Low cost temperature data acquisition


Active Member
Does anybody know of a low cost solution temperature data logging solution? I'm thinking into something like a probe connected to my PC's RS-232 port or USB. I dont even need software because I can capture the data into a file over time.

The modules ones that I saw around are +$200 without the thermocouple. I want it for a temporary project and would not like to invest much money on it.

I would go with the 1-wire sensors, they are dirt cheap (you can get a free sample from Dallas Semiconductor, don't have the url handy right now), and get the interface from: http://www.midondesign.com/ , but there are cheaper 1-wire interfaces.
Not real sure this will help but just FYI. Tempature can be read very simply with two piece of unlike metals. When you twist two unlike metals together it will emit a small current relative to the heat or coolness. Here is a list of the metal types. Search for Thermocouples to learn more.

RJTK......T and K.......... Cu/CuNi and NiCr/NiA1
RJRS......R & S......... ...Platinum 10 and 13% 
RJN........N...................NiCrSi - NiSi
RJE........E............. ......Nickel-chromium / copper nickel

Here are a couple of quick links
http://forum.allaboutcircuits.com/index.ph...8&mode=threaded (Project/Circuit Diagram)
electron said:
you can get a free USB interface for this 1-wire stuff now ?!
Just got mine in the mail. I ordered 2 but only got one, but who's complaining.
Will anyone be able to provide a How-to on these free parts. I just got samples of the ones rupp listed but Im not sure what to do with them. It would be nice if we could work up a How-to
The ideas are very good, but I'll provide some background to help you help me.

I will be helping a friend to make some school experiments around freezing sea water. He needs to collect data over time, but the probe needs to resists the corrosion, water, salt corrosion and low temperature. The iButton would be great if it had internal power to save the data over time. The USB with the blue-dot would corrode.

The bi-metal (thermocouple) idea is good, but I would have to build some circuitry to condition to signal before cpaturing, and them build a transformation table in the PC to linearize the signal created by the thermocouple - it is never linear. The RTDs are much more linears but much more expensive too.

In order to avoid building the electronics, I might buy the circutry from http://www.phanderson.com and a thermocouple to connect to it, but it is not a cheap solution anyway.

I would suggest staying "linear" for simplicity in the data collection. Look at Relco temperature probes. They are not cheap, but they have an incredible linear range and I believe the probe is resistant to the elements you will subject it to.

Of course you would then have to find a way to record the analog voltages and assign "engineering units", but there are plenty of data logger's out there.

Of course none of this stuff is cheap though.
Here is another option for reading the 1-Wire temp sensors

I have not purchase one of these myself but they do look interesting.


I have had one of Peter Anderson 1-wire readers (256 1-wire plus i/o) for 2-3 years and it works great.
I placed the order, and got this email:
I am in receipt of your request for samples of the DS 9490 R which requires business manager approval.

Unfortunately your request has been denied. This may be due to several reasons such as pricing or availability.

You may contact Maxim/Dallas direct at 1/888-629-4642 to purchase.

E. Leal
Sample Dept
I placed a sample order on Tuesday , but haven't reeceived it yet or been rejected. For me, the DS9490R didn't indicate it required special approval, but the DS9097U-S09 did. time will tell..