The Bud Toaster - (currently: Model 14, version 3)
- Thread starter Hippie Dickie
- Start date
okay, cool, carry on.
testing update ...
blew another 3.3v regulator - it appears i need to put in more heat shielding to protect the chips, or spec extreme environment components.
is really okay since i'm re-doing the physical structure anyway ... i've redesigned the PCBs to go back to a cylindrical enclosure. sometimes the first ideas are the best ideas. i abandoned this earlier approach when i switched to surface mount devices for the control electronics.
This design will push the PCB much further away from the heater coil and i can install several layers of foil and air as a thermal barrier.
It will probably end up as a cylinder 1 3/4" diameter by 3" high. Plus the external battery pack.
blew another 3.3v regulator - it appears i need to put in more heat shielding to protect the chips, or spec extreme environment components.
is really okay since i'm re-doing the physical structure anyway ... i've redesigned the PCBs to go back to a cylindrical enclosure. sometimes the first ideas are the best ideas. i abandoned this earlier approach when i switched to surface mount devices for the control electronics.
This design will push the PCB much further away from the heater coil and i can install several layers of foil and air as a thermal barrier.
It will probably end up as a cylinder 1 3/4" diameter by 3" high. Plus the external battery pack.
Bubar
Well-Known Member
It might be worth using a commercial pcb service. I haven't ever had to etch my own pcb, because it's pretty damn cheap to have someone else do it. I just prototype in the bread board and get a pcb if I want multiples.
I'm gonna go build your circuit tomorrow. I'm looking forward to some better temperature control. Mine is currently swinging 3-5 degreesC
I'm gonna go build your circuit tomorrow. I'm looking forward to some better temperature control. Mine is currently swinging 3-5 degreesC
i totally agree ... i am all over the commercial pcb services and i've downloaded several circuit layout software packages to do it. However, that assumes the geometry is determined but i am still moving stuff around to get the best physical form factor.
see, i live with my vaporizer right by my keyboard, which is where i live 18 hours a day, every day. i want this vape to be as small as physically possible.
The vape i'm using now has a pcb with the mosfet/diode on the same board as the PIC/vreg/MAX. In this next model i've cut that in half and have two 1"x1" boards to fit into the bottom of the cylinder.
i'm still trying to get to an actual pre-production prototype. So far, all these vapes have been one-off designs ... just not ready for fabrication.
see, i live with my vaporizer right by my keyboard, which is where i live 18 hours a day, every day. i want this vape to be as small as physically possible.
The vape i'm using now has a pcb with the mosfet/diode on the same board as the PIC/vreg/MAX. In this next model i've cut that in half and have two 1"x1" boards to fit into the bottom of the cylinder.
i'm still trying to get to an actual pre-production prototype. So far, all these vapes have been one-off designs ... just not ready for fabrication.
i don't understand your question: The coil is fed directly from the battery pack -- one end of the coil goes to the +battery, the source of the MOSFET goes to the -battery and the drain of the MOSFET completes the circuit to the coil.
The PIC and the MAX are connected to the voltage regulator output. The VReg can only put out 100 ma (max), but the coil wants 12 amps. Check the diagram:
PLEASE REFER TO POST #1 FOR THE UP-TO-DATE SCHEMATIC
i haven't bothered to install those three 330ohm resistors between the PIC and the MAX. i think i should - there is a potential failure mode that could lock up the PIC.
You can get batteries from A123Systems.com -- it's the 6 cell developer pack.
You can also get single cells from Hobby King and the charger:
http://www.hobbyking.com/hobbycity/store/uh_listCategoriesAndProducts.asp?idCategory=218
The charger seems to have disappeared from A123Systems websites. They just did an IPO so things might change for supporting developers ... maybe not.
EDIT: i just noticed that R8 should be 2k ohm. i'll fix the diagram some time and repost it.
The PIC and the MAX are connected to the voltage regulator output. The VReg can only put out 100 ma (max), but the coil wants 12 amps. Check the diagram:
PLEASE REFER TO POST #1 FOR THE UP-TO-DATE SCHEMATIC
i haven't bothered to install those three 330ohm resistors between the PIC and the MAX. i think i should - there is a potential failure mode that could lock up the PIC.
You can get batteries from A123Systems.com -- it's the 6 cell developer pack.
You can also get single cells from Hobby King and the charger:
http://www.hobbyking.com/hobbycity/store/uh_listCategoriesAndProducts.asp?idCategory=218
The charger seems to have disappeared from A123Systems websites. They just did an IPO so things might change for supporting developers ... maybe not.
EDIT: i just noticed that R8 should be 2k ohm. i'll fix the diagram some time and repost it.
Bubar
Well-Known Member
When I hook stuff up from the same power source it seems the heating element saps all the power, and the pic doesn't turn on. If your vape works , then this must not happen with your circuit. What about your circuit prevents the heating element from fucking up the pic? After thinking about it, it may be the diode you have across the heating coil. I have a .1uF and a 10uF capacitor across the supply lines, but do not have the diode or voltage regulator.
Oh and how many amphours are in the batteries you use.
Oh and how many amphours are in the batteries you use.
Here is the spec sheet (PDF): http://www.a123systems.com/cms/product/pdf/1/ANR26650M1A_Datasheet_APRIL_2009.pdf
Summary:
Nominal voltage: 3.3V
Nominal capacity: 2.3Ah
Core cell weight: 70 grams
Internal impedance: (1kHz AC) 8 m? typical
Typical fast charge current: 10A to 3.6V CCCV
70A continuous discharge
120A, 10 sec pulse discharge
Cycle life at 10C discharge, 100% DOD: over 1,000 cycles
The 3.3 voltage regulator i use must get at least 3.3v + 1.5v dropout, so say, about 5 vdc. Is your supply unregulated that the 12 amp draw drops the voltage below 5vdc?
What type of power supply are you using? Can it put out 12 amps?
Here is a 12vdc, 12amp power supply from Hobby King for $30US
http://www.hobbyking.com/hobbycity/...me=GTpower_SuperCon_AC_Adapter_100~240V_12.5A
But 12v is too much for the PIC and the MAX -- you simply must use a voltage regulator to get it down to 5v or 3.3v -- the problem with using a 5v regulator is the batteries will die too soon unless you get an extremely low dropout regulator.
Summary:
Nominal voltage: 3.3V
Nominal capacity: 2.3Ah
Core cell weight: 70 grams
Internal impedance: (1kHz AC) 8 m? typical
Typical fast charge current: 10A to 3.6V CCCV
70A continuous discharge
120A, 10 sec pulse discharge
Cycle life at 10C discharge, 100% DOD: over 1,000 cycles
The 3.3 voltage regulator i use must get at least 3.3v + 1.5v dropout, so say, about 5 vdc. Is your supply unregulated that the 12 amp draw drops the voltage below 5vdc?
What type of power supply are you using? Can it put out 12 amps?
Here is a 12vdc, 12amp power supply from Hobby King for $30US
http://www.hobbyking.com/hobbycity/...me=GTpower_SuperCon_AC_Adapter_100~240V_12.5A
But 12v is too much for the PIC and the MAX -- you simply must use a voltage regulator to get it down to 5v or 3.3v -- the problem with using a 5v regulator is the batteries will die too soon unless you get an extremely low dropout regulator.
How do you know the PIC isn't running? Is the BLUE LED on? Does the GREEN LED flash the 3 flashes every 5 seconds?
Bubar
Well-Known Member
What I meant was in my version. I know the pic doesn't turn on because it doesn't write to the lcd screen. I use a +12v line for the transistor and have the pic on a +5v line of the same computer power supply. But it freaks out and doesn't work. If I use a separate 5v wall adapter and the 12v line from the computer power supply it works.
I know your version works fine. That's why I'm asking what makes it work like that.
I know your version works fine. That's why I'm asking what makes it work like that.
There should be no current draw by the coil until the PIC turns on the MOSFET.
The PIC should work even if the coil is disconnected.
Also, are you sure the 5vdc line is active? Doesn't sound like it is.
Check your wiring.
The PIC should work even if the coil is disconnected.
Also, are you sure the 5vdc line is active? Doesn't sound like it is.
Check your wiring.
you have to hold BUTTON THREE (BTN3) for 2 seconds and the GREEN LED will come on and stay on as long as you continue to hold BUTTON THREE.
Then release BUTTON THREE (3) and the GREEN LED will go OFF and the RED LED will go ON.
The RED LED is ON when the pwm is activating the MOSFET.
If you do nothing when the power is applied, the GREEN LED will flash 3 times every 5 seconds.
Then release BUTTON THREE (3) and the GREEN LED will go OFF and the RED LED will go ON.
The RED LED is ON when the pwm is activating the MOSFET.
If you do nothing when the power is applied, the GREEN LED will flash 3 times every 5 seconds.
Bubar
Well-Known Member
Is there a reason it would just turn off? I turned it on, then walked away for a while. It was heating up before I left, and then when I got back it was not hot or pumping any juice. The Led was doing the 3 blinks in 5 seconds thing. Does it have watch dog or somehting. 2 times I've turned it on and wandered for about 10min. When i get back its totally cold and drawing no current.
Also side note: I used a irf640 as a mosfet. It doesn't work with the LED in there, because it clamps the voltage at 3.3V and the irf640 needs 5v. In case anyone else tries to build this, make sure your mosfet gate voltage is lower than your supply voltage... lol
Also side note: I used a irf640 as a mosfet. It doesn't work with the LED in there, because it clamps the voltage at 3.3V and the irf640 needs 5v. In case anyone else tries to build this, make sure your mosfet gate voltage is lower than your supply voltage... lol
Yes, there is a reason it turns off.
There is a maximum of 7 minutes run time ... then it automatically turns off, except for the 3 blinks every TWO seconds, which signals that the seven minutes session has been terminated.
3 blinks every FIVE seconds to signal "READY TO START"
3 blinks every TWO seconds to signal "END OF SESSION"
There are other blink patterns to indicate error shutdowns - missing sensor, too hot, battery near dead (test not implemented yet).
When running on batteries (or connected to a power supply that can deliver 12+Amps), it hits vape temperature range in 40 seconds. It is ready to toke continuously after 90 seconds. The temperature control algorithm maintains the SETPOINT temperature regardless of air flow (inhale rate).
Also, on power on, the Watchdog Timer is set to time out if the go signal (press BTN3 for 2 seconds) does not happen within 5 minutes.
RE: MOSFET - interesting, i hadn't even considered the gate voltage. as it turns out the Fairchild FDB6030BL i'm using allows 1 to 3 volts as gate voltage with +20v absolute max.
There is a maximum of 7 minutes run time ... then it automatically turns off, except for the 3 blinks every TWO seconds, which signals that the seven minutes session has been terminated.
3 blinks every FIVE seconds to signal "READY TO START"
3 blinks every TWO seconds to signal "END OF SESSION"
There are other blink patterns to indicate error shutdowns - missing sensor, too hot, battery near dead (test not implemented yet).
When running on batteries (or connected to a power supply that can deliver 12+Amps), it hits vape temperature range in 40 seconds. It is ready to toke continuously after 90 seconds. The temperature control algorithm maintains the SETPOINT temperature regardless of air flow (inhale rate).
Also, on power on, the Watchdog Timer is set to time out if the go signal (press BTN3 for 2 seconds) does not happen within 5 minutes.
RE: MOSFET - interesting, i hadn't even considered the gate voltage. as it turns out the Fairchild FDB6030BL i'm using allows 1 to 3 volts as gate voltage with +20v absolute max.
The duty cycle is set to start out at maximum (duty cycle value of 1023.=$FFFF) until the coil temperature passes the "Trip" point. At this temperature (around 400F), the duty cycle is set to ZERO until the coil temperature peaks - in the case of overshoot, this is the point where the set point error changes from increasing to decreasing.
This is when the temperature control heuristic takes over to determine the correct duty cycle value to maintain the set point temperature.
How long does it take to reach set point? 3.3A??? i think the power supply is limiting the output.
Although ... the duty cycle value for maintaining set point IS about 1/3 of the total range, so 3.3A sounds about right AFTER reaching setpoint.
And ... you're using twice the voltage of my battery pack. i've been meaning to plug into the car battery to see how much faster it can reach vape temperature. Could be bordering on "world's fastest vape" territory.
i write the MAX6675 reading (12-bit value) and the duty cycle value (10-bit value) in two ring buffers in the EEPROM so you can read the last 30 seconds or so of readings. Look at end of VARS.ASM for the layout of values in EEPROM.
This is when the temperature control heuristic takes over to determine the correct duty cycle value to maintain the set point temperature.
How long does it take to reach set point? 3.3A??? i think the power supply is limiting the output.
Although ... the duty cycle value for maintaining set point IS about 1/3 of the total range, so 3.3A sounds about right AFTER reaching setpoint.
And ... you're using twice the voltage of my battery pack. i've been meaning to plug into the car battery to see how much faster it can reach vape temperature. Could be bordering on "world's fastest vape" territory.
i write the MAX6675 reading (12-bit value) and the duty cycle value (10-bit value) in two ring buffers in the EEPROM so you can read the last 30 seconds or so of readings. Look at end of VARS.ASM for the layout of values in EEPROM.
AND, MOST IMPORTANT ... i hope you're using all glass vapor path. It ain't no Bud Toaster if it ain't all glass vapor path, you know.
Hennessy1414
Terrorist
...and this is why I follow every page of this thread. Even if I don't know half of the lingo being used.
keep the science coming, boys
keep the science coming, boys
Bubar
Well-Known Member
The power supply is not limiting anything. This is a nice PSU, it can supply 18A at 12v or 40A at 5v. I couldn't get it to work at home. But when I had it at school euning at 6v probing with an ampmeter and it only drew 3.3A, but it started at .3A and gradually increased. Idk wtf is going on. I ordered the transistor you used. I hope that will fix everything.
Btw. Old computers are an excellent source of power supplies.
Btw. Old computers are an excellent source of power supplies.
How long does it take to reach 430F/220C?
Also, there is an internal zener in that Fairchild MOSFET. So the external snubber diode is probably not needed. i don't know enough ee to figure out if it is needed or not - to be determined.
Removing it would shrink the pcb quite a bit.
Does your PIC "sing" when it's operating? My Bud Toaster emits a pitch that the gf and the cats can hear -- i seem to have lost that tone in my hearing range. i set the PWM frequency to maximum, assuming that would give me more control, but it may be overkill - to be determined.
There's a free program that figures out all the timer config values based on the clock speed. Do you have that?
Also, there is an internal zener in that Fairchild MOSFET. So the external snubber diode is probably not needed. i don't know enough ee to figure out if it is needed or not - to be determined.
Removing it would shrink the pcb quite a bit.
Does your PIC "sing" when it's operating? My Bud Toaster emits a pitch that the gf and the cats can hear -- i seem to have lost that tone in my hearing range. i set the PWM frequency to maximum, assuming that would give me more control, but it may be overkill - to be determined.
There's a free program that figures out all the timer config values based on the clock speed. Do you have that?
usr/share/vape
Well-Known Member
Well, with qualifications.Bubar said:
I recently found out (the hard way) that all my old cpu power supplies output 11.4VDC or less. Two of them were PC Power & Cooling and another was an Antec (the other three were China somethings). The China ones don't surprise me - but the others are quality brands. They're all at least 10 years old - too many miles I guess. I still use them for non-sensitive devices, but I have one widget that gets all pissed off if voltage drops to or below 11.4VDC (the hard way referenced above). YMMV.cheers
new thought ... what wire gauge are you using to supply power to the coil?Bubar said:
i recommend at least 16 gauge, and ideally, 12 gauge, wire to deliver enough current.
i see current limiting effect when using 18 gauge and smaller.
Bubar
Well-Known Member
The wire gauge is prolly it... lol The gator clips I was using were 22awg. I just don't have any 16 awg around. I think 18 awg is fine, since that's the size my power supply uses and it claims to be able to pump 40a @ 5v through them. But right now, unless I'm at school with a regulated 6v power supply I can't get the vape to go. If I short my gate to the + rail it works, but there is too much voltage loss from the pin of the pic to trigger the gate. Once the new parts arrive I'll be chillaxin. I haven't gotten around to taking a pic, but I really like my set up right now. I have a mason jar with a hole in the lid with the oven tube in it and holes in the side for the other connections. The power cords have been hacked up for the heating element to be insulated with teflon instead of plastic. I use ceramic fiber pads as thermal insulation.