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

[Hippie Dickie]

pushing ahead on the next 3 beta units ... here are a batch of batteries:

just need to Dip-It them to give them a plastic coating.

Well ... i've got 2 more batteries in the box, i'll attach the Deans connector to them, too, then dunk them all in Dip-It to plastic coat them.

i was having a problem locating a charger for these LiFePo4 (A123Systems) batteries ... i was using a $50 charger (now out of stock) that needed 12vdc input, and a $15 AC to 12vdc supply. However! today i located a $11 charger that does the whole shebang. i'm very excited! i've got to order a dozen of these puppies stat!.


... baby steps

 

[Hippie Dickie]

step 2 ... wrapped all the batteries with electrical tape before the Dip-It ...

Funny thing while soldering the Deans connectors to the battery tabs -- my 25 watt soldering iron is dead, i.e. just can't get hot enough anymore. i plugged a 30watt iron sitting in a tool box and finished the batteries lickity split -- had to go back and resolder the batteries i had already done and now they're done right.


... pitter-patter, pitter-patter ...

 

[Hippie Dickie]

working on procurement issues ...

The first battery charger i got for these A123Systems batteries was $100, and required a 12vDC input ... well, over the years i have a box full of vape specific AC to DC supplies, so that was covered.

Then i found a $50 charger, and 12vDC supply for a total of $75.

Now, a year or so later, i can get an AC wall wart that supplies the CCCV (constant current, constant voltage) output required for LiFePO4 batteries for $10. Not as much current (2A vs 8A ... so it takes 60 minutes to charge instead of 15 minutes), but for that price, a very reasonable tradeoff. And, for that price, two will be included in the basic Bud Toaster package, along with 4 batteries.

 

[Hippie Dickie]

and about the case ...

This one is a whole lot harder to come to grips with ... i kinda know what will be right, but i sure can't find it. i am hoping when i get the laser engraver/cutter i can do some interesting dovetail joints and make a wood box that just snaps together ... uh, you can see i'm talking out my bud on this one -- at least until i get some hands on with the laser and some 1/4" cherry.

Anyway, i've been taken by the Pelican line of indestructible cases, and thought i had one selected:

The outside dimensions are 8" by 5" by 3", and when i misread the interior dimensions i thought 4 batteries (on end) would fit with no problem ... but i didn't understand the reduced dimensions due to the foam lining and when i stand them up, the lid can't close.

Oh well, everything needed for a day's sessions (4 to 8) does fit okay:

Still needs a little stash box, but that is doable in this space.

 

[Hippie Dickie]

oh, now i'm REALLY excited ... i just got an acceptable quote for a custom made borosilicate oven tube -- 15 mm OD x 38 mm Length, with 1.2 mm wall thickness -- so i can say goodbye to cutting down 100 mm tubes and the wasted glass. Plus, the airgap between the oven tube and the bud vial is smaller so the heat transfer should be more efficient. i really hope this works out as expected.

 

[vap999]

In the VapeXhale Cloud track you recently asked about aerogels.

Aerogel blocks or pieces, such that you could hold, for other than the most cost-insensitive applications, such as defense/aerospace, are somewhat over-hyped, with similar silica-based microporous insulation (such as from MicroTherm) offering advantages over aerogel. The larger and most consistent single-piece aerogel blocks or pieces, what you see in pictures, the best stuff, essentially all goes to NASA or is simply too expensive to actually use.

But contrary to the hype, unmodified aerogel may not be suited for some applications involving achieving a high temperature differential, such as insulating vaporizer heating elements on one side and a handling surface on the other. Because aerogel is translucent, only diffuses light, it does little to stop infrared/radiated heat. This is why many aerogel applications are customized and involve packaging it in vacuum-packed aluminum foil/reflective plastic. Or for applications such as insulating a vaporizer, carbon black or titanium white pigment can simply be mixed in as an opacifier. But keep in mind that silica microporous insulation is as good or better insulation than aerogel for many applications because it's basic design integrates stopping radiant heat.

Granular aerogel, notably Nanogel from Cabot, seems to be the most cost-effective aerogel (as these things go), with gallons (231 cubic inches) sold by Solar Components for ~$60 with U.S. shipping. It is also hydrophobic (getting damp will not destroy it, like some other aerogels), flows freely (like liquid), and can be compressed/crunched into tight spaces and still retain full insulation. Just shake some high-quality (small particle size) artist pigment in before you use it for vaporizer applications, and you should have as good insulation as there is (along with microporous insulation), better than still air, but not as good as vacuum. Nanogel might be particularly useful with the Bud Toaster, since you could simply pour it in and reliably fill all the small spaces/voids before you seal it up. Another reason to use Nanogel from Solar Components is that it is the only way you can effectively get any either aerogel or microporous insulation without a $300-500 or higher minimum purchase (and having to deal with salesmen who want to know all about your business and applications).

In the vaporizer prototypes I've been making (all for myself so far), I have recently gone to using (simply pouring in and compressing) granular Nanogel as a back-up or secondary insulation to fill spaces around the ends of shaped microporous board material, the primary insulation.

Otherwise, probably better insulators than vermiculite, which you mentioned, but nowhere near as good as aerogel or microporous insulation, you might check out castable (mix powder with water and hardens) ceramic foam, such as from Rescor 740 from Cotronics, or high-quality silicone foam. For applications such as the Bud Toaster where worst case scenarios, such as a short, I presume could be incredibly energetic (enough to melt some metals?), forget about silicone which is only effectively stable up to <500?F (can't really be use to insulate heating elements).

 

[Hippie Dickie]

Thank-you for that information about aerogel and nanogel ... this will take several days to digest. Solar Components in particular has some stuff suitable for an alternative energy project i'm starting to work on.

For the Bud Toaster, so far the air insulation seems to be adequate. i'm somewhat hesitant right now to seal it up ... my fantasy is for the user to be able to do parts replacement (i.e. outsource the repair task to the customer). However, a sealed unit would be much safer.

 

[vap999]

For the Bud Toaster, so far the air insulation seems to be adequate. i'm somewhat hesitant right now to seal it up ... my fantasy is for the user to be able to do parts replacement (i.e. outsource the repair task to the customer). However, a sealed unit would be much safer.

At the power levels you are working with, having worked in the safety division of a major consumer products company, the level of risk (and risk vs. benefit) of encouraging consumer repairs is too high by conventional standards. If you presume that only 10% will totally mess things up, or even just 0.1%, that is too much considering the potential consequences (including getting seriously burnt, starting a fire, etc.).

If I were you, I'd seal it up very securely and permanently, even this means considering sealing it up and potting electronics with epoxy (presuming parts don't get anywhere near its temp. limit). Presuming that your production costs are eventually fairly low, why not go the much-admired Magic Flight-type warrantee route and give free replacements (perhaps with a picture of smashed Bud Toaster or return of a key part, such as a circuit board, required)?

 

[Hippie Dickie]

Wow. Thank-you for offering your opinion. It is very much appreciated.

i'm getting impatient to get to the commerical product ... i solve one problem and then i act as if it's all a done deal. And perhaps i have drawn wrong conclusions from the other commercial vape designs. i really need to just slow down and keep working on the design until i get it right.

i can definitely pot the board, since the components do not heat up at all (now, after the MOSFET fix). And production costs are insignificant. i've even considered including a spare unit in the basic package so an individual unit failure doesn't stop the vapor.

And, definitely i will provide a no questions asked replacement warranty ... my goal is to sell a functioning vaporizer, not create a repair hassle for the customer. When i buy something i expect it to work. my customers can expect the same.

thanks again for your comments ... they have been very valuable input to me.

 

[Hippie Dickie]

i thought i was going to pull an all-nighter last night, but i finally managed to get to bed at 4AM (instead of my usual midnight) ...

In my post #580 (6/23) i mentioned that i had located a glass company in India that can make my oven tube to my custom specifications. This is amazing -- i feel like i'm one of the big boys, being able to outsource a custom component to a shop in India (thank-you Google search) -- (i was going to say sweat shop, but i'm sure there is equipment to manufacture culture tubes (rimless test tubes, also know as labware or laboratory glassware) that it's not done by human glass blowers.

About the oven tube. Currently, it is a 16mm (about 5/8") diameter by 1 1/2" long, that is cut down from a standard culture tube that is 16mm x 100mm long (4") -- i throw away the other 2 1/2" of the tube. just a pitiful waste.

Also, culture tubes come in different wall thickness - 0.8mm, 1.2mm, 1.6mm - is the range i've seen. However, when i ask an american tube maker "what is your wall thickness?" the answer is, "we won't tell you that, it's proprietary!". WTF? Considering all the time i've wasted online trying to nail down the parameters of american made culture tubes, i could have flown to India (or China) and it would have been cheaper.

That's the past, this is now.

So, since 6/23 there have been an endless series of emails to India and back. Each response takes 2 fucking days (there is a 10.5 hr time difference - they're living in my future). Alas, buying from India is not like one-click checkout on Amazon.

Finally last night, i just stay up late enough to have near real-time email conversation. 6 days for what should have been a one-click checkout.

Outsourcing is tough, but doable.

Next i'll see if they can drill glass. That would be sweet.

Oh yeah, so i'm getting 1000 oven tubes, each one is 15mm diameter, with a 1.2mm wall thickness, and 38mm long (1 1/2"). This is probably my first real step toward making a commercial vaporizer. Cost is $0.11 each USD (or $0.20 each if you include shipping charge). versus $0.43 each for a too big tube i have to cut down and anneal the end. No brainer, eh?

Not only cheaper, better ... the current oven tube i'm using is the 0.8mm wall thickness. And it does crack - whether from microfractures caused by the initial scoring/snapping process to cut them down to 1 1/2", or the nichrome wire crushing the tube over many heat/cool cycles, or knocking the vape, i don't know. Thicker wall glass handles thermal stress better.

And i've been wanting to reduce the gap between the oven tube and the bud vial to improve heat transfer from the oven tube wall to the bud vial ... and improve heat transfer to the incoming air stream before it hits the bud.

It's a kick to buy something in rupees.

 

[andrewburgess]

i'm somewhat hesitant right now to seal it up ... my fantasy is for the user to be able to do parts replacement (i.e. outsource the repair task to the customer). However, a sealed unit would be much safer.

At the power levels you are working with, having worked in the safety division of a major consumer products company, the level of risk (and risk vs. benefit) of encouraging consumer repairs is too high by conventional standards. If you presume that only 10% will totally mess things up, or even just 0.1%, that is too much considering the potential consequences (including getting seriously burnt, starting a fire, etc.).

i wonder if offering it in kit form would help with that? then the user could decide whether to seal it at the end. i'd sure prefer to buy that way and i'd certify that i was competent to assemble and repair it in any way you want (hopefully a checkbox on the order page, easy for you and me)

offering kits is one of the advantages of doing business with a small company.

besides this thread is practically an assembly manual as it is right now. i wouldn't need more

 

[vap999]

i'm somewhat hesitant right now to seal it up ... my fantasy is for the user to be able to do parts replacement (i.e. outsource the repair task to the customer). However, a sealed unit would be much safer.

At the power levels you are working with, having worked in the safety division of a major consumer products company, the level of risk (and risk vs. benefit) of encouraging consumer repairs is too high by conventional standards. If you presume that only 10% will totally mess things up, or even just 0.1%, that is too much considering the potential consequences (including getting seriously burnt, starting a fire, etc.).

i wonder if offering it in kit form would help with that? then the user could decide whether to seal it at the end. i'd sure prefer to buy that way and i'd certify that i was competent to assemble and repair it in any way you want (hopefully a checkbox on the order page, easy for you and me)

offering kits is one of the advantages of doing business with a small company.

But the problem with selling something as a kit remains that no matter what you sell, you can't ignore common sense in your safety assessments; you can't sell or do something you reasonably should know is hazardous or foolish. In this case, just consider the basic demographics -- vaporizer users, most actually using one while working on their Bud Toaster --and ask yourself whether 100% will follow instructions (if they read them) or otherwise avoid causing a short circuit.

If the circuitry is high-power and the device is going, particularly being sold, to someone you don't know personally, seal it up.

 

[Hippie Dickie]

Currently, i'm thinking a kit is not where i want to go ... as andrewburgess says, the thread is an assembly manual. i'll do a new thread for the production Bud Toaster to clean up some loose ends, add a complete parts list, and eliminate the "drama" ... lol, "drama" as if (i'm kind of low key, usually -- i do have some hot buttons) ...

Ok, this damn project is dragging me down the path ... so now i'm waiting for a quote for the two drilled pieces (they said they can do the drilling), yeah, i just need to wait for my present to arrive at the future when they got my rfq email. There's a real strong sense of the tail wagging the dog on this step, but, really, it has to be done ... doesn't hurt to ask while i've got them on the line. Enough glassware for 300 Bud Toasters. Hard to wrap my mind around that thought.

Also, i did a lot of reading last night about the "3.3" borosilicate glass that cradles the vaporflow -- pretty great stuff. Glass rules!

 

[Hippie Dickie]

Okay, like i'm not heart-broken or anything ... still it would have been nice ...

See, it turns out that the cost of a drilled tube is $0.50 more than the glass cost -- i.e. $0.60 each in 1,000 quantity. Now from experience, i can hand-drill one tube per minute, so 60 per hour, or $30 per hour to drill my tubes. Definitely not done by laser. Drat ... i don't mind spending machine fabrication dollars in India, but i do mind spending manual labor dollars -- maybe that's a distinction without a difference. But anyway, i think i can laser-up my workshop to automate the drilling.

whew ... at least i'm saving some upfront money today. i don't see a downside from deferring the expense ... too many other details to work out first.

 

[Hippie Dickie]

in for a penny, in for a Euro ...

i did accept a quote for these two glass pieces, but NOT drilled -- basically $0.10 a piece for a 1000 piece minimum of each. There are plenty of shops around that can do laser drilling ... at least i'll have the glass.

Total shipping per piece will be less with this larger order, too ... see, i'm actually saving money by spending money.

Drilling the glass is such a critical manufacturing step. The Bud Toaster is the only vaporizer with an all-glass vapor path that uses a glass screen instead of a metal screen. (Vapolution has a "bump" in the glass to hold the bud in place.) i think it makes a BIG difference to have a totally all-glass vapor path -- no plastic, no metal, no ceramic, no fan blades, no silicon diaphram -- and, really, it's not that hard to do.

The other critical step is cutting that damn 1 3/8" diameter, 1/4" deep groove in the wood discs. Then today i finally discovered that end mills come in 1/16" diameter - that will work just fine. So a simple CNC machine for a couple grand and i can just knock these units out.

Well, there are a few other details to work out ... as always ... like tracking down a digital thermometer factory, maybe get a custom casting for the size and shape of the case ... i would never have guessed that so many people barbeque, but it's a market that has created exactly the right kind of digital thermometer with a range up to 300C ... but the wrong size and shape for my need.

baby steps ...

 

[Hippie Dickie]

ch..ch..ch..ch..ch..chargers ...

The first LiFePO4 charger i got was direct from A123Systems, "only" cost $100, and only accepted 12vDC as input -- it was expected to be connected to a car battery by R/C folks for their cars and planes.

i still have some AC to 12vDC power supplies (and plastic boxes) left over from earlier Bud Toaster model designs (before i switched to 5vDC as a power source). That's the black box underneath the brightly colored charger unit on top:

However, this charger is no longer available ... i did find an althernative unit, also made by A123Systems for "only" $50, and a commercial AC/12vDC/6A power supply for $15, so a total of $65 ... much better.

Then, recently, i found this charger:

A typical wall-wart that provides a 2A charging current - universal AC input from 120VAC to 240VAC, 50/60 Hz -- just needs a different shape plug to work in EU. Only costs me $10. But takes 4 times longer to charge a battery -- 1 hour instead of 15 minutes. A reasonable trade-off ... i'll put 2 of these in the package and still be able to drop the price to $400.

 

[Hippie Dickie]

delayed gratification ...

and, of course, India goes on a nationwide strike over increased petrol prices the day i placed my order.
India, where the common laborer makes $3 USD per day.

gosh, being born in America is a totally unearned gift ... thanks to whoever was responsible for that.

 

[Hippie Dickie]

outsource update and sw update ...

the custom sized glassware is enroute, actually shipping a few days earlier than the 15 days specified in the order.

i think i have finally corrected the implemention of the Bubar startup modification ... that is, the PIC starts applying current slowly and measures the temperature rise to keep it at about 10F per second. This should finally allow any DC power supply from 6vDC to 20vDC without getting hot so fast there is massive temperature overshoot and the PIC detects maximum allowed temperature and shuts off.

Also, sometimes i overload the bud vial and there are several minutes of vapor left to enjoy ... so after a session it can be extended for 2 more minutes at a time without going through all the heuristics of the startup sequence. This is being extensively tested and appears to work real good!

 

[andrewburgess]

i've got to get a better understanding of PID before i can attain the temperature accuracy that is obviously possible.

This http://forums.labjack.com/index.php?s=&showtopic=4167&view=findpost&p=12992 is the clearest description of a method of PID tuning I've seen (I haven't tried it).

 

[Hippie Dickie]

Thanks for that link ... i understand the basics of PID theory, but i don't understand how to apply that explanation to my situation. And i would say there is a GREAT amount of hysteresis in this "plant", so that tuning advice needs to be adjusted somehow, someway for the Bud Toaster.

And, actually, i am really happy right now with my modified PID algorithm. The only tuning i think it needs is to reduce the negative gains just a little bit -- it overcompensates when the heater is above SETPOINT, which causes a little oscilation, until the BIAS value is back at the proper point. The perfect BIAS value is probably a transcendental, which is hard (i find) to approximate with integer math.

what i still can't wrap my mind around is, why is the PID so general purpose (i.e. trial and error gain adjustment) when the thermal response can be accurately measure and the equation totally specified -- thus providing a precise value for the BIAS needed to hold any given SETPOINT.

But thanks for the post -- i like to revisit the fundamental issues to see if i can push the rock a little farther up the hill.

.... flash ....

My custom glassware is now at the sort facility in Brussels. India to Brussels is 5 days by air. WTF?

 

[villekille]

ch..ch..ch..ch..ch..chargers ...

The first LiFePO4 charger i got was direct from A123Systems, "only" cost $100, and only accepted 12vDC as input -- it was expected to be connected to a car battery by R/C folks for their cars and planes.

i still have some AC to 12vDC power supplies (and plastic boxes) left over from earlier Bud Toaster model designs (before i switched to 5vDC as a power source). That's the black box underneath the brightly colored charger unit on top:

https://www.icmag.com/ic/picture.php?albumid=13768&pictureid=423844

However, this charger is no longer available ... i did find an althernative unit, also made by A123Systems for "only" $50, and a commercial AC/12vDC/6A power supply for $15, so a total of $65 ... much better.

Then, recently, i found this charger:

https://www.icmag.com/ic/picture.php?albumid=13768&pictureid=423845

A typical wall-wart that provides a 2A charging current - universal AC input from 120VAC to 240VAC, 50/60 Hz -- just needs a different shape plug to work in EU. Only costs me $10. But takes 4 times longer to charge a battery -- 1 hour instead of 15 minutes. A reasonable trade-off ... i'll put 2 of these in the package and still be able to drop the price to $400.

I have been designing and making prototypes of my portable vaporizer for a little more than a year now. Today I found this thread and I was interested to some similarities, like using a123 -batteries, but on the other hand, my vaporizer uses completely different technology to warm the herb and electronics also different, and I use the AVR microcontroller.

There are some tricks when charging these, and I see that you have that wall wart for 2 amps. It just looks a little similar to one I tried, bought from Ebay, cheap 1a and 2a wall warts, I tried the 1 amp version and it didn't have any current control, so actually there was nothing limiting the current when I tried to charge the batteries, and the wall wart was able to supply about 1.45 amps for few minutes until the whole wall wart exploded into pieces and started to burn So I am interested to hear about your wall wart, is it current limited and what voltage rated? I have tried different approaches, like building microcontroller controlled switching charger, linear charger, and so on, but a wall wart would of course be easiest solution! So your batteries are charged one at a time, not in series..that is of course helpful, no problems with balance, even though these batteries seem to keep balance just fine anyway.

 

[Hippie Dickie]

When i first got the A123 batteries i set up a series module with balancing cable and after about a dozen charges the balancing cable broke where it came out of the charger. At that point i decided the batteries should be individual units for better long-term reliability.

i haven't tested the output of the wall wart charger - but LiFePO4 compatible is SUPPOSED to mean that it is CCCV (constant current, constant voltage). Since there is a single chip solution for this (Microchip has one) i am just assuming it is designed and working properly. Your experience is an eye-opener ... guess i better cut one open and see how it's made to be sure.

How do you monitor and control heater temperature?

 

[villekille]

OK, I just have not found such wall warts, if they are marketed to be lifepo compatible! I know the Microchip single chip solution, MCP73113, it is a linear LiFePo4 -charger with pass on transistor but the chip is annoingly small and difficult to handle in prototyping, as it has its contacts under the chip and is only 3x3 mm or was it 4x4 mm in size and you have to have proper vias under it and be able to solder the heatpad under the chip to conduct all heat away, as it can dissipate quite a lot of heat when charging empty batteries..and it has a maximum charge current of 1.1 A.

So far I have not found other single chip options for LiFePo4, but Texas Instruments has a nice easy chip meant for LiPo but it can be adapted for Lifepo just by changing 2 resistors in the circuit, and it is in SOIC-8 package and uses external PNP transistor so heat management easier.

I think we have a little different approaches since in my vaporizer, the 2 LiFePo4-cells are internal to the product, and only thing the end user sees is a charging jack at the backpanel, so they have to be connected in series because otherwise charging and using the product would not be possible at the same time, since my heater element requires 12 volts to work and if I would have the batteries in parallel, the losses from converting 3.3v -> 12v would be too great.

I am controlling the heater temperature with PWM, but I don't use any temperature control or monitoring. I have found out that the device works just fine without any specific control, so no need to make it more complicated. You have much higher currents in your circuit, which enables the element to get hot quicker, mine takes about 2 minutes to get to the right temp but after that it works well without any control, I have so called BOOST button which the user presses when inhaling, so the heater gets maximum power then, to prevent cooling down too much.

There is many times more thermal mass in my vaporizer, but still I get about 1 hour of vaporizing time, but I think your vaporizer gets more done in same time as warm up is quicker so not that much of the running time is wasted on warming up!

 

[villekille]

And more thermal mass means less need for quick adjustments of temperature. My design is not as pure glass on glass because of many reasons. In my design the hot heating element is fully covered with glass but I find that heat transfer is better with metal, and its thermal mass makes temperature more stable..and of course durability for years is guaranteed..so I have used a metal tube and inside it is the heating element and air passes between this hot glass element and the metal tube and I find it work fine - as the hottest temps are only seen with this glass covered heating element, and the metal tube warms only to a maximum of about 140 degrees celcius, there is not much to worry about any metal fumes..and on the other hand, while the temperature of this glass covered heating element can be changed quickly by giving it more PWM, the metal tube gives nice and stable "basic warmth" for the incoming air I don't need to think too much of the temperature, it just seems to work fine and vaporize the material I don't know if there is reason to fear metal too much, after all the melting points for various metals are very high

 

[villekille]

On the other hand, I have to admit that when I put a metal screen made from thin brass on the heating element, which warms up to about 350+ degrees celcius, there is some smoke forming and the colour of the mesh changes..so it is true that in hot temperatures metals react. But when I use a metal screen made of thicker steel, and not let it touch the heating element, it seems to be perfectly fine for the purpose, no colour changes, no smoke..after all, some 200+ degrees celcius is nothing for steel