Building a Digital Vaporizer (Brain Dump)

[Hippie Dickie]

the Bud Toaster thread is about a dozen below this current one, but here is the direct link:

http://www.fuckcombustion.com/viewtopic.php?id=1024

The PIC code is very simple with only 35 instructions. More of a RISC (reduced instruction set computer) then most other processors.

i haven't posted the code yet, but i do have my circuit diagram in the thread.

 

[skippymcware]

Here are a few more details about this version:

You can think of this prototype as my version of a lightbulb vape. You guys have already pointed out some glaring errors that hinder this thing. This version, however, was never meant to be perfect and never will be. Although I may change it around some, this setup has some innate flaws that will always hold it back. In other words, this vape is more of an exercise to start playing with these components and get some vaping in. Before I am done with this version, I will probably put it into an enclosure to see what kind of difficulties that brings up.

Bill of Materials:
Auber PID Autotune Temp Controller: $35
6" 316 Polished Cleaned and Capped Stainless Steel Tube (.5" OD x .065" wall thickness): $8
300 Watt Coil Heater w/ integral J Thermocouple (.5" ID): $133
Zero-Cross 25 Amp Solid State Relay: $15
Wire: $2
Terminal Block: $4
Switch: $4
Misc.: $15

Total: $216
*If I were to use a nichrome heater the total would be about $100.


I just sat down and did some measuring. It takes the heater 79 seconds to get to 400 F. So far, it looks like keeping the heater at 430 F and drawing slowly gets you an air temp around 375 F at the exit.

On another note, many people have told me that I could not run a coil heater like this in the open air because it would overheat. This seemed crazy to me since the heater has a thermocouple inside it. After calling various heater suppliers and asking their advice, I got the feeling that running one of these heaters without forced air would be fine. Indeed, this thing is rock solid at the set point temperature and varies, at most, 10 degrees. A variation of 5 degrees is much more common. Because there is so much thermal mass in this system, the heater is unaffected by air flow through the tube. More or less, it has no idea what is going on with the air temperature. My plan in the future is to use a .020" wall thickness tube I have and see what that does.

Here is a rough game plan over the next few years (yikes!):

Gameplan:
1:Thick tube, auber controller, zero cross SSR, 300W heater w/ thermocouple (try with and without turbulator)
2:Thin tube, auber controller, zero cross SSR, 300W heater w/ thermocouple (try with and without turbulator)
2:Thin tube, watlow controller, SSR, Nuwave phase control relay, custom nichrome coil heater, in stream thermocouple, forced air
3:Add heater thermocouple and custom logic board between heater and controller to eliminate forced air
4:Replace watlow controller with custom board. Eliminate previous logic board.
6:Replace nuwave phase control relay board with custom board
5:Add fan control and driver
7:Integrate all boards into one

I am going to do some more testing and post the data up here. It should be interesting.


Vap,

Lots of good points. Although it doesn't really change the problem, that mount structure is actually copper plated steel. I will definitely be changing that mount over to wood in the near future. You are right that my goal in the long run is a fast acting system, but this is far from that. Right now, I am just trying to make some vapor and play with this stuff. Since this Auber controller only has a .5 second sampling time, there is no point chasing a fast response system with this setup. I ended up with a 300 watt coil heater with a built in j thermocouple. This was very expensive, but allowed me to get the project rolling ASAP. Coil heaters are sold by a ton of industrial heater manufacturers, and the cheapest one I found was at high temp industries. As noted in my plan, I will switch over to insulated nichrome wire for the 3rd version. As for the tube diameter, I have considered all sorts of options. I think you are on the right track with a smaller inner diameter and thinner wall thickness. I would like to stick to stainless steel, but am considering aluminum. I am a little scared off of copper because it is a pretty established fact that it is unhealthy to cook in copper bowls on a regular basis. This may not cross over into this application, though.


voltaic,

You sound like you have a solid game plan their. If you are offering, I could definitely use some help figuring out a phase control circuit. I want to use this atmel IC:

http://www.atmel.com/dyn/resources/prod_documents/doc4766.pdf

Here is an application guide for atmel power controllers:

http://www.atmel.com/dyn/resources/prod_documents/doc4673.pdf

This is what they generally recommend for air heater control:

http://sigma.octopart.com/65664/datasheet/Atmel-T2117-3ASY.pdf
*This confuses me a little because so many people have told me I need to use phase angle control in order to have the best air heater control.

Here is the text from an email I got back from an Atmel engineer:

"Thanks for your interest on our products. In the technical point of view the ohmic heater load is suitable for phase control. But as I know, in Europe it is not allowed to operate heaters in combination with phase control on public mains. Therefore, zero cross period control has to be used (e.g. T2117B). Let me know the needed voltage and load condition as well the volume of the project behind. Possibly I can provide then a suitable proposal."

I have a lot to learn about working with IC's, but there are some steps in the plan to break into it slowly. Let me know if you think I am on the right track here. I would be very grateful for you advice. I will do anything I can to help you out with your project. Honestly, CFD really isn't my thing, but I have taken a class or two in it. Let me know what I can do.

P.S. I can't believe you like assembly code. Something is wrong with you people!

That's it for now,
-skippy

 

[skippymcware]

I made some changes to the test bed. Here are some pictures:




I removed that heavy mount and replaced it with two small L shaped mounts that I hose clamped the heater tube to. Previously, I had a 40 degree split from one end of the heater to the other at 530 degrees. Now the two ends are equal. I think one of the benefits of using a commercial heater is that the heat distribution is very even across the length of the heater. I also added a terminal block. This just cleaned it up a bunch and made me less afraid I was going to short the controller or burn the house down. I also mounted the switch with something sturdier than the alligator clips I was using instead of proper wiring. (high class, I know)

So, I am still messing with what the best set point temperature is, but I think 440 is too low and 530 is too high. It is hard to sort it out because I end up getting too baked before I have a chance to accomplish much. My last trial ended on 530 F and that was still too high. I have only had one really good session with this thing, so I will sort it out over the next week. Once I do that, I can start measuring stuff and trying to figure out what to change and how to change it.

Not really a substantial update, but there you go.

-skippy

 

[Hippie Dickie]

how dark does the vapor poop get?

 

[skippymcware]

haha, waaaay too dark at temps over 530. If I had to put a name to it, I would call it jet black. At 430 it doesn't generate much vapor at all. This system has a large thermal mass, so it is tricky getting it dialed in. You have to give it a bunch of time to adjust when you make temperature changes. It is really kind of annoying. The heater core adjusts almost instantly. Then you have to wait for the rest of the system to catch up

 

[skippymcware]

All you software and EE heads. I need your advice. I am looking at the ATtiny84 microcontroller. Here is a link:

http://www.atmel.com/dyn/products/product_card.asp?part_id=3829

Since I will be working in C, I figured this unit would be a good choice. It also has some PWM outputs if I wanted to go down that road in the future. My current plan is to use two K thermocouples each with a MAX6675 IC. One thermocouple will be on the heater, and one in the air stream. Once you have those two signals, you can set a max temp for the heater and do a control loop based on the air temperature. By doing this, the heater will only get as hot as a predetermined temperature when there is no flow across the air stream thermocouple. Do you guys see anything wrong with using this chip? It seems like it is a good compromise between the much larger and more complicated ATmega chips microcontrollers.

Also, I was thinking of getting this Kanda development/training board.

http://www.kanda.com/products/kanda/AVR-TRAIN.html

It is a little $$$, but I think it might make my life much easier. Any thoughts?


On another note, I vaped with the heater temp at 460 F the other night. That worked pretty well. The flavors are starting to come through and the herb looks much better. I still may be a little too high. I am going to laugh if I end up back where I started. I guess that's how it goes sometimes. I have changed the whip setup around a bunch since I started, and that may have been the problem to begin with. I am pretty excited that the heater temp is coming down because it means I can use insulated nichrome wire. This will open up a new realm of cheapness and experimentation.

-skippy

 

[Hippie Dickie]

the two thermocouples with MAX6675s should give you very good thermal control. But that approach of anything in the air stream (other than glass and bud) is a compromise i am not willing to make.

the programming board for the PIC is only $35. I don't know anything about the ATtiny84, is it cheaper than the $1.38 PIC?

 

[skippymcware]

Hippie,

Yeah, the max6675 is a really nice product. I definitely owe you one for pointing that out. The ATtiny84 is $2.90. Honestly, I could care less whether it is free, $5, or even $10. I just want something that I can program in C and will be able to accomplish my goals. The ATtiny24/44/84 also have on-chip temperature sensors, which I thought might come in handy. It would be nice, at least in the development phase, to have a temp. alarm if the electronics were getting too hot. It could also be used as a control input for any fan that I might run. That is miles down the road, though. As for the board, that product is a great deal more than a programming board. It is a development board as well as a bunch of software and tutorials. This would not be useful for more experienced users, but is great for me. That being said, I can see the argument for not getting something that has done most of the work already and saving some money.

 

[Hippie Dickie]

also check Wikipedia and YouTube for training information. they are both incredible resources to learn something new.

And there are C compilers for the PIC line for additional money. The development system (with assembler) from Microchip is free.

 

[vap999]

I am pretty excited that the heater temp is coming down because it means I can use insulated nichrome wire.
-skippy

You could increase efficiency very simply, and reduce the difference between your thermocouple temperature and the temperature of heated air, by wrapping the heating pipe in aluminum foil (which certainly is safe at the common oven temperatures it will be exposed to). With your heater now having no insulation, an incredible amount of heat must be lost. A little bit of aluminum foil would greatly reduce heat loss at the critical point, where the heater wire is wrapped around the pipe, by convection and radiation. Since your dealing with temperatures not over 500?F, you could also consider using silicone or teflon tape; and still avoid the hassles of insulation (usually fibrous or powdery, soft or brittle, etc.).

 

[Hippie Dickie]

i played with teflon tubing to insulate the nichrome wire and it makes me a bit nervous being so close to the melting point. and, as i found out, it is quite easy to exceed the melting point. shit happens, ya know?

and i've used the aluminum foil to reflect the heat back into the bud -- that works. but i need to use an air gap to not short out the nichrome coils.

 

[skippymcware]

Just to be clear, when I say I want to use insulated nichrome, I am talking about something like this:

http://www.pelicanwire.com/hightempteflonheatingwire.htm

I couldn't tell whether or not we were on the same page. This stuff actually has a perforated coating wrapped around it so that you can coil it around a metal element without shorting it out. As you can see in the link, the different insulation give you varying maximum temperatures. Unfortunately, they also reduce your heater effectiveness.


Vap,

It is definitely a future goal to insulate the heater. For the moment, I have stopped trying to figure out what the best way to do that is. I can figure that out once I get closer to the finish line.

-skippy

 

[vap999]

Skippy and Hippie Dickie:

Both of you are mentioning problems with heating element isolation/insulation. Something to consider is use of a heat-conducting (relatively, for a ceramic), electrically-insulating, high-purity alumina (inert, nontoxic) ceramic adhesive (the mix the powder with water, hardens in a couple hours type). For example, see check out Cotronics Resbond 920 (at http://www.cotronics.com/vo/cotr/ca_electricallyresistant.htm). There are also similar premixed products sold in smaller (cheaper) tubes and cartridges, sometimes even on Ebay, but these can have a very short shelf-life after initial use.

This would add thermal mass, but could solve some heat conduction problems and make your wire-wound heaters much more mechanically robust.

 

[skippymcware]

hmm, that stuff looks pretty good. I will check it out if I end up needing something like that.

On another note, I have been looking into thermocouple data loggers. I am going to get this one:

http://www.omega.com/ppt/pptsc.asp?ref=OM-EL-USB-TC&Nav=tems08

I will let you know if it is worth it. I am hoping this will help me start to develop a baseline that I can refer to as I change things. Also, I plan to use one of these thermocouples:

http://www.omega.com/ppt/pptsc.asp?ref=IRCO_CHAL_P13R_P10R

You can see the time constants on these things are very very small. Once I get this rig setup, I will be able to post some temperature over time data up here.

-skippy

PS

I have settled on 450 F as the best heater temp for this first vape. I will post some pictures of the herb to get everyone's opinions.

 

[Hippie Dickie]

ceramic adhesive (the mix the powder with water, hardens in a couple hours type)

i tried that when i was using the 22ga nichrome (before i switched to 16ga), for that very purpose of adding some thermal mass. It did provide the electrical insulation but was hard to use for my particular configuration, that is, nichrome wire wrapped around the outside of a test tube.

i didn't have the high current power source nor the PIC temperature control, so it was not particularly favorable ... too hard to form around the glass, would be impossible for the user to repair when (not if) the glass broke) ... plus it took too long to reach vape temp and there was a nasty odor from outgassing of the ceramic paste.

The 16ga wire solved the physical rigidity requirement. And the glass can easily slide into and out of the coil when it breaks.

The PIC solves the thermal mass requirement by being able to adjust the temperature virtually instantly to any size (volume) toke.

i have come to believe that current control is vital for best vaporizer performance ... that is, the performance i want from the vape that i use.

@skippy
i log the thermocouple readings in the EEPROM of the PIC. i can store about a minute worth of readings. it has been a very valuable programming debug tool. i've also used the EEPROM to log other readings - such as button array voltages from the A-to-D converter when i was debugging user input (3 button array), and duty cycle settings as the algorithm adjusts to the temperature reading from the MAX6675.

it's a bit of a drag to constantly plug/unplug the PIC to pop it into the programmer board to unload the EEPROM, but really the best option available for like no money spent.

 

[skippymcware]

Hippie,

Using a microcontroller to gather my data would be ideal, but I am just not ready for that yet. I am trying nail down some of the basics before I start trying to do anything with custom control.

Here are some pics of the vaped herb after a session at 450, 440, and 430.




Honestly, it still looks dark to me. I did my best to correct the color in the pictures, but the green's are hard to see in that light/with my camera. Let me know what you think. I suppose we have to keep in mind that some of the herb may be getting some conduction heat from the hot whip adapter.

-skippy

 

[Hippie Dickie]

i like my vapor pooop to get darker than that. when i look with my 60x pocket microscope (Radio Shack) i don't see any remaining trichomes -- well, maybe one or two.

i'm measuring the temperature outside the air flow.

one thing i notice is that as i inhale, the glass tube used as the whip will heat up on the bottom 2" that is closest to the heater. i'm guessing the air flow is being super heated by the restricted airflow.

 

[skippymcware]

Ok,

Quick update. I mentioned that I was getting some gear to step up the quality of testing I am doing. One of the things I was looking into was a microscope to look at the trichomes. I got this idea from a thread here that demonstrates this and has great reference pictures. Here's the link:

http://fuckcombustion.com/viewtopic.php?id=627

I looked at all sorts of microscopes. My main goal was to get an inexpensive one because I will never use it again/for anything else. Ideally, I would have liked to get a dino-lite digital microscope:

http://www.bigc.com/products_handheld.php

My problem with this one is that you can't set yourself up with even the cheapest version for less than $150. As a substitute, I was strongly considering getting one of these:

[url]http://www.amazon.com/gp/product/B001UQ6E4E
/ref=s9_simx_gw_s0_p114_i2?pf_rd_m=ATVPDKIKX0DER&pf_rd_s=center-2&pf_rd_r=0NVMGW3JVTCF6SANA6AW&pf_rd_t=101&pf_rd_p=470938631&pf_rd_i=507846[/url]

Before I did any of this, however, I wanted to see if the idea would even work for me. To do this, I went to radio shack. Here is the first one I tried:

http://www.radioshack.com/product/index.jsp?productId=2179604

100x wasn't powerful enough and the light is mounted at a poor angle so it doesn't really illuminate the herb. So, I went back to rad shack and got this one:

http://www.radioshack.com/product/index.jsp?productId=3493411

This guy was not only on sale, but worked infinitely better and had 200x, which is what I really needed. On top of this, I tried sticking my digital camera in the lens and it actually allows me to take relatively decent pictures. Super ghetto, I know, but it gets the job done.

On another note, I received my thermocouple stuff today. To recap, I got the following:

http://www.omega.com/ppt/pptsc.asp?ref=IRCO_CHAL_P13R_P10R
http://www.omega.com/pptst/OM-EL-USB-TC.html

The bad news is that the 36 awg thermocouples are about the thickness of a hair and don't have any sheath or covering on them. I completely missed this fact when I ordered them. So, if anyone reading this has any interest in ordering thermocouples, get something like this instead:

http://www.omega.com/ppt/pptsc.asp?ref=5LSC_5SRTC

Live and learn, I guess.

Now that I am equipped to do some actual testing and data collection I can finally post some very meaningful results up here. I will do this asap.

Following up from my last post on the set point of my vaporizer, I was looking at some vaped weed from a heater temp of 450 and, as hippie suspected, there were still some trichomes in there. I guess all I am saying is that I have finally started to narrow this down into a manageable temp range.

More to come...
-skippy

 

[Hippie Dickie]

RE: bare wire thermocouples ... or you can get this

Zeus PTFE (Teflon) Sublite Wall Tubing 28 AWG .013" ID x .0030" Wall 48" Length $12.25

from www.SmallParts.com to use as insulation on the leads.

 

[skippymcware]

Here are some goodies for you...

I setup the thermocouple and logging system on Friday. Here are some pictures of the thermocouple installation:





That silicone around the thermocouple insulation is food grade. The thermocouple is 30 awg with glass braid insulation. The logger samples at 1 Hz.

Here is a pictures of some data I gathered:



You can see clipping from the sample rate being too slow. I can't get anything faster without designing it myself or paying a shit ton more money. This will have to do. One interesting thing I came upon is that, at very low flow rates, the temperature goes up with increasing flow rate. At some point, it reaches a maximum and begins to decrease with increasing flow rate (as one would expect). This trend continues until the flow rate is essentially infinite and the air outlet temperature is the same as the air inlet temperature. Let me use paint to illustrate this a little better:



Let me know what you think about this. I am pretty sure this is what's going on, because the air temperature would sometimes go up when you hit it harder and sometimes go down depending on how hard you were hitting it. If this is the case, I am going to need to try and avoid that circled area. In this region, hitting harder makes the temp increase. Given how fast the air temperature can increase, and how slowly these systems cool down, it seems that this would be a difficult region to control.

Lastly, in order to really nail this whole thing down, I would love to be able to measure flow rate. Any thoughts? It's gotta be cheap. I know there are some plastic medical devices that do this, but they are all pretty pricey as far as I can tell. i have thought of using a plastic bag or some sort of volume that I could calculate and then deflate or inflate.

More to come,
-skippy

 

[NinjaVape]

Hey Skippy,
Firstly congrats on your research, very impressive. With regard to flow rates/voulmes, try this:

Take a large jar that you know the volume of (ideally it would have markings on it).

Turn it upside down and place it in a bowl of water so that the air is trapped inside.

Insert one end of a plastic tube into the inside of the jar, near the top.

Insert the other end of the tube into your mouth.

Start a stopwatch and toke.

As the air leaves the jar it will be replaced with water from the bowl.

When you are finished toking stop the stopwatch and measure how far up the jar the water has gone (OR just measure how much less air is in the top OR how far the water in the bowl has gone down depending on which is easiest for you)

You then know how much air you toked in a given time and can calculate the average lpm (or whatever) flow rate

Hope that helps....

 

[Hennessy1414]

theres 3 members with that same name now... vapeninja, vaporninja, and now we have the smarts from ninjavape

glad to see you at this site brother! always looking for the 'cutting' edge of vape tech. glad you are able to deliver just that. Looking forward to your posts in the future

hell yea skippy! go skippy go!

How much have you spent on this 'science experiment' so far? looks like things are coming along nicley

Have a good one,

 

[NinjaVape]

I screwed up registering last night (wrong email address) and ended up registering vapeninja and then ninjavape...

Glad to be here though, the world needs Vape tech if there's going to be any hope of a future!

 

[vap999]

Check out the digital air heaters (solder rework stations) at CircuitSpecialists.com, such as http://www.circuitspecialists.com/prod.itml/icOid/7343
It has a digital controlled temperature hot air gun, with full air flow display and adjustment, readily adapted nozzles (with many others also available) and auto shutoff and safety features. All for <$100.
Check out the very interesting retractable arm/heater holder on another model at http://www.circuitspecialists.com/prod.itml/icOid/9752
They have a conventional 1-piece 120 VAC unit at http://www.circuitspecialists.com/prod.itml/icOid/8010
These detached heater units like they could be very useful for vaporization. I realize that there are many digital hot air guns, but units with a lightweight hand-held heater and running at just 24 V DC appear attractive, particularly if the fan was quite.

Circuit Specialists also has cheap digitally controlled soldering irons, such as at http://www.circuitspecialists.com/prod.itml/icOid/7307 and http://www.circuitspecialists.com/prod.itml/icOid/9748
And a cheaper unit with analog control at http://pickels.powerhostingusa.com/prod.itml/icOid/9747
One unit at http://www.circuitspecialists.com/prod.itml/icOid/7760 claims temperature control within 3C (6F). On all or most the digital models, the thermometer can be recalibrated (e.g., to reflect heated air temperature).
To me, these soldering irons look cheap and functional enough (but probably not fast-reacting) that one could attach an air heater/exchanger to the tip and have the core of a digital vaporizer.

All or most of these units run on 24 V DC, with converters in the unit (no wall adapters). The company has its other related products listed at http://www.circuitspecialists.com/level.itml/icOid/6388 and http://www.blackjacksolderwerks.com/

Does anyone see any problems with using or adapting these for vaporization? Are there any other similar interesting products? [These are the cheapest versions of these products I've seen so far].

And since these integrated units appear to be cheaper than many, if not most, controller units on the market and also convert from 120 VAC to 24 V DC, might the controllers alone be a good deal? Might these or something similar be a platform for hobbyists like me (not an expert in electronics, not wanting to deal with 120 VAC or to have to wire or solder) to simply plug-and-play with digital vaporizer heaters?

 

[skippymcware]

Ninjavape,

Good thinking. I had thought of doing that idea except with displacing the water. My only gripe with that idea is that there will be some pressure in the container. I don't know if this will throw it off, though. I will give it a shot just to see.


Hennessy,

I would guess I have spent around $700 so far (including all of the tools). In my mind, this is worth it. I have learned so much already. I actually bought a bunch of PC case fans, a power supply, and a controller to do fan testing as well. I came up with some interesting stuff but have not had the time to write it up yet. I will get around to that once I need to incorporate it into the design.


vap,

Those look like good little unites. I would try and find out what the heating element is like. That is all that really counts. If it is a ceramic honeycomb, you are all set. As for actually using it, I couldn't say whether or not they will work for you. I was very excited about using the steinel heating element, but I had to give up on it because repackaging it was going to be a total pain. I think you may run into the same problems. Good luck.

-skippy