Results 1 to 10 of 10

Thread: Android bluetooth controlled roaster?

  1. #1
    Junior Member
    Join Date
    Dec 2006
    Posts
    3

    Android bluetooth controlled roaster?

    Gene Cafe Coffee Roaster $850 - Free Beans Free Freight
    Hi all,

    Im a long-time lurker and first time (I think) poster here. So fristly, Gday and thanks for a years of highly informative* reading. My question relates to PID type roast control: Im thinking of getting an Android phone and wondering whether or not if would be possible/viableto use the phone as a roast monitor/controller, Given that:
    [list bull-blackball][*][/list]Anyone can develop and distribute Android apps
    [list bull-blackball][*][/list]
    There are Input/Output controllers such as those made by SparkFun which enable the phones to control peripheral devices
    [list bull-blackball][*][/list]A thermocouple and an SSR would presumably(?) work with such an IO device.

    I probably dont have the smarts to develop this but was wondering in anyone has heard of anything along these lines or has thoughts on whether this might be viable.[/list][/list]

    Cheers, Glen

  2. #2
    Senior Member
    Join Date
    May 2006
    Posts
    158

    Re: Android bluetooth controlled roaster?

    yeah, totally doable. buy this: http://www.sparkfun.com/products/10748

    write some code, do some wiring. victory.

    issues? youll be wiring 240vac so unless youre an electrician this is kind of illegal (i wont tell anyone though). also, youre writing code that will control 240vac circuits including heating elements and such so any minor bugs in the code could be highly destructive. no idea how the sparkfun board will accept a thermocouple input - not sure how the analog inputs work so you may need some sort of inline transmitter to turn the thermocouple junction into a 0-5V analog signal. also, i only have experience with 1 roaster (a friend has a gas has garanti) and im not sure how it would take to PID control - the time constant involved may be too large for PID control to be practical over the course of a ~15m +/- roast. what do you do if someone calls while youre in the middle of a roast? using an android over say just a simple arduino or dedicated PID controller introduces quite a few interesting problems.

    personally, id be inclined to take a few variables out of the equation - if youre seeking more control over your roasts then perhaps limiting your control system to a purpose built PID module and an SSR would greatly reduce the amount of problems you may encounter and the amount of trouble shooting required. if however youre trying to develop custom roast profiles or something and you are comfortable with the risks involved with using a your code in a high level device to control a large, expensive, potentially dangerous piece of equipment then the android may be a very good tool to achieve this.

  3. #3
    Junior Member
    Join Date
    Nov 2011
    Posts
    4

    Re: Android bluetooth controlled roaster?

    Quote Originally Posted by 505447594C6A585A475B5C5B526A350 link=1329918147/1#1 date=1329948411
    yeah, totally doable. buy this: -link removed so I can post it-

    write some code, do some wiring. victory.

    issues? youll be wiring 240vac so unless youre an electrician this is kind of illegal (i wont tell anyone though). also, youre writing code that will control 240vac circuits including heating elements and such so any minor bugs in the code could be highly destructive. no idea how the sparkfun board will accept a thermocouple input - not sure how the analog inputs work so you may need some sort of inline transmitter to turn the thermocouple junction into a 0-5V analog signal. also, i only have experience with 1 roaster (a friend has a gas has garanti) and im not sure how it would take to PID control - the time constant involved may be too large for PID control to be practical over the course of a ~15m +/- roast. what do you do if someone calls while youre in the middle of a roast? using an android over say just a simple arduino or dedicated PID controller introduces quite a few interesting problems.

    personally, id be inclined to take a few variables out of the equation - if youre seeking more control over your roasts then perhaps limiting your control system to a purpose built PID module and an SSR would greatly reduce the amount of problems you may encounter and the amount of trouble shooting required. if however youre trying to develop custom roast profiles or something and you are comfortable with the risks involved with using a your code in a high level device to control a large, expensive, potentially dangerous piece of equipment then the android may be a very good tool to achieve this.
    Id rather use a commercial PID and use the Android to control the profile of the PID over time. Perhaps using one of the cheap Android pads, then you dont sacrifice a phone.

  4. #4
    Junior Member
    Join Date
    Dec 2006
    Posts
    3

    Re: Android bluetooth controlled roaster?

    Quote Originally Posted by 5B53455D5F360 link=1329918147/2#2 date=1330359457
    Id rather use a commercial PID and use the Android to control the profile of the PID over time.* Perhaps using one of the cheap Android pads, then you dont sacrifice a phone.
    My understanding of commerical PIDs is that theres little if any ability to store multiple roast profiles. Editiing profiles is cumbersome and data logging generally needs to be done via a seperate device (eg beanbay multimeter). With a Sparkfun Android breakout board, you could connect your phone/tablet via USB/Bluetooth to the roaster as required and otherwise use it as you normally would. Main advantages over commercial PID units and PICs would be:

    1) Control, record and store your roast info from a single handheld device with a nice touchscreen GUI.
    2) No programming hardware required a la PICs.
    3) Minimal equipment outlay if you already have an Android phone. (SSR, thermocouple, Sparkfun board)
    3) Open source software platform (easily share programs and roast profiles between users.

    Im currently looking into the feasability side of it but my programming skills are pretty rusty!

  5. #5
    KJM
    KJM is offline
    Senior Member
    Join Date
    May 2007
    Posts
    688

    Re: Android bluetooth controlled roaster?

    magicpossum - a COTS Ramp/Soak PID typically has space for storing about half a dozen profiles... Can be had for about $75 too ;)

    The Google IOIO board will need some work done for a thermocouple input - you need to do the buffer/amp and cold junction compensation. You can get an I2C interfaced chip to do it. The costs start to mount up though... The phone nicely solves the user interface part as you noted though - and that is the tough and pricey bit.

    A fun project. The other device you might like to consider to use as the control brain would be the RaspberryPi. It should be available from tomorrow :o

    /Kevin

  6. #6
    Senior Member
    Join Date
    Dec 2010
    Location
    Toowoomba QLD
    Posts
    417

    Re: Android bluetooth controlled roaster?

    Hi Glen,

    Dunno about Anroid stuff sorry, but Im working on a similar idea using an FT232RL (also available from sparkfun) via USB with a laptop. Im using the SPI (part of the FT232RL) with a MAX6675 thermocouple amplifier and am trying to get round to working on the heater control - probably just use a triac based circuit for that. All pretty cheap so far about $50 all up.

    Pete

  7. #7
    Senior Member
    Join Date
    Oct 2011
    Posts
    1,376

    Re: Android bluetooth controlled roaster?

    A few choices here. http://developer.android.com/guide/topics/usb/adk.html
    This concept has come up once or twice before. Would love to see someone do it.


  8. #8
    KJM
    KJM is offline
    Senior Member
    Join Date
    May 2007
    Posts
    688

    Re: Android bluetooth controlled roaster?

    Quote Originally Posted by 72475647111B220 link=1329918147/5#5 date=1330421530
    All pretty cheap so far about $50 all up.
    I went another route - I just bought a TC4 board. This has 4 thermocouple inputs and a bunch of useful stuff. The current rev of the board has pretty much everything you need in the one place. Just needs a microcontroller to make it work - it is designed in Arduino shield format so it is easy to interface to.

    I have a bunch of Software Engineering final year students writing code for it so it can run off a RaspberryPi or an Arduino platform ;) . Excellent project for the group of students and something useful falls out the end ::).

    /Kevin

  9. #9
    Junior Member
    Join Date
    Dec 2006
    Posts
    3

    Re: Android bluetooth controlled roaster?

    Thanks for all the responses, people. Ive learnt a lot in a few days. Ive also put some feelers out on other boards to see if anyones gone down this path yet. Kevin, the RasberryPi project looks interesting - back to the future of computing?

    Glen.

  10. #10
    KJM
    KJM is offline
    Senior Member
    Join Date
    May 2007
    Posts
    688

    Re: Android bluetooth controlled roaster?

    Behmor Coffee Roaster
    Quote Originally Posted by 4D41474943504F5353554D200 link=1329918147/8#8 date=1330659096
    Thanks for all the responses, people. Ive learnt a lot in a few days. Ive also put some feelers out on other boards to see if anyones gone down this path yet. Kevin, the RasberryPi project looks interesting - back to the future of computing?
    Not really - the microcontrollers of today are just getting more complex. Controlling something like your average roaster could be trivially done on a PIC16 series - a 16F877 for example. But modern software engineers tend to want to use tools that cant target such limited devices (the 877 has a whole 268 bytes of R/W storage and a whole 8K instruction store).

    The thing you need to watch is how you deal with the Analogue inputs. Be aware that the need for cold junction compensation is something you cant really dodge and you need some kind of amplification. The Seeback for a K-type thermocouple is 41uV/C - if you have a microcontroller running off a +5V rail with 10bits of resolution (the 16F877 PIC for example) you can resolve only 5mV steps at best. To get to micro volts, you need 1000x the resolution which is a 20+bit ADC. Those are NOT cheap! The simple solution is to use one of the purpose built devices for K-type thermocouples. Last time I tried to source some - I could only get surface mount ones. Lovely stuff, but I cant solder SMD :o Not anymore!

    So there are some hard rows to hoe in this. The TC4 board was cheaper than I could buy 4x of the thermocouple interface chips - and it had them soldered and checked! So it was a no-brainer for me.

    /Kevin



Similar Threads

  1. Coffee roaster android app
    By 10ett in forum Home Roasting - Tips, Tricks, Ideas
    Replies: 16
    Last Post: 2nd April 2012, 08:30 PM
  2. new controlled roaster
    By Brekel in forum Roasters
    Replies: 18
    Last Post: 12th November 2006, 08:59 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •