roaster control widget

[mp]

Hey Milo ... I am curious ... have you considered the AD494CQ ... it has +- 1 degree accuracy vs the AD595AQ has +- 3 degree?
/quote]
Good point... I should have looked at the datasheet closer.... I ordered the AD595AQ's... I can always change to the AD595CQ's later... same pinout.

I wonder if the AD494CQ is code compatible with the AD595AQ?

 

[milowebailey]

Hey Milo ... I am curious ... have you considered the AD494CQ ... it has +- 1 degree accuracy vs the AD595AQ has +- 3 degree?
/quote]
Good point... I should have looked at the datasheet closer.... I ordered the AD595AQ's... I can always change to the AD595CQ's later... same pinout.


I wonder if the AD494CQ is code compatible with the AD595AQ?

 

Are you sure about that part #.... there is a AD594CQ... a AD594AQ, a AD595CQ and a AD595AQ... the AQ's are +/-3%... the CQ's are +/- 1%

The 594's are for J type thermocouples the 595 are for K type.

[mp]

Hey Milo ... I am just going by what this article says:

"The trick with these thermocouples is that you need a thermocouple amplifier to translate their output into a voltage that is linearly correlated to temperature. This makes it very easy to interface to the Arduino. Enter the AD595. The AD595AQ has a +-3 degree accuracy and is available at Sparkfun ($17.95). The AD494CQ has +- 1 degree accuracy and is available from Digikey. Digikey also has both versions in ROHS (lead free package) versions for 10 - 20 bucks more each if you would like. These are 14 pin DIP (will fit in a breadboard) packages. Even though they are 14 pins, you can breadboard one without a lot of effort."

Check out the Instructables web site.

Hope that helps.

[milowebailey]

Hey Milo ... I am just going by what this article says:

"The trick with these thermocouples is that you need a thermocouple amplifier to translate their output into a voltage that is linearly correlated to temperature. This makes it very easy to interface to the Arduino. Enter the AD595. The AD595AQ has a +-3 degree accuracy and is available at Sparkfun ($17.95). The AD494CQ has +- 1 degree accuracy and is available from Digikey. Digikey also has both versions in ROHS (lead free package) versions for 10 - 20 bucks more each if you would like. These are 14 pin DIP (will fit in a breadboard) packages. Even though they are 14 pins, you can breadboard one without a lot of effort."

Check out the Instructables web site.

Hope that helps.

I think that's a typo... or an obsolete part...

[MMW]

Thanks for this...I don't think I'm going to reconfigure my hottop for control with one of these (though the possibilities abound) but can you say multi-input USB temperature logger for ~$60?

The inner-tinkerer (? is that even a word) is itching to get out.

[milowebailey]

Thanks for this...I don't think I'm going to reconfigure my hottop for control with one of these (though the possibilities abound) but can you say multi-input USB temperature logger for ~$60?

The inner-tinkerer (? is that even a word) is itching to get out.

$60 would be for controller and the first thermocouple.... each additional is about $25 more... those little buggers are not cheap.

[MMW]

Yeah...from my (modest) perusal of digikey it looks like the dang amps cost as much as the thermocouples.  Still pretty dang cheap, though.  Still, it's an interesting project and I'll be following along.  Please keep us updated!

[mp]

I think that's a typo... or an obsolete part...

That may very well be.

 

[milowebailey]

I got the Arduno today... installed the drivers and in 5 minutes had a little program running on it...  They said it was easy.... I never thunk it would really be easy.

This is gonna be a blast.  I also ordered an LCD display so I can display bean temp... and who knows what else.

I'm very interested in what folks think the variables in the profile should be....

[J.Jirehs Roaster]

I'm very interested in what folks think the variables in the profile should be....

I don't have a sonafresca that I can tinker with but my current goal is to be able to plug in a number of different electric heat sources (one at a time) for a number of roasters and roasting experiments, I see the potential for fan speed control but most of my current configurations would not use the fan speed and I believe with the sonafresca you would not want to mess with fan speed...

My frustration is not getting ramp hold to work on the PID and my online research has left me to believe the PID is pretty limited for coffee roasting and that something like your milowwidget would give us more repeatability roasting platform to roasting platform, bean origin to bean origin ....

[mp]

I got the Arduno today... installed the drivers and in 5 minutes had a little program running on it...  They said it was easy.... I never thunk it would really be easy.

This is gonna be a blast.  I also ordered an LCD display so I can display bean temp... and who knows what else.

I'm very interested in what folks think the variables in the profile should be....

Hey Milo ... great to hear you got the Arduino.  Since it was so easy to do that much in so little time then you should be able to complete this project in 2 or 3 working days. 

Yeah ... the LCD sounds like a great visual aid to what is going on.

Wouldn't it be cool if the Arduino could sense when the first crack of the roast comes to an end?  You could work on a super cool automated green coffee profile that automatically adjusts the temperature depending on the type of roast you want to achieve.  Maybe even if it had the smarts to detect the beginning of the 1st crack for getting a light roast.  If it could detect the end of the first crack you would get a slightly darker light roast.  Likewise you could detect the beginning of the second crack which would mean you beans would be darker again or the end of the second crack which would put your beans to a fairly dark roast.  It might be called the universal bean profile ... or in this case the MiloVersal profile.  Just thinking out loud here.

The LCD would be great to provide the total roast time, the current temperature, the stage length.  If you can turn off the heat and only leave the fan on perhaps a message saying "cooling" would come on.

There are just so many possibilities.

 

[J.Jirehs Roaster]

Wouldn't it be cool if the Arduino could sense when the first crack of the roast comes to an end?  You could work on a super cool automated green coffee profile that automatically adjusts the temperature depending on the type of roast you want to achieve.  Maybe even if it had the smarts to detect the beginning of the 1st crack for getting a light roast.  If it could detect the end of the first crack you would get a slightly darker light roast.  Likewise you could detect the beginning of the second crack which would mean you beans would be darker again or the end of the second crack which would put your beans to a fairly dark roast.  It might be called the universal bean profile ... or in this case the MiloVersal profile.  Just thinking out loud here.

that would be cool...
...and at the end of second crack it could automatically release the halon to put out the fire



~~off to search for tiny heat resistant microphones~~

[milowebailey]

Hey Milo ... great to hear you got the Arduino.  Since it was so easy to do that much in so little time then you should be able to complete this project in 2 or 3 working days. 

You sound like my boss now.

Wouldn't it be cool if the Arduino could sense when the first crack of the roast comes to an end?  You could work on a super cool automated green coffee profile that automatically adjusts the temperature depending on the type of roast you want to achieve.  Maybe even if it had the smarts to detect the beginning of the 1st crack for getting a light roast.  If it could detect the end of the first crack you would get a slightly darker light roast.  Likewise you could detect the beginning of the second crack which would mean you beans would be darker again or the end of the second crack which would put your beans to a fairly dark roast.  It might be called the universal bean profile ... or in this case the MiloVersal profile.  Just thinking out loud here.

I've thought about this... if there was a way to capture the sound with a microphone... I think you could characterize the sound 1st crack makes... 2nd would be a little more difficult, but not impossible.

[MMW]

I've thought about this... if there was a way to capture the sound with a microphone... I think you could characterize the sound 1st crack makes... 2nd would be a little more difficult, but not impossible.

I dunno, it might not be that difficult at all.  The roaster is going to have a characteristic (moving air, rotating drum, etc.) that should be pretty easy to profile.  If the PC can keep up, a little blind LMS magic might even be in order.  Now all you are looking for are the cracks...should be pretty broad frequency, short time events.  A short time Fourier Transform or wavelet transform should be able to keep up in real time and visualize the snaps (they'd show up as thin lines across frequency in the display) and you'd need that for the detection anyway.  From there, the math to detect the cracks are pretty straightforward and I'd wager that a good bit of the code can be found open source/public domain that could be adapted to this particular job.

I think a more interesting application might be an uber-PID that could follow a set roast profile while controlling airflow and temp.  I have a burning desire to have a roast profile that is piecewise linear when plotted as bean temp vs. log time-- one slope for drying, another for ramp to FC, and another for ramp to SC.  (Don't ask why, you stick a EE in the oil field and you get crazy stuff like this).

[peter]

Hey Milo ... great to hear you got the Arduino.  Since it was so easy to do that much in so little time then you should be able to complete this project in 2 or 3 working days. 

You sound like my boss now.

I have a hard time believing Mrs. Milo would do you like that.   

I've thought about this... if there was a way to capture the sound with a microphone... I think you could characterize the sound 1st crack makes... 2nd would be a little more difficult, but not impossible.

I thought that's why God gave us ears, to hear coffee cracking...