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

[Bubar]

LCD ain't that bad. Mine only takes 4 or 6 pins, I don't remember right now. I'm using an 18series pic anyway, so I have a ridiculous number of pins. Once I finalize stuff I'll put it on a smaller pic.

Also, what gauge nichrome are you using? I have been using 16awg, but I will switch to something finer, because 16awg is a bitch to bend into shape.

 

[stickstones]

Is anyone familiar with how to suggest this to the person or persons that manage FC?

noted.

 

[Hippie Dickie]

Bubar - i'm still using 16 ga Nichrome 80 (medical grade). Yes it is hard to form. i make jigs for stuff like that, and (after 35 years) i've equipped a nice little work shop. Sometimes i just wonder where all this shit came from.

i started out with 22 ga -- but now that has too much resistance per foot to allow the high currents needed for this super fast heat up to vape temp. i'm thinking about going to a nichrome ribbon to reduce the resistance even more. Anything is possible with a PWM and a finely tuned duty cycle algorithm -- i feel like a PWM grants virtually unlimited control -- together with a fast clock cycle.

How's your code coming? Do you have the Watch Dog Timer configured? That was about the first thing i got working - that and the 100 msec clock. i've programmed too many infinite loops in the 42 years i've been hacking code to not have a failsafe backup when i'm dealing with heater and batteries that can hit 800F so fast.

 

[vtac]

Maybe, it's time for a new main FC Forum (one that would permanently host multiple ongoing threads) to host these discussions. Perhaps, it could have a title along the lines of "Hobbyists Discussion," "Building Your Own Vaporizer," "Technical/Engineering Aspects," etc. Doing this would separate discussions of commercially marketed vaporizers, now in the "Vaporizer Discussion" forum, from discussions of those not yet or ever to be commercialized or available to others, separate out technical discussions of parts and designs, etc.

Is anyone familiar with how to suggest this to the person or persons that manage FC?

We certainly have had a lot of very productive, informative discussions is this and related threads, such as "Building a Digital Vaporizer (Brain Dump)." In my opinion, one of the best things that could be done by FC to advance the state-of-the-art and availability of a diversity of vaporizers would be to encourage those building and tinkering around with these on their own to share and discuss their work.

I am probably not the only one hesitant to start a new thread about experimental, non-available vaporizers within the mainstream Vaporizer Discussion forum that is dominated by discussions of marketed products. Giving hobbyists, developers and techies our own little corner could well draw out more who are now just lurkers, facilitate initiation of more threads (more topics), etc.

These vape building threads are uber awesome and if the contributing parties feel a new section would help, personally I'm all for it. I've been subscribed to the existing 2 threads since their creation.

 

[Hippie Dickie]

i don't know if a separate section is needed, but whatever management decides is fine with me ... as i learned from my time in CA in the early 70s, i just go with the flow (and i ain't no dead fish, either).

 

[Bubar]

My code has been done for several weeks. I coded it up in C so it was a breeze. I don't have a watch dog timer. I guess I could add one, but I haven't done it yet. My vape doesn't really overheat. If thermocouple connection breaks it turns off.

 

[niall]

Maybe, it's time for a new main FC Forum (one that would permanently host multiple ongoing threads) to host these discussions. Perhaps, it could have a title along the lines of "Hobbyists Discussion," "Building Your Own Vaporizer," "Technical/Engineering Aspects," etc. Doing this would separate discussions of commercially marketed vaporizers, now in the "Vaporizer Discussion" forum, from discussions of those not yet or ever to be commercialized or available to others, separate out technical discussions of parts and designs, etc.

Totally agree, and loving this thread and others like it. This site is a great resource.

 

[Qbit]

I've just read through this thread, HD, and I'm most impressed, but I wonder if the complexity of your device could mean for more things that can go wrong. And I wonder how complex electronics mixes with the heat. But then I don't know that much about it all.

 

[Hippie Dickie]

i agree it is more complex than an eterra or PD, but have you ever seen all the parts that go into a volcano? Reliability is a reachable goal. A vape that doesn't power on is just not acceptable. It better always work, or be a 2 minute repair job that anybody can do.

i've been testing the Bud Toaster concept from the bottom up -- from a very simple device to this computer controlled, temperature feedback version. And i think that is the key ... exhaustive testing ... excruciatingly comprehensive testing ... testing, testing, and then more testing.

Fortunately, i have a day job so there is no rush to get this vape to market.

So far, the points of failure are always the same (other than stonerly dropping of the glass -- usually the vial, not the stem, by the way) -- power delivery to the heater. It appears that high current flow eats copper. And these failures are due to the prototypiness of whatever current model i'm testing. Should be very easily fixed in production. Not rocket science, nothing to invent, just apply best practices. And have a warranty.

And, really, three chips is not a complex device these days. The improvement over all the past models is staggering ... just imagine ... cordless, blazing fast warm up, and accurate temperature control during the entire vape session. For $20 in parts. Plus batteries.

 

[MacRadish]

just imagine ... cordless, blazing fast warm up, and accurate temperature control during the entire vape session. For $20 in parts. Plus batteries.

Imagining.....

OK, I'm down for one.

 

[Qbit]

How many hits or how much time at heat could you expect to get out of it using typical batteries? (Did you already mention that earlier?)

 

[Hippie Dickie]

With the present configuration - wire gauges and printed circuit trace widths/copper thickness -- i'm getting 6 sessions of 7 minutes each. i get at least a dozen hits each session. The truth is i always lose count.

Depending on how much bud in the vial, i sometimes pull the plug at 6 minutes. This is usually enough time to work all day for me.

Then a 15 minute recharge.

i've been using just 2 sets of batteries to see if there are any longevity vs performance issues with these Li Phosphate cells. They are supposed to be good for more than 1000 charges and won't lose storage capacity, but the technology is too new to have good performance data yet.

These are the cells the new auto industry is focusing on. These fuckers - $20 each right now - will be dirt cheap in a year or two.

 

[Hippie Dickie]

This is a true bummer.

Image Shack has been hijacked.

i have deleted all my images there and modified this thread to remove the URL links to the images.

Fuckers.

 

[Hippie Dickie]

Here are the latest pictures:

This shows the top deck of the vape, and below that, the heater coil wrapped around the oven tube, and below that, the command and control circuit board:

The vape body is flipped over and the batteries are visible inside.

These pictures show the new improved coil design, where the middle of the coil is at the bottom of the oven tube.

Here is a close up of the control board, showing the coil mounting clips at the top center.

This is by far the best coil attachment scheme i have come up with in the 8 years of fucking around with this project. Solid, gas free connection, using pure copper conductor.

This is how the coil looks when attached to the control board (firewall removed for clarity):

So here is the coil with the firewall in place:

And a different angle:

 

[skippymcware]

I am still blown away that you managed to print your own board. Anyway, that thing is looking really good. All that work has definitely paid off. What are the dimensions on this latest version?

Keep it up and thanks for your help,
-skippy

 

[Hippie Dickie]

Printing boards is very easy. Radio Shack has a starter kit with double-sided copper clad, etchant, a resist pen (Sharpie), and resist remover (isopropol alcohol). Print the pattern on a laser printer -- pattern should be reversed like for printing a t-shirt -- and iron it onto the board. Then soak to remove the paper backing.

The outer dimensions are 4.25" long, by 2.5" wide, by 1.75" high.

 

[skippymcware]

hmm, i had no idea radioshack had that. I will have to check it out. Where is the thermocouple mounted or going to be mounted?

 

[Hippie Dickie]

i got the bare-wire thermocouple with .010" diameter leads. i will mount it in a 1/16" thick piece of mica and it will go between the firewall and the oven tube, between two loops of the coil.

i'll take some pictures when i get it mounted. It is vital to have good insulation -- i have shorted out the coil with the lead of the thermocouple a couple of times and it is not pretty when that much current flows through the short -- it melts the teflon pretty quickly.

 

[Hippie Dickie]

Alas, my prototype (m13 v6-2) has developed an intermittant problem and isn't working right now. So rather than try to debug what will eventually be decommissioned soon anyway, i'm pushing ahead with getting m13 v6-4 operational.

here is a progress shot:


i just need to attach the batteries and i can start testing this sucker.

 

[Hippie Dickie]

M13 v6-3 ... first toke on 07/22/09 at 6:30 pm

gosh this has been a thrash.

i think i burned up a MAX6675 by overheating during soldering. i do not like this silver solder, it just doesn't flow like with a little bit of lead in it. the solder joints always seem to be cold solder joints. i think it may be time to learn the hot air pen technique. i got the pieces-parts to do the McGyver hot air pen using the Radio Shack desoldering tool and an aquarium air pump, but i haven't assembled it yet. And i'm not sure the solder paste i got at Lowes will work properly. And may be i need some flux ... yet another exercise left to the reader.

And now the thermocouple readings seem to be intermittent ... getting a reading of coil temp equal to zero is not good.

i also think the copper clad i'm using is junk. i've got wide-ass traces and still i'm only getting 6F per second temp increase -- only half of what the prototype was getting -- i think the copper is restricting current flow - reminiscent of a model several interations ago, when i was using 22ga power cable and saw a vast improvement when switched to 16ga cable.

BUT! at least i have proved the circuit is good. i think it is time to get some boards made at a foundary with 2oz copperclad.

i think i can see a homemade reflow oven in my future ... where did i put that toaster oven?

 

[Hippie Dickie]

Whew! A nice resolution to a very, very frustrating day of debugging.

See, i was getting a zero reading from the thermocouple (as reported by the MAX6675 to the PIC 12F683). Now, if the thermocouple was disconnected (UNcoupled, as it were), the MAX sets a bit saying the thermocouple is missing (actually, an open circuit), and the code checks the bit and aborts the session when set.

But there was no program abort (i can tell because the red LED goes out when it should be on, and the green LED flashes a special pattern - 2 quick dits with a 2 second period - so that wasn't the problem.

Now, along the way to checking the code as a possible cause of the zero reading, i noticed the eePROM was getting cleared just as soon as the chip was programmed. So that HAD to be a coding problem.

So began an unfruitful 4 hours of code checking and cross-checking and getting nowhere.

Then, for no good reason it dawned on me to try a different PIC chip (i've been buying parts for 5 devices at a time -- hoping to built at least one functional vape and maybe a spare, cause, damn it all, the just broken prototype has proven, beyond a shadow of a doubt, the need to ALWAYS have a spare vape).

So i popped a spare PIC into the programmer, and the eePROM value stayed set the way it should be set. DUH! Like i say, i'm not a hardhead -- i'm a softie through and through (balls to bone, as the Oracle would say).

Put the correctly programmed PIC into the vape chassis, and then i was hot to go (so to speak).

Now, back to the problem of the zero reading from the thermocouple ...

If i held the thermocouple , jiggled it, wiggled the MAX, i could get a sporatic reading. So i disassembled the probe i had rigged up, uncoupled it from the MAX and fashioned a new pcb attachment. Still, the readings were erratic.

Using a continuity tester (for the 10th time) i sensed a weak PCB connection to one of the pins on the MAX - the one for the thermocouple + lead. Resoldered that pin, and the damn thing is working again.

This hassle got started during testing late last night -- the bare lead of the thermocouple must have touched the coil, created a path to ground, and immediately melted (the teflon insulation, too). Damn these batteries are intense.

And, now that i think about it, i'll just bet the eePROM on the PIC got fried at the same time. Damn good thing these parts are all so cheap.

So i finally got to take the afternoon toke at about 11:00pm. The batteries are re-charging now. It's getting late. Time to go to bed. Hopefully tomorrow i can make up some of the work i skipped to do this debugging.

But i had to get it done for a special visitor to the digital cave tomorrow and Saturday -- i just HAD to have a working vape i'll sleep well tonight.

 

[Hippie Dickie]

at work (finally!) ... and the first two vape sessions have worked perfectly. whew.

now i've got to squeeze it all back into the enclosure.

after i get some work done.

 

[Hippie Dickie]

here is the breadboard layout of the version 6-3 vape:



looks a mess, but this is merely a functional test. i changed a lot from the prototype (previous vape) that was the first vape to use my pulse width modulator design , so i just need to see that i haven't shot off the rails.

By the way, while working on the Bud Toaster, in all its manifestations, i have come to develop a great respect for the people (engineers? rather, probably artsy types) who do industrial design -- design that is way beyond the trivial making of a thing (shit, any McGyver can do that), but to think through the total lifecycle ... to make it easy to fabricate ... to envision better functionality ... and to make it pretty, for pity sake.

 

[Hippie Dickie]

(this is more like a diary than a discussion thread it seems)

plugged in the power yesterday and there was a flash and a puff of smoke ... i just hate it when that happens. And it doesn't happen due to bad design necessarily, just the flimsy nature of the prototype setup -- too much flexture and movement of the leads. However, perhaps it functions as an "accelerated testing to failure" procedure.

i think the failure this time, was that there was another short caused by the end of the bare thermocouple leads -- plus lead to ground ... fried the voltage regulator and evaporated part of the ground trace on the PCB. Which was a real hassle since the broken trace was underneath the socket for the PIC processor.

So i replaced the voltage regulator and soldered a jumper around the broken trace and voila! it's vape time again.

So this round of testing has mapped out some tasks:

i definitely need to relayout the pcb to move that ground trace.

more parts - couple of bad PICs, fried regulators, dropped resister chips -- damn they are so very small.

better battery connection -- i have been trying to avoid putting a Deans connector on each battery, but it sure looks like that is the best solution. Then each battery is removable, and rechargable, on its own. But that causes problems for the enclosure design -- viz. how to make a reliable hatch cover.

At least this prototype is vaping. That's a big help for design.

 

[Qbit]

Don't be too proud of this technological terror you've constructed. The ability to destroy a planet is insignificant next to the power of The Force.