DIY $15 Temperature Probe and Readout

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

goatchze

Well-Known Member
Joined
Nov 13, 2008
Messages
108
Reaction score
2
Location
College Station, TX
So I'm beginning to set up a RIMS system and would like to monitor temperature at several places in the process. However, thermocouples and readouts can start to get both bulky and pricey! What I want is a simple measurement device and a simple readout.

So, what to do but build my own circuit! Here's what I've come up with.

temp_probe.jpg


This circuit uses the LM34 IC for the temperature measurement. It prodcues 10mV/degree F. I set up a voltage divider to divide this value by 10, giving 1mV/degree F. Then I just hook up a 200mV voltmeter to readout my temperature directly!

I've ordered the parts and they just arrived. I'll be doing some testing this evening and will post up photos and results. My total cost was $50 which will allow me to set up 3 measurements and 3 read outs. That's a little over $15 each!
 
What do you put the sensor into, some sort of probe? Sorry, complete newb at this stuff, but really interested.
 
Maybe replace the 100kOhm resister next to the thermistor with a pot so you can calibrate?

? The 100kOhm resistor is just there to keep the current down flowing through the indicator LED. It doesn't do much of anything else.

The LM34 is precalibrated and reads linearly. It's guaranteed accuracy is 1F at 77F and claims +/- 1.5 up to 300F. (We'll see when I get it all set up)

I'm not 100% how it works (whether there's a thermistor incide the IC, which I imagine there is), but it's output is voltage.
 
What do you put the sensor into, some sort of probe? Sorry, complete newb at this stuff, but really interested.

The LM34 itself is REALLY small (5mm..the same size as a standard LED). I plan on putting it either inside of an acorn nut or inside of a copper tube with a cap. Then it's going to go into a "T" joint so that it is exposed to the wort that's flowing by. I have the TO-92 version which is all plastic and must be isolated from the liquid. If you can find the TO-46 version it is inside of a metal case whose top CAN be directly exposed to the fluid.

You can also put it in a wand if you want to.
 
I've ordered the parts and they just arrived. I'll be doing some testing this evening and will post up photos and results. My total cost was $50 which will allow me to set up 3 measurements and 3 read outs. That's a little over $15 each!

Where did you get the lm34's? I tried to get some samples from National and they wanted $54 shipping for 4 samples.. "Due to popularity" they stated.. LOL
 
What exactly is an LCD Voltmeter panel? It it some fancy LCD panel that is a voltmeter or is that just attached to an LCD voltmeter and your using its output as a temperature readout as it is calibrated? and the LED is just a "hey, im on" light?

edit, I see you descreibed it as a voltmeter, so would it really be "$15 temperature probe with readout*"

*multimeter not included
long as you have a digital multimeter
 
I ordered most of the parts from BGMicro (www.bgmicro.com). They have LM34DZs, which are unfortunately the least accurate LM34s, for under $2. I would prefer a LM34A but couldn't find them anywhere.


It's an LCD panel display and a voltmeter in one. It's the most expensive part of the set up at $8. It reads in mV up to 200mV. The voltmeter will read the LM34s output voltage which directly correlates to temperature. The voltage divider gets the decimal in the right place.

No multimeter is required. You could certainly use one, but they're big, bulky, expensive, and overkill in this case. I wanted a really simple piece that I could bolt on to a small control panel. Using a multimeter would make that tough.

And yes, the LED is just a "hey I'm on light" and isn't needed, but who doesn't like lights to come on when they flip switches?? :)
 
It's an LCD panel display and a voltmeter in one. It's the most expensive part of the set up at $8. It reads in mV up to 200mV. The voltmeter will read the LM34s output voltage which directly correlates to temperature. The voltage divider gets the decimal in the right place.

Cool! I guess Im not surprised they make such things... This seems almost too easy ;)
 
Yup, it's that Pfetzer valve, right there! Yup.

It's all ball-bearings these days. Now you prepare that Pfetzer valve with some 3-in-1 oil and some gauze pads, and I'm gonna need 'bout ten quarts of anti-freeze, preferably Prestone. No, no make that Quaker State.
 
Sorry, I didn't have time to mock up the circuit last night. SWMBO had a long to-do list for me when I got home! Hopefully I'll get it working tonight.

I'll include a parts list once I can confirm that the thing works!
 
Will you be using a separate circuit for each with its own LCD, or will there be some way to "scroll" through the separate measurements using one screen?
 
Success!

Last night I almost had the circuit built but SWMBO took over my time. This morning I was able to take a quick look at it and fix a simple mistake. The result? It works!

01-15-10_0952.jpg


The thermostat at my office is set to 75F (yes, it's hot in here). It reacts fairly quickly to small changes (such as putting my fingers over it) but does take a bit to make major swings (such as placing it in the freezer). But once it's stabilized it responds pretty well to minute changes.

Voltage divider works like a charm. The only thing missing is I forgot to short a jumper on the LCD panel to put in the decimal place. Other than that it looks good!

Now for more testing and then a more permanent setup with a PCB.
 
Will you be using a separate circuit for each with its own LCD, or will there be some way to "scroll" through the separate measurements using one screen?

At the moment I have 3 LCDs and 3 LM34s. I want to have a temperatures sensor in two in the MLT to get a good profile. I'll have a PID controller on the outlet of the RIMS heater to get that measurement. I plan to use the 3rd LM34 just upstream of the RIMS heater so that I can get the temperature across the heater (and therefore be able to approximate my circulation rate).

In the future I'm going to want to measure just in front of my CFC and just after it, so two more LM34s. I plan on using the same screens that I have and just use a toggle switch. This is why I included the indicator LED; I want it show me which probe my panel is measuring.

Thanks for the links to other LM34s. After I get it all working I'll probably upgrade to the higher accuracy versions. Of course, this is going to push the price up a bit (they're $6-7 instead of $2-3).

Parts list coming up next.
 
Parts List:

PM 128-E: LCD Panel Voltmeter........................$8.95
LM34DZ : IC Temp Sensor...............................$1.89
200Ohm Precision Resistor 0.1%.......................$1.77
1.8kOhm Precision Resistor 0.1%......................$2.31
9V Battery Snap............................................$0.41
Push On/Push Off Switch.................................$0.99
1kOhm Resistor 5% (pack of 200)......................$2.95/200
Red LED........................................................$0.49

Total............................................................$16.82

Throw in the PCB board, soldering, and wire, it probably comes out to $20 instead of $15.

Also, the 0.1% resistors may be overkill; you could probably get 1% for cheaper. I don't think the 5% resistors are accurate enough.

I ordered everything from BGMicro except the high precision resistors which came from mouser.com.
 
Parts List:

PM 128-E: LCD Panel Voltmeter........................$8.95
LM34DZ : IC Temp Sensor...............................$1.89
200Ohm Precision Resistor 0.1%.......................$1.77
1.8kOhm Precision Resistor 0.1%......................$2.31
9V Battery Snap............................................$0.41
Push On/Push Off Switch.................................$0.99
1kOhm Resistor 5% (pack of 200)......................$2.95/200
Red LED........................................................$0.49

Total............................................................$16.82

Throw in the PCB board, soldering, and wire, it probably comes out to $20 instead of $15.

Also, the 0.1% resistors may be overkill; you could probably get 1% for cheaper. I don't think the 5% resistors are accurate enough.

I ordered everything from BGMicro except the high precision resistors which came from mouser.com.

I was getting hung up about tolerance once and an EE pointed out I could just buy a bag and test them and I'd probably find one out of the100 that was well within the range I needed.
 
If I have this right in my head then you'll have a small pcb with some elements and 3 wires to the the ic as your temp probe, will the length of the wires affect the accuracy of the readout?
 
If I have this right in my head then you'll have a small pcb with some elements and 3 wires to the the ic as your temp probe, will the length of the wires affect the accuracy of the readout?

I put it all together on a breadboard to make sure it worked. Now I'll solder it all up to a PCB.

The length of the wires may affect the readings, but not much. I'm only expecting 3, maybe 4' of wire from my readouts to the IC. Voltage drop should be very small across the wires.

The LM34 does have a limited ability driving capacitance, which means you are limited on how long the wires can be as is. However if it becomes an issue there are several options to get it working.
 
How did this turn out? I would be interested in making one for mashing. Did you ever test the accuarcy around 155 range?
 
ekjohns,

It worked out well. I've been traveling a lot lately and haven't brewed in a while, but I got the sensor(s) working just fine. I placed the IC temp sensor inside of a "probe" made of copper tubing with an end cap and the back end sealed with silicone and JB weld.

I replaced the breadboard in previous picks with the little do-it-yourself circuit boards from radioshack.

Accuracy of a temperature measurement is always a little difficult since I'm comparing with a device that has errors of it's own. Still, the homemade sensor matches well to my store-bought sensor. I have done some quick tests at higher temperatures, but I haven't tested it exhaustively.

I'll see if I can't get some pictures up soon.
 
Back
Top