Go Down

Topic: LM35 in water? (Read 299 times) previous topic - next topic

jimLee

Set up a circuit for reading water temp with a couple LM35s. Seemed like a really simple undertaking. Everything worked as expected on the bench. So we coated the LM35s in liquid electrical tape, let 'em dry and plunked em in the water tank. They went bonkers.

Well, maybe the liquid electrical tape wasn't cutting it? We made up a new set coated in either epoxy or polyester resin. Can recall. Same thing, works great on the bench, goes wild in water. So we finally potted the sensors in Sharpe caps poured full of polyester resin. There's no way water can get in there. Same thing! We even made up a set using different wire 'cause we thought the insulation may be seeping. Nothing has worked.

Has anyone seen anything like this?

Ideas?

Thanks!

-jim lee

Peter_n

My idea : use a digital temperature sensor like the DS18B20.
http://www.adafruit.com/product/381
Adafruit has also food-grade heat shrink for it : http://www.adafruit.com/products/1020

The Arduino is connect to a computer via USB, or via a power supply somehow a little connected to the mains.
I suppose the water is grounded in one way or the other. Somehow there seems to be a lot electrical noise in your circuit. Either from the water or from the arduino or from the mains or a computer... probably from everything together.

Can you explain what happens, what are the values ? You say "bonkers" and "goes wild", for some people that is the temperature changing half a degree, for others it is releasing the magic smoke  :smiley-confuse:

Off-topic: Why isn't there a smiley which is releasing the magic smoke ? (and I don't mean a smoking smiley).

cattledog

I have seen commerially available waterproof lm35 sensors available on alibaba and this one here from a Greek supplier http://grobotronics.com/temperature-sensor-waterproof-lm35.html?currency=USD&sl=en

I certainly don't trust Instructables, but the tutorial on waterproofing the lm35 looks reasonable, and you'd hope if the thing didn't work they would have indicated some problem.
http://www.instructables.com/id/Waterproof-a-LM35-Temperature-Sensor/

I am investigating an underwater temperature measurement system to determine the thermocline in a lake, and want to send a probe down 20m. I have the issue of wire length as well as being waterproof. I know there are waterproof versions of the DS18B20, the lm35, and 10K NTC thermistors, all of which I have used successfully at short range. As Peter_n suggests, have you looked at alternatives to the lm35?

I

jimLee

1/2 degree would be fine. It does a random swing around +/- 10 degrees? I'd have to look bcd at the data. This was both running off a 9V battery & USB port. Different tests. I actually plotted values to see if there was some sort of pattern and I could see none.

Just plain weird.

-jim lee

Peter_n

#4
Feb 27, 2015, 10:28 am Last Edit: Feb 27, 2015, 10:28 am by Peter_n
A 9V battery is not good to power the Arduino.

We have to know more to say something about it.
How is your circuit ? How long are the cables ? which type of cable ? Can you make a photo of it ? and so on.

It is not weird, it is normal. The LM35 is not only a temperature sensor, it is also an oscillator !
Did you see the datasheet, section 8.1.1 page 14 ? http://www.ti.com/lit/ds/symlink/lm35.pdf
Long wires with capacitance turns the LM35 into an oscillator.

That doesn't mean that the other things are okay. Everything should be done right to get accurate temperatures.


polymorph

It sounds like it could be electrical noise in the water, or as Peter_n suggests the extra capacitance between the wires caused by the water may be causing it to oscillate.

In addition to the suggestions on page 14 of that datasheet, you might add a 10uF capacitor from Vs to Gnd right at the LM35.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - tinyurl.com/q7uqnvn
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

jimLee

Found the issue. Putting the twisted power,Gnd,Signal wire in the water was the culprit. The water raised the capacitance between the wires causing the sensor to go unstable. There must be some sort of feedback inside the sensor. Anyway, 2k resister inline from the sensor to the output wire solved it. Then a larger resister, I didnt get the size, I'd assumed 10k or so, between the output wire & ground wire knocked the RF interference out of the picture as well.

Everything is humming along just swimmingly now.

Thanks!

-jim lee

polymorph

Great!

And thank you for letting us know what solved the problem.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - tinyurl.com/q7uqnvn
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

Go Up
 


Please enter a valid email to subscribe

Confirm your email address

We need to confirm your email address.
To complete the subscription, please click the link in the email we just sent you.

Thank you for subscribing!

Arduino
via Egeo 16
Torino, 10131
Italy