I have an Arduino Duemilanove clone, powered from a 9v wall transformer, DFRobot Ethernet Shield and i've connected and distributed 11 DS18B20 therms - each one is (currently) driven from a different pin on the Arduino and supplied 5v from the Arduino 5v pin.
What i'm worried about, in general and in particular because some of the sensors are outside and one is sat in snow at the moment, is what happens if there is a short on the 5v.
Is this likely to cause damage, or even worse a risk of starting a fire?
I tried to simulate it to see if it sparked and it just seemed to cut out - is this because of the inbuilt resettable fuse in the Arduino?
Would it be safe if that happened and went unnoticed for days in that shorted position without sparking?
Is there anything I can do to protect against this? - would running from battery help?
Considering the voltage there is only a small chance on fire especially due to water. A bigger problem is that you will loose the ability to do readings.
Simplest way is to put the DS18's in some thin plasticfoil (bag) and tape it tight so no water can enter. Place it in such way (opening below) that water has to crawl up to reach the contact. Water can crawl up - Capillary action - Wikipedia - but most of the times it runs down.
What I was wondering is why you use a separate pins for every sensor? (I can think of reasons but I like to hear yours)
Considering the voltage there is only a small chance on fire especially due to water. A bigger problem is that you will loose the ability to do readings.
Loosing readings is no problem at all! - it's only a bit of a hobby and the readings are not controlling anything.....it was whether there was a risk of something going wrong and burning the house down I was concerned about!
Simplest way is to put the DS18's in some thin plasticfoil (bag) and tape it tight so no water can enter.
Won't putting it in a bag affect the readings?
I've currently wrapped the cable in electricians tape which overlaps into the plastic top of the sensor a little bit but leaving the majority of it uncovered.
What I was wondering is why you use a separate pins for every sensor? (I can think of reasons but I like to hear yours)
Simply because i'm new to Arduino and I was hacking some code I found online so it was easier to stick each sensor on a different pin and read up on multiple sensor networks (and parasitic power) later.
Nah, the bag won't significantly affect temp sensor readings unless the temperature is changing very rapidly. The temperature of the bag (and the small amount of air inside) will equalize to any external temperature change in a few minutes at most.
I've taken sensors like that and just coated the entire mess in about 1/4" of epoxy, also, just leaving insulated wires leading off. The coating of epoxy is totally water tight and prevents any other mechanical failures also.. and temp readings only take a few more seconds to stabilize. the epoxy really doesn't provide that much of a thermal barrier. Since the top of your sensor is plastic (waterproof) if you're careful, you can cover the rest of the sensor and wires, but leave that plastic top exposed.
As you tinker, you're going to find epoxy is an invaluable tool. Hot melt glue is a close second. Avoid cyanoacrylate stuff ("super glue") if you can as it tends to offgas and corrode plastics and foul connectors. Duct tape, electrical tape, and Phenoseal caulk are also need-to-have adhesives and waterproofers...
Once you've got 'em waterproof, mount two next to each other out of the wind. Have a small rag leading to a cup of water on one. Read the two temps, run them against a lookup table or simple algorithm, and you'll have an evaporative relative humidity meter..
Nah, the bag won't significantly affect temp sensor readings unless the temperature is changing very rapidly. The temperature of the bag (and the small amount of air inside) will equalize to any external temperature change in a few minutes at most.
Excellent! - i'll try do that then!
How do you seal the bag?
Once you've got 'em waterproof, mount two next to each other out of the wind. Have a small rag leading to a cup of water on one. Read the two temps, run them against a looking table or simple algorithm, and you'll have an evaporative relative humidity meter..
Like was said above, make sure the "open" end of the bag faces down, and a good bit of tape or hot glue will do a decent job. It might not be sufficient for a permanent installation, but for the level of goofing off that you're doing, it should be fine. It's not like the sensors cost a fortune
The other part was just a suggestion for a project. Relative humidity is measured by comparing the temperatures of two thermometers, one wet and the other dry. As water evaporates, it cools the wet thermometer. The drier the air (the lower the humidity) the faster the water will evaporate from the wet thermometer.. thereby creating a wider temperature difference between the two (wet and dry) thermometers. The temperatures are run against a lookup table or simple algorithm, producing Relative Humidity. Given you know the temperature, that also allows you to calculate the Dew Point.
There are sensors now which do this all much more conveniently and without the water these days of course... but this method (wet/dry bulb) is the original and Gold Standard for humidity measurement
That is the ultimate effect of any measurement. A measurement disturbs the system it is trying to measure. But for these sorts of levels the disturbance is negligible.
That is the ultimate effect of any measurement. A measurement disturbs the system it is trying to measure. But for these sorts of levels the disturbance is negligible.
Or as I say in my signature block, "Measurement changes
behavior"
