Long temp sensors
Hi. I bought 2 of the temp sensors off the site to test with and they work accurately. What I'd like to do now is either buy new sensors or lengthen the ones I have. What's the best way to do this?
I'm about finished building a solar water heater and I want to run the sensors from the panel outside into my computer in the house.
I found one of these bad boys through a Google search LINKY.
Could I just extend the 3 leads with cat5 cable or something similar?
Don't quote me (Or someone else might be able to confirm), but I think that if you make the leads longer, the reading gets skewed, so you could probably lengthen it, but you would need to recalibrate the readings to suit.
Why would you use that instead of the ones you have bought from us? Ours have a larger response range and work the same otherwise. Just extend the leads.
This is correct. The added length of wire will scew the readings. But it should be a constant skew so you can correct for it in software.
Originally Posted by Chester
I wasn't sure if I could simply extend the leads or not and have things still work correctly.
Originally Posted by 2k1Toaster
Temp sensors are usually low impedance (eg LM335; typically under than 1 Ohm) hence lead length shouldn't be a problem if your voltage measuring circuit is high impedance and the supply current isn't causing a major voltage drop.
Good deal. I got the sensors and I made a 30' cable. Quick side note: where can I buy the little black things that connect to the brain? What I got didn't work so well. Anyway I have my 30' cable and I used the "Temp �C = 100*(reading in V) - 50" formula that they recommend.
The reported temp comes back within a few degrees below the actual temp and it also seems to swing roughly 5 degrees F. It has me rather puzzled.
I'll make a short cable tomorrow and see if that clears up the issues.
Remember an antenna is just a wire. You are making a 30ft antenna and carrying a relatively small signal over it. So even lights will introduce a 60Hz (or 50Hz if you are in Europe) noise. It will not be as accurate. Also the formula will need to be adjusted a little bit because the signal will deteriorate a little bit.
Originally Posted by scotty562
Version 5 will fix all of these issues. :)
Try small caps across the line...
I'd try across the supply at the sensor end, and across the signal (to GND) at the brain end.
Probably a greencap (ceramic) or similar of .001uF to 0.1uF (1nF-100nF).
STOP PRESS - suggest 1nF for the output to gnd (at the brain end) - I just read the specs....
See page 17 if still a problem. Also p10. (Of the linked adafruit pdf datasheet.)
Our sensors already include capacitors at the sensor.