Who is here? 1 guest(s)
 Print Thread
4-channel TC meter and datalogger project
randytsuch

Quote

bvwelch wrote:
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 )


I would be in for a PCB.

Do you do any signal conditioning (amplify) of the TC signals, or do you just feed them straight to the A/D?

Randy
 
bvwelch
Hi Randy,

The chip mcp3424 handles everything. It has a built-in amplifier, but I just have its gain set to 1 for now, since it is 18-bits and a 2.048 volt reference. Datasheet: http://www.microc...e=en536354
Edited by bvwelch on 05/29/2010 1:16 PM
 
bvwelch
By the way, since we're going the 'all I2C' route, here is a way to use our ADC board with just a PC - no Arduino at all!

http://www.acrona...B-I2C.html

Let me know if this appeals to any of you.
 
bvwelch
Well, this board shows how little I know-- here's a nice little Arduino with built-in USB:

http://spiffie.or...ts/iduino/
Edited by bvwelch on 05/29/2010 2:55 PM
 
allenb

Quote

bvwelch wrote:
By the way, since we're going the 'all I2C' route, here is a way to use our ADC board with just a PC - no Arduino at all!

http://www.acrona...B-I2C.html

Let me know if this appeals to any of you.


I'm probably just doing some wishful thinking but does this mean ( Acroname specs below) that I could still use my old Dell?

2C Devices Any registered or unregistered I2C device.
Host support Drivers available for Windows XP64, XP, 2000, ME, 98, CE.NET V4.2, Linux, Mac OS8, OS9, OSX.
Voltage Self powered - 5V available
Current Up to 70mA available for your own circuits
On-board Pull-up resistors 4.7k
Digital I/O 1 Input and 2 I/O available when not using I2C.
Analog Input 2 10-bit channels available when not using I2C
I2C clock rate 100 kbps
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
randytsuch

Quote

bvwelch wrote:
By the way, since we're going the 'all I2C' route, here is a way to use our ADC board with just a PC - no Arduino at all!

http://www.acrona...B-I2C.html

Let me know if this appeals to any of you.


Hi Bill
Not trying to be critical here, but for $35, I don't see an advantage to using this. An arduino board is cheaper, and I would guess that would allow for more flexibility in the long run.

Randy
 
milowebailey
bvwelch

I'd like to know more about this A/D... what Arduino code are you using to read it. For the AD494 there is code written... but if you have some that would work with the A/D I'd be interested in going that way.

I'm interested in only 2 thermcouples, but that should work too.... right.

I'm developing a roaster control so I need the other I/O of the Arduino and if the A/D will buy me back one input that would help.

I've also looked into seeduino... which has even more I/O than the Arduino.
 
bvwelch
Allen - you don't have USB on that old laptop, so you can't use that USB-I2C gizmo. And besides, I can't support Win98. :-)

Randy - I agree, the Arduino is better, but just wanted to put it out there to see what response I got.

Milowebailey - I'm no expert, but I do making my living writing software, for going on 30 years now, and my favorite is so-called 'embedded' programming, like the Arduino. I can either track down existing Arduino libraries, or write new ones from scratch if we need one. They will be open source and you're welcome to use them, suggest improvements, or make improvements on your own, glad to share with you.

In the case of this mcp3424, the Jeelabs guy provides a library: look for 'analog plug' : http://cafe.jeela.../hardware/

Also, there is good details in the datasheet: http://www.microc...e=en536354 . If the chip has a 'weakness', is that it is relatively slow. Only 3 or 4 samples per second, but that is plenty good for roasting, in my opinion.

You mentioned roaster control, which I've dabbled with in the past ( I was using a picmicro back then ). I look forward to working on that, after the data logging is finished. Hope we can collaborate on that.

About I/O pins - I think you'll love the I2C protocol that these chips use. With I2C, you can 'daisy-chain' many chips/modules together, and only use a total of 2 Arduino pins. The guy that I'm getting my ADC 'module' from, has several 'plugs' (as he calls them), and they just snap together sort of like leggos. He has open-source libraries for all of them. Even though he sells his own style of Arduino boards (he calls them JeeNodes), they will work on any Arduino or clone.

One in particular that you'd like is the I/O expander, which gives you 8 more digital I/Os. Another nice one is the LCD 'plug', which I'm using to drive an LCD. (Not shown in that photo).

For example, in my prototype setup, I can run the 4-channel ADC, the ambient temp sensor, and the LCD, all with just 2 I/O pins from the Arduino. Nice. The JeeLabs guy is in Holland, but here's his state-side web-shop: http://www.modern...ns/jeelabs
 
bvwelch

Quote

randytsuch wrote:
I would be in for a PCB.
Randy


Hi Randy,

Do you favor an all-in-one board that has the usb, cpu, and adc, screw terminal block, etc? Or do you prefer to leave off the usb and cpu? I may try both approaches in Eagle and see how things fit and get a cost estimate from 4pcb.com.
 
randytsuch

Quote

bvwelch wrote:

Quote

randytsuch wrote:
I would be in for a PCB.
Randy


Hi Randy,

Do you favor an all-in-one board that has the usb, cpu, and adc, screw terminal block, etc? Or do you prefer to leave off the usb and cpu? I may try both approaches in Eagle and see how things fit and get a cost estimate from 4pcb.com.


Hi Bill,
I had just been considering an add on board, and had almost bought the seeduino board, it looks pretty nice.

But, thinking about it, an all in one board would have some advantages. I think it would be cheaper, and you have all you need, except for display and keypad, which are optional anyway, on one board.

I was thinking the all in one would be more work, but since you have to wire from the arduino to the ADC board anyway, and still build the adc, it probably isn't that much more work, if any. Then if you factor in the extra effort to mount two boards, versus one, seems like a draw there.

Looks like I talked myself into voting for the all in one.

Let me know if I can help in some way.

Randy
 
bvwelch
I ran into one snag with the all-in-one approach - for whatever reason, the nice SMT version of the cpu chip itself is unobtainium - we'll have to use thru-hole (DIP) part.

I'll see if everything will fit, if not, we can go with a 2 board set-- our TC board and something like the RBBB -- it's PCB is very cheap: $2 in qty 10.
 
Randy G

Quote

bvwelch wrote:
I ran into one snag with the all-in-one approach - for whatever reason, the nice SMT version of the cpu chip itself is unobtainium - we'll have to use thru-hole (DIP) part..

I wouldn't stress over that detail. Two separate boards may even have benefits in some installations.

Life's too short to drink bad coffee.
 
JETROASTER
Still hoping... any chance of a pressure channel ?
Scott
 
bvwelch
Hi Scott,

As far as I know, we should be able to handle pressure fine -- we'll have extra I/O for experimenting.

Have you identified sensors besides the ones CHAD found at Omega?
 
JETROASTER
It's not my field of expertise. I'll stick with those unless
anyone feels there's a better fit.
I think whatever works better in your design is how I will go. Scott
 
randytsuch

Quote

bvwelch wrote:
I ran into one snag with the all-in-one approach - for whatever reason, the nice SMT version of the cpu chip itself is unobtainium - we'll have to use thru-hole (DIP) part.

I'll see if everything will fit, if not, we can go with a 2 board set-- our TC board and something like the RBBB -- it's PCB is very cheap: $2 in qty 10.


The DIP is probably a better choice for most people, I think some people wouldn't want to deal with soldering a SM IC.

No problem if it ends up as 2 boards.

Randy
 
bvwelch
soldering SMT isn't nearly as bad as I thought. Sparkfun has some nice video tutorials of smt soldering. The big 'oh, that's how they do it', for me, was -- they don't worry about solder bridges at all - they just go back and wick it away. Easy.

But I also figured I'd offer an 'assembled board' or 'partially assembled' board - maybe trade for some greens. :-)
 
randytsuch

Quote

bvwelch wrote:
soldering SMT isn't nearly as bad as I thought. Sparkfun has some nice video tutorials of smt soldering. The big 'oh, that's how they do it', for me, was -- they don't worry about solder bridges at all - they just go back and wick it away. Easy.

But I also figured I'd offer an 'assembled board' or 'partially assembled' board - maybe trade for some greens. :-)


I solder SMT stuff, but I have access to really small tips, microscopes and other special tools which make it easier to do. I was really speaking for others, but maybe I am selling people short.

If you offer a board with the part soldered, that problem goes away.

Randy
 
bvwelch
I must have been confused - I see the nice small SMT cpu chips now at mouser. I may look at the all-in-one idea again... I also worked on a 'shield' version of the project. A shield might appeal to a wider audience... Any thoughts?
 
farmroast
Great to see more work being done in this area of controls. I've been totally enjoying JimG's first prototype of a roast BT speedometer. It has added a whole new and better perspective on my roast profiles.
Ed B.
DreamRoast 1kg roaster, Levers, Hand Mills http://coffee-roa...gspot.com/
 
randytsuch

Quote

bvwelch wrote:
I must have been confused - I see the nice small SMT cpu chips now at mouser. I may look at the all-in-one idea again... I also worked on a 'shield' version of the project. A shield might appeal to a wider audience... Any thoughts?


All things being equal, I think most people would want the approach with the lowest total cost to get a working system.

I would guess that an all in one board approach would give a lower cost.

But, the shield approach might make it more flexible.

I think either way will be fine, flip a coin?

Randy
 
JETROASTER
I would still vote for versatility even at a at 25-35% increase...my 2 cents. -Scott
 
bvwelch
Thanks for the feedback guys. Here's a first cut of an Arduino shield: 4-channels for TC's, with provision for experimenting with inputs like millivolts, 4-20mA, etc. There is a generic 'prototyping area' also, since there is a lot of unused room on the 'shield'. An Arduino shield is about 2x3 inches. The board looks big in the PDF because I zoomed in before I printed the PDF.

http://github.com...ch/arduino

Thanks for any suggestions! -bill

ps: the 'github' site takes a little getting used to-- if you can't read a file, click on the 'raw' link. Sorry for the inconvenience.
Edited by bvwelch on 06/10/2010 11:41 PM
 
JimG
What a great thread! I'm just getting (sorta) caught up. Thanks to all who've contributed to developing the proposed Arduino shield.

My analog rise meter set up is still working very well for me. I designed it to fit a specific situation, so it lacks a lot of the features that this Arduino shield will offer. (Since I already have a couple of PID's hooked to my roaster, I had no need for either a temperature display or the ability to do roast control).

I'm in the process of getting rid of the external DMM now needed to display the 0-200mV output from my analog gizmo. It will soon be replaced by a PM128 ($10, eBay) digital panel meter so the whole gizmo will be self contained.

But since I've already got an Arduino up and running here, I'm very anxious to experiment with this new shield, and I'm wondering about 3 things:

1.) How the TC signal (which is in the 5's of millivolts) gets amplified? I know the ADC can be configured with a gain up to 8X, but seems like there needs to be another decade or so of amplification somewhere?

2.) Why I can't open the .sch and .brd files in Eagle? I'm running version 5.10, and it does not like those files at all.

3.) If there are short term plans to order a run of shield PCB's? If so, I am in.

Jim
 
bvwelch
Hi Jim,

Your project was an inspiration for me for sure! I am excited that you have an interest in an Arduino version - I'm sure you have lots of practical experience that I lack.

The 18-bit ADC has plenty of resolution - in fact, I leave the gain set at 1 for now. The ref is 2.048 volts so with 18 bits that is a lot of resolution (15uV per bit) with approx 40uV per degree C. Please double-check my numbers, here is the datasheet: http://ww1.microc...22088c.pdf

I definitely want to make the Eagle files accessible - in fact I hope others can help make them better - this is just a first cut to get some ideas out there for folks to throw rocks at.

Until I figure out a better, just click on the 'download source' link on the top-level github page - this will give you a zip file of everything. I think this is the link to the 'download source' tab:
http://github.com...ves/master

Thanks! -bill
Edited by bvwelch on 06/11/2010 9:59 AM
 
Jump to Forum: