Cheap Humidity Sensor w/ Resistor

[krazatchu]

Hi all...

After seeing this idea mentioned a few times on various threads, I gave it go...

The resistor's shell (R8) was scraped off... As the resistors core absorbs moisture the reistance varies...

It seems to be very sensitive and very fast.... It will detect the moisture from my finger from about half a cm... it takes less than half a second to respond either way...

To test if it was sensing the heat instead of the moisture, I tried a wet piece of paper... and also tried wrapping my finger in plastic.. In both cases, it seems to be sensing only moisture...

Perhaps this will be of some use to some one...

Michael

 

[dknguyen]

Have you tried characterizing it yet (with a regular humidiity meter?)? How about breathing onto it? or waving it it over a humidier?

 

[krazatchu]

Breathing onto it also works... tried that from about 30cm distance...

What do you mean by "characterizing" ??

EDIT... Oh, I see.. no, I don't have any thing to reference it too...

Thanks,
Michael

 

[AND_ECE]

Have you tried characterizing it yet (with a regular humidiity meter?)? How about breathing onto it? or waving it it over a humidier?

Hi.. alberta!! I am from Philippines, can you give us idea for our project design? ... i really-really appriaciate,if you help for our project design

 

[whiz115]

interesting...
we should call it Humidistor™

i've noticed with my multimeter that as humidity raises... resistance
falls... is this the case?
do you think we can make it more compact with only one IC?

 

[blueroomelectronics]

Humidity sensors tend to be capacitive not resistive.

 

[krazatchu]

Hey all...

I was away for a while, was moving my office to bigger location~~
Needed more space for my toys~~

Anyhow, blueroomelectronics, do you know of any reason why they tend to be capacitive?

whiz115, we better get that name registered right away~~ It can be made with just the lm393... and configured for either hi or low output depending on which resitor you choose for the sensor...
R8 was where I placed the Humidistor, however the outpt can be reversed and the invertor droped if you use R9...

Thanks,
Michael

 

[blueroomelectronics]

They appear as both, I guess the cost determines popularity.
**broken link removed**
**broken link removed**

 

[krazatchu]

Hey~~

Thanks, for that article... It's very informative...

 

[Conradb]

2W/3W resistor attempt

I have not put anything in here for size comparison, I think it's a 2W; I used a 2.2MOhm (I got a few in a maplin lucky-bag). The one on the right was scraped too far and went open-circuit, and the original is show on left.
The one in the middle came out right, but strangely enough it owns up to being a 4M Ohm now, which probably means I just started to scrape away the oxide a tad. Even without a bridge circuit, my multimeter is able to detect a humidity change over ~10 second interval. So I'll be building the bridge circuit to amplify it nicely; if I use a stable op-amp. Anyone got more photies, of their attempts, etc. I will build the circuitry this weekend and draw up some of my stats and past those with a in-focus picture for the curious.

BTW, I am not too sure about humidity measurement - I always thought that 100% humidity refered to the '% chance for condensation on a body at ambient temperature', and that 100% is not "underwater" at all?

 

[stevez]

Relative humidity refers to the capacity of air to hold moisture as a vapor. 100% RH simply means the air can hold no more water vapor. If the air over a body of water were 100% RH then no water will evaporate into the air. One could argue that at 100% RH it is possible that objects could become wet with liquid water as a result of condensation.

Moisture affects the electrical (among many other) characteristics of many materials. Temperature does too. The choice of material for a sensor is driven by it's behavior over a range of conditions that would include RH and temperature - how repeatable it might be, how linear, how stable, etc. Manufacturability is certainly a consideration. Some electronic sensors I used in the past had a range that was limited to no more than 80%. Exceeding that could or would foul up the calibration even if returned to below 80%.

Still, your sensor might be quite useful for some applications.

 

[krazatchu]

If your camera doesn't have a macro mode (usually flower icon), try a scanner.. just lay them on the bed and scan them...

The LM393 Op amp I used is specifcally suited for output to digital circuits..
You may want to choose something else...

Thanks

 

[carl_m817]

i'll consider this one of yours! im interested =)

Hi all...

After seeing this idea mentioned a few times on various threads, I gave it go...

The resistor's shell (R8) was scraped off... As the resistors core absorbs moisture the reistance varies...

It seems to be very sensitive and very fast.... It will detect the moisture from my finger from about half a cm... it takes less than half a second to respond either way...

To test if it was sensing the heat instead of the moisture, I tried a wet piece of paper... and also tried wrapping my finger in plastic.. In both cases, it seems to be sensing only moisture...

Perhaps this will be of some use to some one...

Michael

hi there my friend!

i actually have an on-going project, that is a Weather Observatory and Humidity reading is included.

i am constantly watching this topic to see the development..,
and now i have decided to explore about your circuit myself..

but can i ask a few questions?
1.) did you try comparing the reading from that scraped resistor (with the actual hygrometer?
-i have my digital hygrometer here and i'm planning to compare my own reading (as soon as i do my experiment).
3.) what are you reading from that scraped resistor by the way? (voltage? resistance?)
2.)how would i scrape the resistor? enough to expose the steel/iron?

thanks for this post!
pretty interesting!!

 

[carl_m817]

so i'm here again-

i already bought the materials for the circuit. but my problem is:
22Mohms is not available so what i did is:
2 10Mohms + 1 2Mohms resistor

any suggestion how would i do the scraping?

 

[Torben]

so i'm here again-

i already bought the materials for the circuit. but my problem is:
22Mohms is not available so what i did is:
2 10Mohms + 1 2Mohms resistor

In series that will give you 22Mhm:. [Edit: That's not taking into account the tolerance of the resistors.]

any suggestion how would i do the scraping?

Myself, I'd probably use my pocket knife, because it's comfortable in my hand. Something else may suit you better. Say, an X-Acto or utility knife, or a razor blade, or. . .


Torben

 

[krazatchu]

Hey there...

I don't have anything to reference it too... I would be interested to see how it compares ...

The op amp is just comparing the voltage differences between its inputs..

So... as long as the reference resistor and the scraped resistor are the same, it will work (R8 & R9) ... They can be almost any value...

I used 22M ohm... but I also tried 1M ohm and lower... the lower values were not as sensitive... this might have been because of the lower resistance.. might also have been because the lower values have less surface area (not as many wraps around)...

I used an X-acto, you could try a hammer if you like...

Thanks,
Michael

 

[carl_m817]

and this is what i did..

i replaced the:
-22M ohm with just 1M ohm
-[variable resistor]940k ohm with 500k Ohm [variable resistor]

i scraped R8 using a scissor, just like cutting an ordinary paper [desperate act..LOL!) but of course i did it with care.

before running the circuit, i tested the scraped resistor's behavior with an ohm meter.
My findings:
-when i hold the scraped part, resistor value lessens
-[this is actually disgusting...sorry about this]i licked my finger then touched the scraped part & surprisingly, the reading approaches 0 fast! then slowly goes back to the original value(1m ohm)
-i tried blowing it, nothing happens

running the circuit
-same test above, same results[ohm meter readings while circuit is on]
-but the LED never lit.
-i expect, it should have lit as i gone biasing..[potentio and R8 test]
-as i vary the potentio[500k ohm] the input at pin3 of lm393 varies[as i expected]
-but when i did the test with R8, which is conected to pin2 of lm393, it never affects the input voltage for the said pin. which means, lm393 output at pin1 is just dependent on the potentiometer
--R7 prohibits high signal for the inverter[always low regardless of the opAmp output since they are parallel], causing it to always output a high resulting to an off state LED..
-but when i short pin1 and pin2 of lm393 the LED lits.,


any comment on this?
Thanks a lot!

-carl-

 

[Conradb]

The photo I attached above of 3 resistors was done with a 'stanley' or craft-knife, it is not the best suited tool, I think it's not the knife you want to worry about, and not the total resistor value much either. Basically you want the bigger ones as they are easier to handle, and clamping the resistor in a vice is not going to work well if it damages the thin film. I can suggest only one thing, have 3 or 4 to play (my first one was a failure), and scrape (not cut like if you wittle wood), but scrape like you are sandpapering, first I scraped untill the color-bands went, then I carried on evenly all round as I started to get through. Takes about 10 minutes. Do not worry if you do not expose all of the surface, since that will just improve the scale of your readings and sensitivity.

 

[Hero999]

There must be a more reliable method of removing the coating.

Have you tried paint stripper?

In any case, if you use chemicals you must wash the resistor and bake it to evapourate any traces left behind.

 

[BeeBop]

When I did this, I used a flameproof (ceramic) resistor, and it was easy; it took about one minute.
For the plastic ones, Hero's suggestion seems right on the money, but I still think it'll be more work than if you tried a ceramic.

To try and decrease the humidity, and test that way, I used a pile of silicon sacks that come packed in tools or other moisture sensitive items.