See this excellent EEVblog video. 2257 Joules/g
In humid areas, most of the energy used in air conditioning goes to condense out the water vapor.
See this excellent EEVblog video. 2257 Joules/g
In humid areas, most of the energy used in air conditioning goes to condense out the water vapor.
Well, buying electronic products from Amazon can be somewhat of a gamble. Many times the descriptions and specifications are translations from other languages and they don't come out quite right or they are somewhat misleading.
I don't recommend buying anything from Amazon unless it has a product/user's manual with detailed instructions and specifications or the give manufacturer's part numbers so you can find the relevant information.
I'm located in the US so I buy products from places like Adafruit, Sparkfun, Jameco, MPJA, Digi-key, Mouser. However, I do sometimes take a chance with Amazon but with the expectation that it may be lost money.
but the fan is facing in the opposite direction, so assuming that the water is condensing on a surface near/below the cool side of the peltier module, it should work, right? it seems like all peltier modules need a fan for cooling the hot side.
The fans in the picture of the device you selected are BOTH blowing onto the heat sinks. One heat sink is the hot side the other heat sink is the cold side. How will dew form on the cold side heat sink while the fan is blowing air that is to be cooled? Even if the fan on the cool side is moving cooled air away from that heat sink, that air is continuously being replace by warmer air.
You could remove the fan and the heat sink from the cold side and see if the device will get cold enough for your project.
Peltier modules are heat pumps, so for cooling, you need to remove the pumped heat from the hot side of the Peltier.
A second fan on the cold side might be used to circulate air within a "cold box", to even out temperature variations within the box.
Think carefully about whether that second fan would be useful for condensing water vapor from the atmosphere.
Hi, sorry for the extremely delayed reply. So here's my dilemma: I've purchased this Peltier cooler from Amazon (I know it isn't the best place to buy these things, but it arrived in decent condition with all the parts installed). However, I don't have a 12 V 5 amp battery and can't invest in one, so I have to the Peltier into the outlet.
I'm planning to use a switching connector that Adafruit recommended, and then use male-female connectors to connect the Peltier to the connector?
Do you have any thoughts or advice? Thanks!
A 5A power supply may not be able to supply enough current for the Peltier and if so will simply overheat and shut down.
Power supplies must be rated for higher current than the project is expected to draw. Typical choices are 1.5 to 2X higher (7.5 to 10A, to be on the safe side).
You can find suitable 12V power supplies a thrift shops for very little money (look for video game power supplies, like XBox).
Adafruit says 72W max so that would be 6A and I don't think that includes the Fan.
I would look for a 6.5A or more supply.
What happened to the battery and solar panel?
If you are looking for something cheap, I would go with this supply.
However it does not have a power cord. Maybe you could cut one off of some old appliance/computer
https://www.amazon.com/LEDMO-Switching-Converter-Adapter-Transformer/dp/B01E6RMASC/?th=1
Solar panels are way too expensive and don't produce enough power to power the Peltier, so I scrapped that idea after looking into it. I don't currently posesss a battery large enough for this project, and I don't want to buy the 100 dollar ones online.
You could use a cheap motorcycle or car battery purchased from a local store but you would also need a batery charger.
Sorry if this sounds like a stupid question, but if the required power for the Peltier module is 12 v 5 amps, and the switching converter you recommended provides 12 V 10 amps, would that cause any problems?
No
The peltier will use 5A of the 10A, so you have 5 more amps (A) for powering other things.
Make sense?
Did you read the Adafruit write-up and look at the Adafruit video (go to about 7:28)?
Simply connect to the...power supply and you will see frost appear on the aluminum plate within a minute.
Since you probably do not want to create frost and then melt it to get water, you will need to control the temperature of the Peltier's cold side so that it is at or slightly below the dew point.
To do that, you will need a temperature sensor that touches the Peltier's cold plate connected to your Arduino and some means of controlling the power to the Peltier device with your Arduino (PWM through a mosfet comes to mind, though reportedly Peltiers don't like PWM).
Also, as several commenters have already pointed out, this is not a very efficient way to create drinking water, much less water for "irrigation." In addition to everything previously mentioned, take a look at a psychometric chart:
At 68 deg F (green line) & 50% RH (blue line) the psychrometric chart shows that there are 51 grains of H20 in one lb of dry air (red line).
Since...
7000 grains = one pound
and
one pint of water weighs 1.0432 lb = 7300 grains
...to get just one pint of water, would need to completely extract all water from:
7300 grains / (51 grains / lb air) = 143 lb of air
And since air density is approximately 0.0765 lb / cu ft, that means you'd need to completely extract all water from:
143 lb / (0.0765 lb / ft^3) = 1,871 cubic feet of air.
But considering the relatively small surface area of your heat exchanger, I'd guess you'll only get a tiny fraction of the water out of the air that passes through the exchanger.
If it gets 5% out, then you'd need to process 1,871 ft^3 / 0.05 = 37,400 cu ft.
And if your fan moves 15 cfm (a guess), then it would take 40 hours to get just one pint (two cups) of water.
Edit to add: the Adafruit peltier you bought does not have the cold-side fan and heat exchanger mentioned above. So it will take even longer to produce a pint. I wouldn't be surprised if it takes 400 hours!
Whew.
And if you create and then melt frost, it will take much longer...
PS: if you know the temperature and relative humidity (RH) of the ambient air (which you can obtain via other sensors connected to your Arduino), you can use the psychrometric chart to determine the dew point. There are also formulas available on the internet for calculating dew point from temperature and RH, so your Arduino can calculate it, too.
And at about $3/gallon, not terribly economic.
A couple final questions about the power switcher you recommended – would I be able to plug the wires of the Peltier module directly into the power switcher? The Peltier module link is here: link. As you can see from the diagram, the wires are bare (no connectors).
One last thing – I know I'll need a temperature sensor to regulate the coldness of the Peltier, since obviously I don't want frost. Any general advice on that?
Thank you so much for your help and your lightning-fast responses, they are extremely helpful!
Hi DaveEvans, to be honest I was pretty tentative about starting this project because after reading about Peltier modules it became really obvious that they aren't that efficient. But I live near the coast so it gets pretty cold at night (40-50 degrees F) and has 80-90% rel humidity, so I think it should be decently feasible on a good day.... right?
Thanks for your advice on the tempereature monitoring by the way, that's definitely one of the more challenging parts of the projects just because I'm a completely newbie to Arduino. If you guys know any code online that serves a similar purpose, please let me know
The chart I posted will answer your question. Suggest you take a stab at it. At 50 deg F and 90% RH, how many grains of moisture are in a pound of air?
For starters, look at how I did my example, with 68 deg F and 50% RH, which gives about 51 grains per pound. Then follow the same approach, but at 50 deg F and 90% RH.
I would use a DS18B20 for temperature monitoring. There are many tutorials for it on the web. RandomNerd has some decent ones.
Yes, it has screw terminals, it is designed for bare wires.
temperature sensor
You said you had a starter kit. Did you check to see if it had a temperature sensor?
If not, the TMP36 is cheap. The DS18B20 is another.
You could also use a thermistor but if you buy one make sure it has a datasheet with a temperature table or all of the thermistor coefficients.
You will also need to control the peltier. How will you do that?
I've been thinking a lot of Peltier control and one video I watched said that MOSFET transistors can be used to regulate current. I'm looking at this example, and it seems like you would be able to manually control the current running to the motor (or Peltier, in my case) through a similar method.
I don't want to get too ambitious, especially since I'm anticipating a lot of troubleshooting issues, but do you think there's any way to automate this process? I have a temperature and humidity sensor (I finally found it! It's the DHT11) and, I'm thinking there has to be a way to get the dew point readings from that sensor and use it to regulate the MOSFET. Hopefully that makes sense, thanks for your advice!