Who is here? 1 guest(s)
 Print Thread
4-channel TC meter and datalogger project
allenb
From "Home Roasting" category on Home-Barista.com

Any way to have a meter display of the degrees rise/ min. rate?

Post # 7.

This is version 3 of the analog design.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
randytsuch
I actually have a kit of the HB temp rise circuit to build, have not gotten around to it yet.

It seems pretty simple, except it requires a calibration step, which the designer volunteered to perform, after I assemble it.

The thing with this circuit is that it is designed to connect to the temp meter, so you will still need to buy a temp meter, if you don't have one.

It was a little over $40 for the board and the parts (kitted by Jim, the designer). Add around $20 for a meter, depending on what meter. The board was designed to run off a little battery, from what I remember.

Compared to an Arduino, cost about $20-30 for the board itself, depending on which one you get. If this is standalone, you need a display, around $10 for an LCD.

Needs power, but you can power from a USB, if this will always be connected to a PC, otherwise can be battery powered also.

The Arduino solution will be more flexible and allow data logging. There are data logging meters, but they are more expensive.


Randy
 
allenb

Quote

randytsuch wrote:

The Arduino solution will be more flexible and allow data logging. There are data logging meters, but they are more expensive.


Randy


Please explain how one would use the Arduino setup for rate of rise if using it stand-alone. Could it just be switched on with a TC plugged into it and let it rip?

It's sounding like this would be a much more practical approach especially if a need for data logging came along. Cost looks reasonable as well.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
bvwelch

Quote

allenb wrote:
Please explain how one would use the Arduino setup for rate of rise if using it stand-alone. Could it just be switched on with a TC plugged into it and let it rip?
Allen


Yes, the stand-alone mode is simple to use-- hook up your TCs, turn it on, and start roasting.
 
bvwelch

Quote

allenb wrote:
From "Home Roasting" category on Home-Barista.com
Any way to have a meter display of the degrees rise/ min. rate?
Post # 7. This is version 3 of the analog design.
Allen


Neat project. If I understand it correctly, the setup would require: 1) an existing TC meter, 2) a VOM to display the 1 millivolt / degree F rise-o-meter output, 3) some means to split the TC or two probes co-located.

In our project, the Arduino provides both the TC meter and rise-o-meter functions, using just one TC, no splitter requred. All of the information (temperature and rise-o-meter) is displayed on the LCD. We can support up to 4 TC probes/channels per ADC board. The only extra cost for more channels is the connectors for the TCs.

We can also log the data to the PC.
Edited by bvwelch on 05/26/2010 12:25 AM
 
allenb
I hadn't read the thread closely enough to understand why the need for two TC's.

What I see in the setup you are putting together is an elegant approach with the most bang for the buck.

I look forward to seeing more.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
bvwelch
I tried out the data-logging mode today. It worked well. I'll take some photos, but meanwhile here is a screenshot and also the CSV file of the roast.

bvwelch.com/roast2010/roast_2010_5_27_18_48_55.jpg

http://bvwelch.co..._48_55.csv
 
bvwelch
Here's a photo of the prototype board. We plan to get a PCB made soon.

bvwelch.com/roast2010/100_1419.jpg

Next time - the standalone version with LCD.
Edited by bvwelch on 05/27/2010 8:55 PM
 
allenb
Is the reason for the PCB to be able to replicate the arduino onto the board along with the ADC or is the PCB for other purposes?

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
bvwelch
Hi Allen,

Since there are so many different Arduino boards out there, I think we probably want to keep the ADC board separate from the Arduino itself.

But I was planning a PCB for the ADC board, to reduce the time spent soldering the board. But really it isn't so bad, if someone wants to just use the Radio Shack proto-board like I did, that is fine too.
 
bvwelch
A follow-up-- do you consider the idea of a PCB a plus or a minus?

And also, we wanted to, the PCB could have a 'socket' for the arduino, for example the 'Pro Mini' is quite small:

http://www.sparkf...ts_id=9220
 
allenb

Quote

bvwelch wrote:
A follow-up-- do you consider the idea of a PCB a plus or a minus?

And also, we wanted to, the PCB could have a 'socket' for the arduino, for example the 'Pro Mini' is quite small:

http://www.sparkf...ts_id=9220


I think it depends on how each of us will be utilizing the completed unit. for example, I've decided to not use it as a hand held but to instead mount the whole works within my roaster's control box so I can read the rate-o-rise lcd right next to my PID controller which will limit clutter on my bench (which is way too much at present). I'll mount a panel mount TC female jack to the side of the control box chassis.

With this in mind, if the PCB makes it easier and less cumbersome to mount the works into my control box then I'm all for the PCB!

Hope this helps,

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Randy G
Looks like I sneezed and missed a lot here.. Where was I? Good question. I have been working on some profiles for the Hottop "P" and posted a page on the Hottop USA website for that model, but now back to the work at hand.

Looks like this project is well on its way. I like the graph. If that was gathered in real time, I think we are well on the way with this thing! That along with the data in the CSV file collected during the roast is a wonderful start.

Looking at it, I would build it into the roasted so that I would 0nly have to install a USB connection on the outside of the roaster. I have already installed two thermocouples in the Hottop.

Will it run on 12v DC?


Life's too short to drink bad coffee.
 
bvwelch
Hi Randy G,

Thank you for your interest. Yes, it is a real-time graph. Right now it is very basic, until I get some feedback from you all.

Yes, it can run from 12 volts, or it could get its power from USB cable -- for datalogging, very little power is required. Later, for PID, we might need more power so I'll think about this issue.

I'd love to send a few of these hand-built units out for evaluation, but my budget is tight right now. On the other hand, I can see where folks might be hesitant to buy something sight-unseen. Let me know what you think is a good approach.
 
allenb
Another question, I'll most likely want to plug into the trusty laptop until the roaster control cab mods can be done. Unfortunately, my favorite antique Dell is not USB ready but only has RS232 DB9 for a connector. Would it be much of a pain to transition from USB to my Dell?

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
bvwelch
Hi Allen,

Funny you should ask about that-- I've been meaning to poll you guys about that.

There are Arduino boards that come with built-in USB. But many if not most Arduino boards do not have onboard USB -- they depend on an external 'cable' which can be either USB or RS232. We can support both, so not a problem.

How old is your Dell though? It may not be good enough to run the Java-based Arduino or the Java-based 'Processing' that I was using for the real-time graph. However, we can work around this. Let me know more about your Dell. Or, look back in this thread for the links to the Arduino and Processing software, and just try to install them.
Edited by bvwelch on 05/28/2010 6:54 PM
 
allenb
Wow, maybe it's time to retire the old thing. I had no idea just how feeble it was gettin.

Dell Inspiron (1995 vintage) with Windows 98
-2 GB free space on hard drive
-48 MB free ram (not much on it other than a Fuji PID controller program)

Well doc, let me know if we'll have to put it to sleep. It's been a great companion.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
beanflying
Looking good smile

as most will only want 2 or maybe 3 TC inputs any thoughts as to making maybe the 3rd & 4th channel a simple voltage input channel as an option?. In the case of say a Hottop monitoring fan speed and having it overlay the graph or even power level with an optocoupled input?
My name is Tim and I have a coffee equipment addiction problem smile

Two Hottops - modded
TJ 067 Electric 1kg 5+years old
Insert new 5-8kg Roaster here urgently BBQ grill
 
bvwelch
Hey Tim, great ideas for the other channels. Tell me more about those circuit details. should be do-able. Also, the Arduino has several ADC inputs of its own, they just aren't good enough to read the microvolt levels of a TC.

Allen, that computer could be used to collect the data (it is ASCII CSV) with something like Hyperterm, but that is about it. Sorry. I gave up trying to support Win98 a while back, sorry.
Edited by bvwelch on 05/28/2010 8:59 PM
 
beanflying
Before you chuck the old W98 lappie providing it is still reliable just mothball it. My W98 box is currently running a converted plotter for vinyl cutting and I have assembled some bits to make a CNC foam cutter (roundtoit project ;) ) Both of those need real dos and a proper serial port to work so XP and up is no good.

Back On topic, I havn't looked at the Arduino in any detail doue to lack of time but if it already has spare channels then thats great all it needs is the software to be able to log from them as well as needed.

In the case of most of our smaller roasters the fans are 12V in the main so a simple resistor divider to get the voltage to whatever the Arduino needs is easy. For any of the bigger roasters fan control is normally a baffle plate across the airflow and I am not aware of any of them running AC motors and chopper circuits?

With power assuming it is AC based not gas but is chopped by some form of regulator. It needs a circuit to get it back to something that can be read safely without the chance of hitting the chip with 110 or 240VAC. :eye-popping:

I had a second think about it and maybe the easiest and safest way would be a mini transformer to drop the voltage then feed it into a filtered bridge rectifier network to give say a 0-5VDC at 240V and about 0-2,5VDC at 110V? Then add a trimpot to calibrate it back up to X volts at full roaster power to display it. It may not be perfect but it will be repeatable and you will be able to identify power levels from one roast to the next for comparison.
My name is Tim and I have a coffee equipment addiction problem smile

Two Hottops - modded
TJ 067 Electric 1kg 5+years old
Insert new 5-8kg Roaster here urgently BBQ grill
 
milowebailey
Randy

An Arduino is about $30... if you want true thermocouples then that's the expensive part

http://www.hacktronics.com/Arduino/Arduino-Duemilanove/flypage.tpl.html

Thermocouple is ~$15 each x 4 = $60

www.omega.com or ebay..

and the chip to read it another $20 x 4 is $80

AD595AQ

interface board x 4

$4
http://store.makerbot.com/thermocouple-sensor-v1-0-pcb.html

and a shield

http://www.nkcelectronics.com/freeduino-terminal-block-shield-arduino-compatible.html

~$10 so for $200 you can get a USB interface with 4 thermocouples.

not cheap, but very accurate and flexible.

easy to breadboard and I'm convicted to share this since I owe you the details as I mentioned over at GCBC.

Then I'll toss you some code.....
 
beanflying
I did a bit more reading on measuring chopped AC and a more accurate way to do it would be somthing like this in the voltage input channel http://www.analog.../AD736.pdf
My name is Tim and I have a coffee equipment addiction problem smile

Two Hottops - modded
TJ 067 Electric 1kg 5+years old
Insert new 5-8kg Roaster here urgently BBQ grill
 
bvwelch
Hey Milowebailey, thanks for jumping in, I welcome a little friendly competition. Perhaps we can collaborate on software and hardware too. The plan here is to use CC/BY and MIT open-source licenses. I'll put the source code up on github so you can throw rocks at it.

But you shouldn't discount the mcp3424 out of hand - it is a cool, low-cost alternative to your AD595AQ. The mcp3424 is a 4-channel 18-bit sigma-delta adc that a friend introduced me to. It works great and is very accurate. May I suggest you check out the TC example in its datasheet? We've been using this chip for a while now with TCs and it works fine. Check out the screenshots above.

The AD595AQ is a nice chip, just pricey. In contrast, the mcp3424 cost $4.26 and that is for 4 channels. Add 1 more chip, the mcp9800 (costs $1.26) for measuring ambient temp. You can get the mcp3424 pre-assembled on a PCB for $12 (thats for 4 channels). Add a little dip adapter $0.96 for the mcp9800 and you can see the advantage of this approach. Also, this solution is I2C and very low power suitable for battery operation.

You also mentioned full-sized Arduinos and 'shields' - these are great, if folks have plenty of room. But I was wanting a smaller form-factor. But I fully expect some folks to want to use the full-size Arduino and 'shields', so we'll support both formats.
Edited by bvwelch on 05/28/2010 11:53 PM
 
bvwelch
I've been doing some thinking about the PCB -- it makes sense, if you want the lowest power and a small size. But until we get enough interest to build a PCB ( about $100 will get us around 10 PCBs ), here is another solution.

Use an off-the-shelf temp sensor for measuring ambient. Like this one: http://www.modern...ure-sensor

Combined with the 4-channel ADC module:
http://www.modern...nalog-plug

Some cabling:
http://www.hvwtec...bCatID=100

and just about any Arduino, we are ready to go. Little or no soldering required!
Edited by bvwelch on 05/29/2010 11:55 AM
 
bvwelch
A reminder about thermocouples. The voltage at the meter end, is not absolute -- instead, it is the difference between the temperature at the probe/hot end, and the temperature of the meter/cold end. I call the temperature at the meter end 'ambient temperature', since for a hand-held meter, it will be a few feet away from the roaster.

When I say the 'meter end' of the TC wire, I mean the point at which you connect it to something other than TC wire, like a screw-terminal or a jack. Now some jacks, like the ones from Omega and Auber Instr, have nice screws/tabs such that you can 'extend' the TC wire inside your project box.

So, it is important that we mount our 'ambient sensor' close to the actual place where the TC wire ends, either at the jack or if you use the little green screws like I did in the photo.

This little gotcha is why the chip that Milowebailey uses costs $20 -- it has the 'ambient' sensor (they call it a cold-junction) built inside the chip itself. Nice, but expensive. What we do, is simply use software to perform this function.

Personally I'd say that as long as they are both mounted inside the project box you should be good. But one time, I mounted my project box onto the side of my popper, and I had large errors because the ambient sensor was several inches from the screws.

So long as we keep this in mind, we can save a bunch of money on the project. But if you decide you want to use the fancy chip, that is fine too. We can support either type.
Edited by bvwelch on 05/29/2010 12:50 PM
 
Jump to Forum: