Hi all.

This is a quick test of a humidity sensor. I've had a humidity sensor for a few weeks but never tried it. Finally, I decided to give it a try today. The sensor has been hooked up to the exhaust all this time but I never actually see if it worked.

This is the sensor I have tried (link has hook-up schematic and formulas)

http://www.ohmicinstruments.com/pdf/ABS-300.pdf

The sensor is $50

This was my setup before I started:



The sensor needs to be powered by 15 volts to get warm. In my setup, the circuit with the sensor only draws a total 0.03 amps

This is the sensor hooked up in the exhaust:



The quickly made circuit:



I did not applied the math formulas to map the humidity output. Instead, I only logged the voltage output. The software is capable of applying formulas from the multimeter, however I did not set it up that way (maybe next weekend). The sensor is read by using the Omega HHM28 multimeter (artisan compatible). The sensor outputs in milivolts range from 13mv to 0mv. 13 mv would be very low humidity and 0mv very high. In the software I set it up so that the output is multiplied by 10000 so that it can be seen on the screen (but again, I could have applied the formula in the manual). The formula in the manual is a quadratic equation but only one solution makes sense.

This is the result of the first trial roast:



Notes: I left the fan ON all the time. At the end, I place the bean try blocking the flow of air but did not made much difference. The higher the humidity read, the lower the voltage. It seems like the first crack is easy to identify. At 1C I powered off the heat gun (HG on at dry end). Weight of beans was about 280 grams.

I probably have to figure out many things. Don't know its applications yet.

Cheers

Arpi wrote:
I probably have to figure out many things. Don't know its applications yet.

Cheers

It's certainly an interesting way of seeing the progression of 1st crack.

Thanks Farm.

Oops. I just realized that the meter has a 25mv range option. I had it on auto. The 25mv should give me a much better reading (next time).

Cheers

Hi Allon.

The delta humidity may be helpful in some cases. I am just exploring the possibilities right now.

This is a second roast. This time I put the meter in the 25mv range. What a huge difference! The humidity meter has now double precision (like 4.83, 4.82, 4.81 mv). The sensor does work nicely. Again, this is just the output voltage (it has not been converted to humidity).



On this roast, I decided to finish it with the minimum airflow setting of the heat gun. At the end, you can see a peak on humidity. That is when I placed the tray in the cooler (closer to the sensor)

I am happy with the results so far.

Today I tasted the roast from yesterday and it had some citrus punch (no defects). The one from today is faster and it may even have more punch. Beans are El Salvador Molino de ... (sweetmarias)

Cheers

Shouldn't most of the humidity get blown in the air around "drying", much earlier than FC.

I am confused,
Marko

Hi M.

You are right. My thinking is that maybe it gets stuck in the walls (condensation). There is a long path between the drum and the sensor. If I were to place the sensor closer, I am sure it would get a better reading.

On the other hand, at 1C it is when the beans explode (crack) because of moisture pressure buildup inside.

Maybe the sensor would need to be in a different place.

Cheers

Using this other differential method, I also got more or less the same results

My home made digital humidity meter while roasting





On the graph above, the humidity is not reversed like when reading the voltage.

the results look similar (a peak of humidity at 1C). "Sensor" placed in the same location