I vote for the mobile app to cover most of the intelligence and the core to concentrate on the Smoking control 
hey @coolnewt do you have a fritzing of your initial prototype? id love to recreate, and build along. we can start helping out with some of the firmware development that wayā¦
1 Like
I smell some (really a bit or maybe imaginary) smoke coming out from your house 
no smoke till july 4!
except from the pizza ovenā¦
1 Like
what about generating power from the heat of the bbq? something like a peltier deviceā¦ one side heated from the BBQ and the other cooled with the air being drawn in from the fanā¦ ive seen a camping stove that uses it to charge mobile phones
not sure about patents and stuffā¦ but its awesome technology
2 Likes
This is really cool. I wish they sold the thermoelectric generator separately. It has USB output, so you could plug it into the core and it should just work. So far I've left the power on this project open ended. The reason for this is to allow a little flexibility based on application. If using it only for measuring probe temps then battery power would work, but as Avidan mentioned previously this will likely not work well if powering to fans/blowers. In that case wall power would likely work better and probably this thermoelectric generator as well.
You can get the high-temp Peltier devices on ebay. I donāt have any experience with power generation from Peltier devices but they are really useful for cooling the lab.
These are rated up to 225 C.
1 Like
I stopped by and visited them at Maker Faire last month. If they are able to generate enough electricity with a peltier to power a 40mm fan AND charge a cell phone battery off of a tiny cup of fire, surely you could power a Spark with it and maybe some smaller fans/blowers. Perhaps pair it with a battery-charging shield of sorts in case the temperature differential gets too low at points (Iām not familiar with the temperature curve, if any, of a BBQ smoker).
1 Like
More hardware updates: https://github.com/engirati/smart-bbq
Several updates polishing whatās there currently. Need to add the fan/blower output circuitry. Should be simple, but havenāt spent time on it yet. If anyone has experience in this area and wants to contribute that would be fantastic!
Still open ended on power as well. So far USB powered works 
Iām wiring VIN/5+ from USB to power the LED display. Is depending on the USB to regulate the power ok, a no go or just bad practice? Trying to minimize parts so if I donāt have to add another regulator that would be great. Itās working on the bench ok so far. It was at 3.3V though too despite being rated at 4.6 minimum.
1 Like
this seems like a good fan output circuitry to work with
1 Like
A complete schematic. Iām sure there is tweaking in store 
Iām thinking you could make that 2 PWM fan controller WAAY simpler using 2 N-channel logic level FETsā¦ hereās two in one tiny SMT package:
http://www.digikey.com/product-detail/en/SSM6N58NU,LF/SSM6N58NULFCT-ND/4304408 (15 cents in quantity)
At 500Hz PWM and such low gate charge you probably donāt even need series gate resistors, but a sprinkling of .05 cent resistors for series gate and pull down on gate is not going to take up much space or dollars.
You kind of need to spec in your Fans first thoughā¦ typical computer fans are not that bad on current, less than 0.5A @ 12Vdc. Not sure if a 120mm 90CFM fan is good enough though.
Here is a good fan example used for smokers.
http://www.auberins.com/index.php?main_page=product_info&products_id=400
1 Like
should handle that no problem 
@BDub thanks for the input. Most of the fans I looked at were around 0.5A @ 12V but some were as much as 1A @ 12V. If Iām reading the datasheet correctly, this would actually support up to 4A at 12V. Does this design look right? Iām trying to minimize the types of components used so trying to use either 330 or 1k for R1 depending on what looks right w/ my LEDs. I think either would work for R1 in this design. Is the cap necessary or would it actually interfere with my PWM signal? Thanks again! No making fun of fan drawing 
No, you donāt need the cap. The diode is probably ok to keep, and doesnāt have to be big honkinā 1N4001, but those are cheap and common. Just look around for a suitable cheap SMT diode in SMA package or SOD323 package.
Yeah I figured the fans might be UP to 1A, so I went with a 4A part for better power dissipation at 1A. I would probably not recommend running it at 4A.
330ohm would probably be more than enough for a dampening resistor. This FET has a very low gate charge and at 500Hz you arenāt going to be switching fast enough to start having issues with inductance. You can go nuts calculating a lot of this, but if you just start with a small resistor and look at the Vgs waveform with a scope, then increase the resistor youāll get a good idea of how much is too much. Itās important to have a low enough pull down resistor on there as well for the case where you turn power on and all of the Spark core outputs are hi impedance still while itās powering up. You can build in all kinds of safety for stuff like this as well, probably outside the scope of this project, but it might not be a bad idea to shove an appropriate Polyfuse in series with the two fans. Just guessingā¦ maybe one 2-3A polyfuse for the pair of fans. You really have to pay attention to the holding current, trip current, and how those derate with temperature vs. your expected operating temperature.
1 Like
@BDub thanks for the input. I thought I found a comparable pinned out SMT package, but it turned out to be an obsolete part. This one has such good specs, Iām going to give soldering this package a try.
Schematic updated, pcb to follow.
Looks good! I will suggest a few things though:
- Probe inputs looks like maybe 2.5 or 3.5mm jacks. Connect the TIP to the analog input, and the SLEEVE or RING to GND. That way ESD has a harder path into you Spark Core inputs.
- It might be a good idea to do a Low Pass RC filter with the inputs in a different way so that you have a resistor always in series with the analog input. A 1k and 1uF should be plenty fast for temperature sensors. Then you still have your 10k pull up and thermistor pulldown on the front end. Assuming the impedance of the ADC input is high, you wonāt get much of a voltage divider effect when taking a sample. The large 1uF cap will also help this along, effectively lowering your input impedance.
- Switch inputs could use two resistors each. I know you are using the internal pullup resistors, but ESD again can easily get to the Spark Core inputs from something a user is purposely touching all of the time (i.e. the switches). Minimum 1k series, 10k pull up. Put the pull up at the switch, not the Core, that way the switch can pull it all of the way to ground.
- I would add a diode in series right after the J6 output (input power), just in case anyone decides to plug in a reverse polarity wall adapter. Also a polyfuse in series with the diode.
- R18 should connect to FAN2, not FAN1 (copy/paste error )
- I would make the fan connectors whatever the fans normally come with to save costā¦ whatever you use make sure the connections are not exposed before or after connections are made, since you have live +12V and a switched node. Shorting GND to this by accident could cause some issues.