First Temperature Probe Build

Hey all,

I am planning on making a temperature probe to use when I am brewing... I have seen that many have used the LM34/35 sensors. I am not sure that these are suitable for long lead applications (in excess of 10 ft). Are there any sensors that others are using (in the TO-92 package) that are for long lead applications? I plan on using this sensor in a max. diameter SST tube of 1/2".

Thanks for your time,

MT

That would probably depend on how you are going to use them. I use the Dallas 18B20 one-wire sensors in my fermentation controller (Arduino Clone).

Your going to want the DS18B20's as noted, they are digital so they will work regardless of length.

With maximum bit usage(12 bits) you can get down to .11C measurement points(not accuracy) i think they have a +-2C accuracy, but i think thats on the high end for what most are capable of.

MT_Keg said:

Hey all,

I am planning on making a temperature probe to use when I am brewing... I have seen that many have used the LM34/35 sensors. I am not sure that these are suitable for long lead applications (in excess of 10 ft). Are there any sensors that others are using (in the TO-92 package) that are for long lead applications? I plan on using this sensor in a max. diameter SST tube of 1/2".

Thanks for your time,

MT

Click to expand...

What controller/processor are you using? Some only allow the use of certain probes.

DS18B20s will work on a 10' wire, and a 3.5mm audio jack socket will fit nicely into 1/2" ID. See http://www.jimsbeerkit.co.uk/forum/viewtopic.php?f=6&t=31341&p=346461#p339227

I have a Basic Stamp microcontroller right now that I am doing some Prototyping on... I could easily move to a PICAxe or Arduino. I primarily would like to use it for fermentation temperature control. But I was also thinking of using it for holding the water during single temperature infusion or during multi-rest mashing. Is +/- 2 deg. for a temp accuracy ok for doing this or is it overkill?

Thanks for your responses!

MT

DS18B20s are typically more accurate than that, the range is more like +/- 0.5 oC, which I think is ideal for mashing. You can always calibrate them, in fact I would routinely recommend for any kind of temperature probe just as a sanity check.

I find that the DS18B20s are incredibly accurate, closer to 0.5C like alien said. However, one thing I've found is that you can actually heat up the device (and therefore give you a false reading) if you tell it to measure too often. The datasheet says you can use read time slots to determine if the temperature measurement is done, and they actually complete measurement not in 750ms but closer to 20-30ms. Reading from them this frequently keeps them out of power save mode and they'll go up about a degree or two. This can be easily avoided by limiting your polling interval to 750ms.

I use RJ11 telephone cords for my probe connectors. Just cut off the end of a phone cord and solder it directly to the DS18B20 with some heatshrink. I've connected 2x 6ft cords together using an off the shelf phone splitter and it works as well.

Here are some guidelines on using DS18B20s on long wires:

http://www.maximintegrated.com/app-notes/index.mvp/id/148

alien said:

DS18B20s will work on a 10' wire

Click to expand...

I've got several connected to the ends of 50' telephone cords that work with no problem.