The Halotron

[whammy_time]

Greetings!

Intro (since it's my first real post)
I've been lurking for a few weeks here as from my various google searches this seems like one of the few goto resources for vaporizing. Other than trying a volcano maybe 4+ years ago, I've always used combustion. I quit smoking cigarettes ~4mos ago and took a long break from ganj at the same time. After 3mos, I was a little nervous that it might cause nicotine cravings, but the experiment was a wonderful success!

Now that I can taste and smell again... smoking weed stinks! I started looking into vaporizers for the first time, and definitely got some sticker shock! Perhaps like some others around here (though few, it might seem, by reading through the arsenals in some signatures!), I wanted to see if I could improvise something functional and economical.

I felt quite at home, as there are some incredible tinkerers around here. My goodness were threads inspiring and impressive. I can't help but acknowledge @Pipes for his incredible work in the Bulli and Project threads, which have given me ideas. It's a testament to the cool factor that the former has been active for more than 4 years!? I quite enjoyed @Hippie Dickie 's toaster, the lab glass vape ( @jojo monkey ) and a super recent one that looks like something out of my own home (@rz).

Anyway, inspiring shit and hopefully I can contribute here among my newfound inventors!

The Halotron
Silly name, but it seems people name their stuff so I thought I'd do no less. Somewhere I stumbled on someone suggesting a halogen bulb as a heating element and started digging. I guess it's common knowledge to many of you, but i had ~zero knowledge of vaporizers prior to maybe 3 weeks ago, so this was exciting stuff! I was most intrigued since it was simple and available (easy to get a hold of). I tinker with microcontrollers quite a bit and wanted to try a temperature controlled system using PID.

Here's my current list of parts:

| part                                | source        | cost    | note         |
|-------------------------------------+---------------+---------+--------------|
| 50 watt halogen bulb                | home depot    | $2.12   | 4 for $8.50  |
| ceramic halogen socket/fixture      | amazon        | $2.50   | 2 for $4.99  |
| honeywell 135-104LAG-J01 thermistor | amazon        | $5.94   |              |
| irlb5721 mosfet                     | adafruit      | $1.75   |              |
| computer power supply               | surplus store | $5.00   |              |
| 29/42 to 24/40 bushing              | ctech glass   | $21     |              |
| 29/42 to 24/40 reducing connection  | ctech glass   | $17     |              |
| 24/40 straight adtapter             | ctech         | $17     |              |
| banana jack binding post            | amazon        | <$1     | 10 for $6.76 |
| arduino                             | amazon        | $22     |              |
| breadboard with jumpers             | amazon        | $6.47   |              |
|-------------------------------------+---------------+---------+--------------|
|                                     |               | $101.78 |              |

I was able to borrow the glass components, had the power supply and the arduino/electronics (minus the thermistor), so I only spent $12 on everything. The project wouldn't have been nearly as appealing to try if I'd needed to buy everything... at $100 I might as well just spend another $30-50 to get a commercial vape.


How she looks like
So, finally, here's the gist! Used a scrap piece of oak for now just to hold the light put and manually got the thermistor placed above the bulb, positioned just by wires holding it there. I only had lead solder, so I used makeshift crimp connectors for now... I'd like a better solution, and getting the thermistor past the glass joints proved tough! I would ideally like to have just the actual glass encapsulated thermistor in the air stream, with the wires not in the heat path... if I ran them on the outside of the bottom glass joint, the next one (as expected) was really far from seated. As is, I don't like that the leads are going to get hot as fuck as well (which plays with wire resistance and overall thermal mass). Anyway, here you are...

Closeup of halogen fixture

Glass bushing to surround the light, thermistor over bulb, and a small screen bowl will sit in this opening

Getting ready for test drive; you can see the three glass pieces (one's hidden a bit but surrounds the bottom piece shown above and connects to the longer piece)

Preheating

Test setup to try and find good PID settings and explore the capabilities of the light

Preliminary data gathered with tmp36 sensor to evaluate PID temp control (only good to 125C)

Trying PID values (top labels in form of "P, I, D") with a 150C setpoint once I had my thermistor

Thoughts from the test drive
I ended up setting up a potentiometer so I could change the setpoint on the fly from 150-250C. I wanted options below my true target of 170-220C just in case I was reading high/low. Even at the high setting, I couldn't get any vapor at all using the PID controls! I think there are a couple possible issues:

• the thermistor is super close to the bulb, so perhaps right above it actually is ~200C, but at the top of the glass where the screen sits, it's not there yet. Moving the thermistor higher could help with that.
  
  
• the method of reading temp altogether could be off and I should really calibrate with boiling water or so to make sure I'm accurate.
  
  
• this NTC style thermistor drops in resistance as temperature increases. Because of the formula used to calculate which relies on taking the log of the resistance (simplification), there's really poor resolution at the top of the temperatures. I realized that mathematically, there start to be very discrete "steps" one can measure based on the analog to digital converter. As the resistance decreases, my steps around the 200C mark are really far apart: 237, 222, 211, 202, 195, 188, 183, 178, etc. This is far from ideal as I just don't have the resolution desired.

In any case, no vapey = no happy.

I removed all the controls all together and just treated the bulb like a lighter I guess. Shortly after I saw my first vapors! They were quite thin and I was super paranoid about combusting, which I'd seriously like to avoid in my garage. As mentioned, I've only vaped once so I had no idea what to really expect. I just kept sucking on those whispy streams, having no idea what temperature I was at. Tasted almost sweet and fruity, though... kind of reminded me of fruit loops.

I got more and more bold, leaving it over the light for longer periods. At one point I coughed and think it might have been from some smoke occurring, but if so it was minimal as it didn't smell/taste like it (just felt harsh on my throat). Other than that one hit, everything felt really smooth.

After a few minutes of this, I wasn't really feeling anything and thought I might need more bud (didn't measure, but eyeballed a medium bowl). I let things cool a bit and started breaking up more bud... while doing that, I realized I was lit! Cool! I added just a bit and started 'er up again. This time I popped the top glass off and stirred occasionally. The concept works, just not like I'd planned!


Mods planned
For the next round, I'll be trying the following:

• improve temperature sensing. One uses a fixed resistor to calculate, and if my math is correct, decreasing this value should give me more steps at the top. I'll try dropping from 100k to 50k or 22k. I'd also like to test in boiling water. Lastly, I'll try to move it up just below the screen. I didn't run my computer in my garage, so I have no idea what the thermal drop is when I take a hit, but that might be good to know (can this bulb keep up?)
  
  
• increase thermal mass/surface area. When I read the Bulli/Project threads, I immediately was thinking that more surface area would help vs. just a coil. We want the air to be able to efficiently remove the heat, and surface area helps immensely with that (think of the fins on heat sinks). Because I have it available, I plan to lightly pack the heat chamber with coiled, flat SS from a cleaning scrubbie. I'm not in love with it touching the bulb, but I think it will help a lot... hold some heat to "cushion" when taking hits, and more efficiently heat the air.
  
  
• i'm likely losing a lot of light/IR since it's, well, a bulb. Since it's out of the air path, I thought of wrapping the outside in foil with the shiny side in. Down the road, I'd like to get some SS foil or shim stock and form some kind of reflector tube. I think this will keep the heat in.

Conclusion
Hopefully this isn't too long! I don't have an awesome sense of what's alright on this forum, but have seen some thorough posts so I hope this is alright. At the moment I'm just happy to have found some other awesome and very resourceful tinkerers and join the fun. I'd love any thoughts/suggestions.

Thanks for taking a look/having me aboard.

 

[jojo monkey]

What's up!! Great post!

What's cool is no matter which way you take the project, you have the arduino as your base to do anything.

I like your idea of heating more than the surface of the bulb. The more surface, the better!

To give you anidea about heat, I heat glass to 430-520 on the outside to hit vaping temps. The cartridge heater is running 600-700 on the inside.

 

[whammy_time]

Hey @jojo monkey : thanks for the encouragement! Maybe I should stick to questions about your vape on your thread and mine over here so I don't tax you with twice the replies

Thanks for the info, and I'll definitely try that. I plan to give a plain stainless kitchen scrubbie a whirl tonight. It appears completely virgin (no additives/soaps/etc.) but I'll rinse with water and/or iso and then do a burn over a gas stove just to be sure I've cooked the hell out of it. Lots of surface area there so hopefully that helps. I also plan to raise the thermistor to contact the bottom of my mesh basket "bowl" to see if that keeps it hotter. I could possibly try routing it around the mesh as well so I'm reading post-herb...

Lots of fun to try and the arduino is quite flexible. Someone else here used halogen and I took note of their PID settings to try. I had a helluva time dialing in what I was looking for which is a smooth output from the bulb. My temperature is pretty consistent with zero I or D values, but the bulb just pulses and goes off once it approaches temp. I'd rather have a +/- on the temp and smooth out the bulb output a bit as it makes me think it's not great for the life (equivalent of pulsing the hell out of a light switch all the time!).

Thanks again and see you around

 

[Dynalowrider]

Please keep us informed on this. Very interesting. Doc

 

[whammy_time]

Super long delay. Sorry for those who were interested. I got pretty busy with house projects and work, but also hit a bit of a wall with this guy. Some issues:

- I have yet to wrap my head around a great way to mount the bulb in that socket as well as mounting that socket to something (wood, stainless plate, other) in order to access it.

- I really, really liked the idea of having true temperature control as it's not clear to me how a lot of other vapes do this. I may be totally off, but I'd guess a low cost vape is calibrated based on some sort of resistance measurement, maybe some calculations based on the heating element, etc. I haven't seen any calling out a temp sensor and feedback. When I used the PID on mine, it was awesome to see that bulb ramp up super hot during a draw.

- glass on glass is hard. I want the temp sensor really close to the herb, but that likely means it has to poke through the wall of a piece of glass yet still allow the herb to be loaded as well as not be in the way of any sort of stem or whatnot. Eventually I'd like not to be tethered to my work bench either, so having a way to clip the glass together simply would be sweet so this isn't reliant on me holding it perfectly!

Anyway, for now I just decided to bite the bullet and get a Boundless CF. Seems super reasonable for price and I'll get soooo much more bang for the buck out of my endeavors with a portable. This was a fun exercise and I'll keep tinkering, but for now I'd like to get results

Here's some sketches that probably won't make much sense, but were various ways I was trying to figure this out... the fourth was interesting, as I contemplated ways to get around that this is glass and oddly shaped. If I could sufficiently isolate the air path, I could use something like concrete to hold a glass tube how I want it. This would also help with heat loss which I think is lowering effectiveness a lot with my current design. The glass tube the bulb is in gets crazy hot. This might also let me play with full on-demand. Push a button and start hitting because literally hardly any heat will get out at all?

We'll see and happy vaping!

 

[KeroZen]

I may be totally off, but I'd guess a low cost vape is calibrated based on some sort of resistance measurement, maybe some calculations based on the heating element, etc. I haven't seen any calling out a temp sensor and feedback.

Sorry to disappoint but even cheap Chinese crapo-conduction vapes have digitally controlled and sensored heaters these days.

But that doesn't diminish your work in any way of course!

 

[whammy_time]

Thanks for keeping me honest with myself @KeroZen! I guess that makes sense... not sure why I assumed otherwise? The way I wanted to do it was to have the sensor directly in the air path either before/after the herb chamber to get it super precise.

So do units like the Fenix, Boundless, Flowermate, etc. all bump their current when one is hitting and the chamber drops? I looked around and see tear down pics reporting a temp sensor taped to the outside of ceramic chambers. That's feeding back to try and keep that constant? Looks like I oversold myself... but I'll still keep tinkering Thanks again!

Edit: I came back in part because I just stumbled on the evic tubo! Man... super cool. It got me reinspired to at least make one version that isn't a piece of wood with a light bulb coming out if it on my garage work bench! I actually tried wrapping some SS coils and maybe it's just me, but I swear I can smell that funny toaster smell in the air as the wire heats up. I still have hopes that a bulb from a hardware store could have appeal for some.

 

[Hippie Dickie]

to have the sensor directly in the air path either before/after the herb chamber to get it super precise.

i put a k-type thermocouple sensor on the hottest point of my heater ribbon coil - the air going to the herb reaches this temperature during the first 5 seconds of the draw ... so i find this sensor to be precise enough.

bump their current when one is hitting

i use a PIC 12F683 to read a MAX6675 that converts the k-type thermocouple to a digital value ... my PID algorithm maintains temperature during a hit - actually, right now, the "D" term is a little too aggressive and the temperature rises during a hit (i changed the heater from a heater wire to a heater ribbon that has less resistance and haven't updated the PID terms yet). of course, this has to be balanced against the physics that limits the amount of heat transfer from the glass oven tube heater to the air flow - i.e. surface area of heater, volume of air passing the heater, etc.

i prefer an all glass, only glass, air and vapor path - so i had to make my own.

 

[KeroZen]

So do units like the Fenix, Boundless, Flowermate, etc. all bump their current when one is hitting and the chamber drops? I looked around and see tear down pics reporting a temp sensor taped to the outside of ceramic chambers. That's feeding back to try and keep that constant?

Right. But indeed their sensor is out of the air/vapor path. On one hand it allows to keep your design say completely "all-glass" or "all-ceramic" in the vapes you cited (although it has to be seen whether the sensor being in the path would have any impact in terms of purity, see below)

On the other hand, it adds lag to the readings. So in theory, your system would be less reactive. Add to that extra lag in the other direction if your heating element is also out of the path and needs to transfer its energy through another material or to a heatsink.

In practice, it means you have to chose between "purity" and "responsiveness". @Hippie Dickie chose the former, and he will surely argue (and rightly so) that his device responsiveness is already more than adequate for the task at hand. But this leads us to the second point...

I actually tried wrapping some SS coils and maybe it's just me, but I swear I can smell that funny toaster smell in the air as the wire heats up

Several things here: firstly not all steels are created equal. I don't know which one(s) you tried? Secondly some people are sensitive to metal smells, some are not, you obviously fall in the first category. Then it's also a matter of temperature, was your coil glowing when you tested? Did you have temperature regulation in place?

I have several vapes with SS heaters directly exposed to the air path (Zion, Milaana, Firewood2, MistVape Touch, the Project OG/Eraser/PureSF, my various custom 510 RDA's) as well as using SS mesh screens (as a bowl and between the bowl and heater) and I'm not able to smell nor taste any of them, at least during normal operation.

But if I torch any screen to clean it or if I make my coils glow then I can usually smell the steel for a short while when it's still hot.

You might try other metals too, Titanium has some success in the e-cig world. The upcoming Nomad will be using NiChrome, it's also controversial for some people but apparently (and again when it's not glowing) it's fine and odorless the way he's using it.

Smell is one thing, now for the safety aspect we really don't know. If the heater resistance rises over time, one can assume that several things are happening. Oxidation for one, and it's accelerated by temperature and high currents. But it could well be that over use minute amounts of the heater vaporize or at least separate... where do they go that's an open question?

I read a horrible study about cheap e-cigs once and they found nickel nano-particles practically everywhere in the tank. But let's not draw any conclusion, it was an entirely different setup.

Finally, having the thermocouple (or another kind of sensor) in the air path? Yes, why not. It depends on the two metals used of course, and it would have to be a bare wires one (with the practical implications that go with it) Those having a glass bead at the end would be probably ok. But if you are fine already with SS or other metals in the path, then a sensor close to the bowl could work.

A sensor inside the bowl or in the vapor path, hm, I would forget about it.

PS: there are other SS or Platinum encapsulated sensors that you could in theory mount in a way such as only the cap and not the wires are exposed, but then you are back to the lag issue...

 

[Abysmal Vapor]

Awesome project dude ! I've been following your thread with interest,i've been tinkering on a halogen bulb vape myself for some time,but i broke the glass parts and it is on hold till i decide i can spare some more $ on DIYs..,I have tried aromed ones ,and experience was quite pleasant. Have you done any reasearch on which bulbs are best for application in vaporizers ? Osram has some models with increase IR spectrum ,there is some special coating which bounces heat back on the coil and it glows even more for less watts,i expect these to be best performing,also there are some other which are suited for microwaves and ovens ,durable to high heat,i guess those with be most durable,anyway i got carried a bit away,so to the thermocouple placement theme. There is one model in particular that has hole in middle.. it would be a nice place to put the sensor,although i havent researched yet on them ,are there such with just bare wires without isolation ?

 

[whammy_time]

Hi @Abysmal Vapor ! Thanks for the kind reply, and it'd be great to pow-wow on the idea. I haven't signed in a bit, partly because I got myself a Boundless CF for more hassle-free medicating Some comments/answers to your questions!

your creation: what kind of design did you use, or might you have a drawing? I love to see what others are doing!

bulbs
You mean you tried an aromed bulb in your design? I had not tinkered with different types at all, and that's really interesting to consider! I did think about it when I ran across the aromed, and it hangs me up a bit... would one want a coating on the bulb to keep the heat in, for example? Would that just get the glass hotter? If we want heat, is it best to be less efficient?

In essence, this is just a resistive coil that glows. In my thinking, the halogen is basically an e-cig coil but it's wrapped in glass so I don't have to care about vaporizing titanium, nichrome, or whatever else. I fiddled with stainless and swear I could taste/smell the metallic nature of it. Ick.

I do think it would be interesting if other radiation could be incorporated. For example, if the air or some material (heat sink) or the plant matter were particularly IR (or other) receptive, maybe this would be a better heat transfer method.

There is one model in particular that has hole in middle.. it would be a nice place to put the sensor,although i havent researched yet on them ,are there such with just bare wires without isolation ?

Whaaaa? That's awesome! Took me a second to actually catch you meant the bulb itself had a hole in it! Regarding sensors, there are thermocouples, but I used a thermistor. It's glass coated, which is cool, though granted it has wire leads right next to the glass so maybe that isn't as critical. I don't think the heat source would have to be right by the bulb, and maybe it shouldn't. The bulb itself might be waaaayyy hotter than we need, but the heat transfer equalizes with the speed and volume of ambient air going past it. You could certainly measure the bulb, but maybe it's internal temp is going to be 700F in order to get air to 380F. There are more and cheaper options for reading 380F than there are for reading 700F!

Anyway, thanks for the post and let me know your general setup. I still wrestle with the best airpath. I never wanted my herb right above the bulb as I feared dust that gradually fell and coated the bulb might cause it to explode. That worries me for sure. So... I'd wanted to have the air come past the bulb, maybe surrounded by SS fins, glass beads, SS ball bearings, then around some turn or corner and to the bowl.

Glass, as it sounds you've found too, was not super friendly to prototype with/modify. It's not even always easy to know what to search for and what's available for fittings. And pricey. I'm probably on pause for a little while, but if you have some ideas I could help replicate here too Thanks again!