Homeroasters - Home Roasting Coffee Community » DATALOGGERS - CONTROLLERS - RATE OF RISE METERS » Dataloggers/Controllers/Rate of Rise Meters
Who is here? 1 guest(s)
Page 9 of 31: 1...6789101112...31
 Print Thread
4-channel TC meter and datalogger project
seedlings
Top  #201 Print Post
Posted on 07/01/2010 10:19 AM
Awesome work, Bill. And, now you will have intellectual rites on the whole shebang.

CHAD
Roaster: CoffeeAir II 2# DIY air roaster
Grinder: Vintage Grindmaster 500
Brewers: Vintage Cory DCU DCL, Aeropress, Press, Osaka Titanium pourover
 
allenb
Top  #202 Print Post
Posted on 07/01/2010 5:21 PM
My board arrived today in perfect shape.

Thanks Jim for the great packaging!

I'm curious, will one board's worth of parts meet Mouser's or Digikey's minimum order requirement?

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
JimG
Top  #203 Print Post
Posted on 07/01/2010 7:01 PM
I think only the 0.1uF capacitors, which are pretty cheap, have a minimum quantity on Digikey. I haven't checked Mouser.

I just updated the BuildIt wiki to reflect buying in small quantities. Prices listed are mostly from Digikey, so be sure and check if you can get a better price at Mouser, etc.

Jim
 
Web
allenb
Top  #204 Print Post
Posted on 07/01/2010 9:19 PM
Newark has plenty of the MCP3424's at $3.91 (1 ea).
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
randytsuch
Top  #205 Print Post
Posted on 07/01/2010 11:55 PM

Quote

allenb wrote:

I'm curious, will one board's worth of parts meet Mouser's or Digikey's minimum order requirement?

Allen


Mouser doesn't have a min order.

Digikey charges $5 extra if you spend less then $25.

But neither has that part, you should be able to find everything at Newark though.

Randy
 
JimG
Top  #206 Print Post
Posted on 07/02/2010 8:26 AM
Update on the "rate of rise" part of the project.

Bill and I have split up some duties, and I took the task of writing code to do realtime rate of rise calculations (Bill took the hard stuff).

I've got something that's working now, and have posted 2 screencaps that show the differences between different amounts of smoothing.

The test setup consists of the TC-4 board with two identical bare bead type K thermocouples connected to channels 0 and 1, sitting on my desk next to an open window with random breezes.

In the screencaps, the blue/red traces are the logged temperatures from the TC's. The orange/green traces are the calculated rates of rise. (The rates are multiplied by 10X, so a plotted value of 40 means a rise rate of 4.0 degF per minute).

The differences between smoothing over 4 samples and over 10 samples are pretty dramatic, and I've decided the 10 sample version is the way to go. This is pending, of course, some testing in actual roasts.

Here are links to the screencaps:

http://tc4-shield...CapN04.jpg
http://tc4-shield...CapN10.jpg

I have also posted a zipped up version of all of the files needed for someone to try out my test/demo logger program on their roaster. Feedback on performance with different amounts of smoothing would be very welcome.

http://tc4-shield...100702.zip

This is offered without much documentation, except for comments in the code. At a minimum you would have to manually change the COM port variable "whichport" in the pRiser.pde file. If you want to play with different amounts of smoothing, there are four lines of code in aRiser.pde that would need to be edited. PM me for instructions.

You'd also have to put the library folders TypeK and Riser in your Arduino distribution libraries folder.

This might be enough to hold bleeding edge early adopters at bay until a "release" version of the logger is ready. smile

Jim
Edited by JimG on 07/02/2010 8:38 AM
 
Web
farmroast
Top  #207 Print Post
Posted on 07/02/2010 9:34 AM
Jim
Is it going to be possible to have a data list of RoR readings next to time? With a friends help during a roast we made a list with Time-ET-BT-RoR-V(variac voltage off a Kill-A-Watt). Readings were recorded every 20secs. It gave me the clearest capture of a roast to date for me. The RoR IMO takes away the need for a graph to show the relationship between Time and BT. Maybe I'm just graph challenged :| but side by side lists just seem clearer to me.
Bill I enjoyed chatting with you yesterday on the project :Clap: I think a valuable tool is being created here! Thanks to all!
Ed B.
DreamRoast 1kg roaster, Levers, Hand Mills http://coffee-roa...gspot.com/
 
milowebailey
Top  #208 Print Post
Posted on 07/02/2010 9:56 AM
I like the way this is moving.

Have you guys seen/used zedgraph? very cool graphing classes. I don't know if they would provide a better interface. I use that in my plotting utility. I don't know if you can add it to the Processor program...

http://zedgraph.o...=Main_Page

Here is a screenshot of what my interface looks like
milowebailey attached the following image:
widget.jpg
Edited by milowebailey on 07/02/2010 9:57 AM
 
allenb
Top  #209 Print Post
Posted on 07/02/2010 9:58 AM

Quote

farmroast wrote:
Maybe I'm just graph challenged :| but side by side lists just seem clearer to me.


I don't think your graph challenged. I use graphs at work which for certain applications are indispensable but I feel the same in that it's much easier for me to gain a very quick interpretation of a roast viewing a table versus any graph.

All I've ever gained with a roast graph is a vague idea of "S-curve" steepness but I can do without it.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
bvwelch
Top  #210 Print Post
Posted on 07/02/2010 10:07 AM
Ed, I'd like to make sure I understand what you're asking for. We may already have something that is very close.

First to explain: All of this can be changed -- it is 'just' software. :-)

In Jim example above, the data comes into the PC from the board, looking like this:

time, T0, RoR0, T1, RoR1

We could easily display that for you directly on-screen in real-time, and also save it to a CSV file that you can later review with a spreadsheet program like Excel. You don't have to look at a graph if it isn't helpful. :-)
 
bvwelch
Top  #211 Print Post
Posted on 07/02/2010 10:12 AM

Quote

milowebailey wrote:

Have you guys seen/used zedgraph? very cool graphing classes. I don't know if they would provide a better interface. I use that in my plotting utility. I don't know if you can add it to the Processor program...

Here is a screenshot of what my interface looks like


Nice, why don't we just forget this Processing stuff and go with your approach? It might drop in pretty easily. Unless your code is proprietary? Which is fine, just asking?
 
milowebailey
Top  #212 Print Post
Posted on 07/02/2010 10:25 AM

Quote

bvwelch wrote:

Quote

milowebailey wrote:

Have you guys seen/used zedgraph? very cool graphing classes. I don't know if they would provide a better interface. I use that in my plotting utility. I don't know if you can add it to the Processor program...

Here is a screenshot of what my interface looks like


Nice, why don't we just forget this Processing stuff and go with your approach? It might drop in pretty easily. Unless your code is proprietary? Which is fine, just asking?


Nope, nothing I do outside of work is proprietary. Life is too short to be selfish.

I'm a C# hack and can send you all the code. I used two open source classes and Visual C# express to compile. Happy to hand that off.... you seem way more capable in the SW department than me. <grin>
 
JimG
Top  #213 Print Post
Posted on 07/02/2010 11:26 AM

Quote

bvwelch wrote:We could easily display that for you directly on-screen in real-time, and also save it to a CSV file that you can later review with a spreadsheet program like Excel. You don't have to look at a graph if it isn't helpful. :-)

Actually, through absolutely no fault of mine, it already does this.

In the little window below the code in the Processing IDE, the time,temp1, RoR1, temp2, RoR 2 values scroll by every second. Not very elegant, but it provides the info in real time. You can drag the window to enlarge it to see several lines of data at once.

Also, I left intact Bill's original code that logs everything to a .CSV file.

For a "real" application, all of this needs to be cleaned up. But for proof of concept trials, the present code already provides numerical information.

Jim

PS - What's up with all you "non-visual" learners? Shock
 
Web
allenb
Top  #214 Print Post
Posted on 07/02/2010 4:16 PM

Quote

randytsuch wrote:

Quote

allenb wrote:

I'm curious, will one board's worth of parts meet Mouser's or Digikey's minimum order requirement?

Allen


Mouser doesn't have a min order.

Digikey charges $5 extra if you spend less then $25.

But neither has that part, you should be able to find everything at Newark though.

Randy


I just placed my order with Newark.

The only item they didn't have is the MCP9800. They offered to ship me one from the UK for a $20.00 service fee but I declined. I'll see if my local surplus shop has this item.

I ordered the Arduino from Sparkfun. Seems to be a user friendly site.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
bvwelch
Top  #215 Print Post
Posted on 07/02/2010 5:28 PM
Allen,

I don't think Mouser has a minimum order, and they are willing to ship USPS (cheap). But I have a few extra of those, send me your postal address via a PM and I'll send you one. Be careful tho - if they jump off the tweezers they are just about impossible to track down. :-)
 
allenb
Top  #216 Print Post
Posted on 07/03/2010 9:17 AM
Thanks Bill.

I'll take you up on the offer and I'll make it up to you with a trade of something.

It took me a while to get the hang of Newark's search/selection tool. If you don't start with the most basic descriptor columns and click find or search each time, it gets a little confused.

A call to their customer help line squared me away.

I know what you mean with a chip flying out of the tweezers. I was trying to move an Atmel T2117 zero-cross chip to a board (only had two) and had to go on the hunt a few times. I'll tie a safety line to it till it's docked.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
randytsuch
Top  #217 Print Post
Posted on 07/03/2010 12:40 PM
So, I'm bummed.

My digikey shipment is delayed until Tues, and my Arduino (which I ordered on Tues), is not here yet, I'm hoping it gets here today, they shipped USPS.

I ordered the Arduino from an outfit about an hour away from me, but it shipped from Nevada (I hate when that happens) on Thurs.

Looks like I won't be able to do much on this project over the long weekend.

I started working on a software installation guide yesterday, hopefully it will help others that do this after me, but I can't really work on it anymore until I get some real HW :(.

Randy
 
JimG
Top  #218 Print Post
Posted on 07/03/2010 11:25 PM
I took what I learned using the roast logger yesterday, and made some much needed revisions to the Bourbon release.

A new zip file has been posted on the Googlecode site:

http://tc4-shield...100703.zip

Significant changes:

1. We are only tracking RoR on channel 1 now (presumably you will use this channel for BT). Screen is too busy if we track RoR on ET also. (This can be put back in if someone wants it.)
2. Added on-screen real time alphanumeric displays of elapsed time, chan. 1 temperature, chan. 1 RoR, and chan. 2 temperature.
3. Added ability to capture JPG image(s) of screen with mouse click anywhere in window.
4. Changed names of sketches to aBourbon.pde and pBourbon.pde.
5. Automatically generated log files (CSV) and any JPG's now get put into a LOGS folder.

Jim
 
Web
JimG
Top  #219 Print Post
Posted on 07/03/2010 11:30 PM

Quote

randytsuch wrote:
I started working on a software installation guide yesterday, hopefully it will help others that do this after me, but I can't really work on it anymore until I get some real HW :(.

Randy -

Thanks. That will be a big help for people.

Bill configured the zip file for the first Bourbon release with what should be the correct folder structure for standard Arduino and Processing sketch book locations. I maintained the same format with the release I just posted.

Sorry to hear that your stuff didn't arrive in time for the weekend.

Jim
 
Web
farmroast
Top  #220 Print Post
Posted on 07/04/2010 9:25 AM

Quote

JimG wrote:
I took what I learned using the roast logger yesterday, and made some much needed revisions to the Bourbon release.

A new zip file has been posted on the Googlecode site:

http://tc4-shield...100703.zip

Significant changes:

1. We are only tracking RoR on channel 1 now (presumably you will use this channel for BT). Screen is too busy if we track RoR on ET also. (This can be put back in if someone wants it.)
2. Added on-screen real time alphanumeric displays of elapsed time, chan. 1 temperature, chan. 1 RoR, and chan. 2 temperature.
3. Added ability to capture JPG image(s) of screen with mouse click anywhere in window.
4. Changed names of sketches to aBourbon.pde and pBourbon.pde.
5. Automatically generated log files (CSV) and any JPG's now get put into a LOGS folder.

Jim

Sweet, I too really don't see much value in ET RoR and it does keep the clutter down.
Ed B.
DreamRoast 1kg roaster, Levers, Hand Mills http://coffee-roa...gspot.com/
 
randytsuch
Top  #221 Print Post
Posted on 07/04/2010 10:34 AM

Quote

JimG wrote:

Quote

randytsuch wrote:
I started working on a software installation guide yesterday, hopefully it will help others that do this after me, but I can't really work on it anymore until I get some real HW :(.

Randy -

Thanks. That will be a big help for people.

Bill configured the zip file for the first Bourbon release with what should be the correct folder structure for standard Arduino and Processing sketch book locations. I maintained the same format with the release I just posted.

Sorry to hear that your stuff didn't arrive in time for the weekend.

Jim


USPS came through, and I got my Arduino, along with a LCD, so at least I got to play with that. I hooked up the Arduino, and made the LED blink.

I made a "C:\program files\arduino-0018" directory where I installed all the arduino stuff.
I unzipped the riser stuff into a "C:\program files\arduino-0018\Riser20100702" subdirectory.

I would do the same with the new program, and update my instructions accordingly.

I can redo my directory structure, to match what you guys are doing.

Randy
 
JimG
Top  #222 Print Post
Posted on 07/04/2010 11:27 AM

Quote

randytsuch wrote:
I can redo my directory structure, to match what you guys are doing.

Here's the way my computer is set up:

1. arduino-0018 is in Program Files, just like your's. This doesn't really matter very much, as it turns out. There is not currently any need for customization of the Arduino distribution.

2. In My Documents, there is a folder named Arduino. This is what they call your sketchbook. In the My Documents/Arduino folder are individual folders containing sketches. The name of the folder must be the same as the name of the sketch file (*.pde) it contains. For instance, the My Documents/Arduino/aBourbon folder should contain aBourbon.pde.

3. Also inside the My Documents/Arduino folder is a libraries folder. The IDE is set up to look here (My Documents/Arduino/libraries) for contributed libraries. Each library (TypeK, Riser, etc) gets its own folder under libraries. The *.cpp and *.h files reside inside the folder (TypeK.h is in the TypeK folder, e.g.).

4. Repeat steps 2 and 3, but this time for Processing instead of Arduino. Same folder structure. (None of our stuff so far, however, uses any contributed Processing libraries).

5. I've adopted a sketch naming convention of the distribution name (Bourbon, e.g.) preceded by a lower case a or p, depending on whether it is an Arduino sketch (to be flashed to the chip) or a Processing sketch (to run on the host PC).

Bill set up the zip file for the Bourbon release to match the above. I think you can just unzip the file under My Documents and everything will find its way to the right place?

Jim

PS -- I'm using Windows. The Linux and Apple setups are a little different, but I think the idea is the same.
 
Web
JimG
Top  #223 Print Post
Posted on 07/04/2010 11:34 AM
Screen capture from current Bourbon release (post on the project googlecode site):

tc4-shield.googlecode.com/files/roast201007041035-826.jpg

Roasted today, in Hottop. El Salvador Bourbon, 237g.

See a short discussion of a funny thing with the RoR for this roast on H-B.com thread:
http://www.home-b...ml#p175554

Jim
Edited by JimG on 07/04/2010 12:33 PM
 
Web
farmroast
Top  #224 Print Post
Posted on 07/04/2010 12:57 PM
The RoR really shows fine details of a roasts reactions. This could produce more identifiers of when a reaction is happening and quantify how the beans are reacting to the ET profile during a roast.
Bill, here is my present controls setup BT and ET on the back left dual display datalogger, BT RoR on back right datalogger, countup timer and kill-a-watt meter on the variac
farmroast attached the following image:
dreamroastcontrols.jpg
Edited by farmroast on 07/04/2010 3:31 PM
Ed B.
DreamRoast 1kg roaster, Levers, Hand Mills http://coffee-roa...gspot.com/
 
allenb
Top  #225 Print Post
Posted on 07/04/2010 3:58 PM

Quote

JimG wrote:

Roasted today, in Hottop. El Salvador Bourbon, 237g.

See a short discussion of a funny thing with the RoR for this roast on H-B.com thread:
http://www.home-b...ml#p175554
Jim


I've recently switched to radiant only on my 1lb drum since I've encountered problems with my under-drum heating element.

After the switch to radiant only, I noticed exactly the same thing. At around 400F (BT) when 1C starts to get rolling my BT RoR slows then almost stops in it's tracks for a couple of seconds then after a few seconds resumes the original RoR all while giving off a healthy set of snaps. While this is happening my PID controller has had to go to 100% output to try and bring the ramp segment under control.

I've read many articles covering the endo/exotermic reactions and many of them contradict each other so I've never had a real good understanding of when they occur.

Your theory of vapor release sounds plausible.

I'd like to hear more from someone who has a good grasp of this.

BTW, looks like great progress with the sheild project data logging!

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Page 9 of 31: 1...6789101112...31
Jump to Forum: