a 7 minute session, about .08g

that is as dirty as the herb vial ever gets -- i never bother to clean it.

 

[robert denby]

i love watching the evolution of this device. this is a real interesting piece of work and it looks better now than it ever did.

 

[Hippie Dickie]

major bummer today ... A123Systems has filed for bankruptcy. i really hope somebody continues making the 26650 cells. Must be even worse news for Tesla Motors.

 

[Hippie Dickie]

Okay, so i was rummaging around in the shop last night, trying to find that piece of black leather i have squirreled away somefuckingwhere (could only find the pink leather left over from the Lady Bud Toaster build -- cylinder Model 13), and what do my wandering eyes see, but this:

a AC to 12v,6a power supply (one of two --- now where did i get these from?).

Of course it comes with a stupid plug, so that had to be replaced with a female Deans plug:

Just got the plug installed a couple minutes ago for a first test. And, lo and behold, the PID algorithm works just fine with twice the voltage -- maybe a little more stable. and, something i've never seen before, the heater goes completely OFF for a 1/3 second or so, every now and again during the session. Using the 6.6v battery power, it never goes completely off.

don't know yet if i like this or not -- may have to get some input from my most vociferous critic (my gf) -- freedom of the battery pack versus not having to mess with the recharging task.

Otherwise, the session was completely normal. Special thanks to Bubar whose own prototype showed what can happen at 12v and convinced me to throttle the heater during startup to only 15°F per second increase. Works like a treat.

So, now the Bud Toaster is a "plug in vape", too. And will obviously work off the cigarette lighter outlet of an automobile just fine. BT accepts up to 20vdc ... i should try it with a DeWalt power cell.

 

[LooneyTune]

hi Hippie Dickie, maybe this was mentioned before but I can't find anything in my notes. Did you find that a certain strain profile of herb is more suitable for vapping? I'm mostly thinking in terms of sativa/indica, although nowadays we also have autoflowering, hydro etc.