Water cooling using aquaponics tank

Ken Beck

Well-Known Member
Well i made a barrel aquaponics system. Its not quite fully set up with correct lighting and heating as i used a lot of scrap parts and the in laws needed something right away. Borrowed water from established trout tank with high nutrient levels and made a quick light out of 2 cheapo chinese cobs and an aluminum hard drive enclosure with everything taken out of it so its just the frame. Of course they are using the system for weed lol. I have some pics of the start and will have pics of temp setup later today. I rushed so quickly to do it that i never did step by step pics lol.
 

Ken Beck

Well-Known Member
Kinda dark underneath so it's tough to tell what's going on down there.
Well not too much going on under there, just drain from bell syphon, then the pump is down in the bottom pumping up to the grow bed. Im relying on pics from the in-laws so its an accomplishment actually getting anything to show lol. I still gotta get their final led system set up. Its pretty low on light right now the way it is sitting and the tank water is barely 70°f. The fish would be much happier around the mid 70's
 

Ken Beck

Well-Known Member
Ok, so original plan was one citi 1212 4000k gen 5 cob on a heatsink, then they wanted an entire small room they built to be lit up with one single light, so i quickly put together a system of 4 citi 1212 5000k cobs that i had on hand using a long bar water block and a radiator in the barrel to drop the heat into. They are running a bit on the high end at 300w total.i will upload a pic later today when i can get them to send a pic lol.
 

Ken Beck

Well-Known Member
Has anyone ever dealt with peltier plates? Im wondering if the difference in heat between the water block and the cobs could create power to be fed back through? I know they are only at most 30% efficient, but if i can take 30% of 150watts of waste heat, then the overall efficiency jumps up quite a bit!
 

ttystikk

Well-Known Member
Has anyone ever dealt with peltier plates? Im wondering if the difference in heat between the water block and the cobs could create power to be fed back through? I know they are only at most 30% efficient, but if i can take 30% of 150watts of waste heat, then the overall efficiency jumps up quite a bit!
Not sure what you're trying to do here?

I've heard Peltier coolers aren't worth the trouble in smaller sizes (like smaller than commercial buildings), but that is hearsay and not proof.

Can you discuss your idea in more detail?
 

Ken Beck

Well-Known Member
Not sure what you're trying to do here?

I've heard Peltier coolers aren't worth the trouble in smaller sizes (like smaller than commercial buildings), but that is hearsay and not proof.

Can you discuss your idea in more detail?
The idea is that heating the one side of the plate directly with the cobs mounted to it and mounting the other side of the plate directly to the water block. As the heat flows through the plate, it creates a current recovering somewhere between 10%-30% of the heat that goes through it converting it into electricity. Of course the peltier plates i was looking at were 12v range so they would be best of use to power all the 12v pumps directly and any excess into a small grid tie inverter. Im not expecting crazy amounts of power, but it is definitely enough to make a difference in total power consumption. For example, 300watts of power goes into the led cobs, 150watts is given off as waste heat into the peltier plate, as the heat flows through the plate to the water block, it creates somewhere between 15watts and 45watts as long as the light is operating. This is a total efficiency gain of 5%-15%
It may not sound like much, but when you run the lights anywhere between 12-18hrs per day, it adds up quick. Worst case scenario a payback of 5.5yrs at 300watts of lighting 12hrs per day at 10% efficiency. best case though at 18hrs per day at 350watts maxing out a power supply (one I'm looking at is 545watts of power capacity) is about 1yr 8months. If lets say i had a larger water block and 1000w of power to the leds with the plate in between, the return on investment is then as short as 7months. Its just a thought, but something that i at least want to test out. They aren't expensive and have many uses even if i didnt get the power expected out of it. Another option would be to use them to cool the water for the trout tank and keep cobs cool as well. Just run power to have a side on the pad cold. Sure beats the hundreds it would cost to buy just a 1/4hp chiller.
 

ttystikk

Well-Known Member
The idea is that heating the one side of the plate directly with the cobs mounted to it and mounting the other side of the plate directly to the water block. As the heat flows through the plate, it creates a current recovering somewhere between 10%-30% of the heat that goes through it converting it into electricity. Of course the peltier plates i was looking at were 12v range so they would be best of use to power all the 12v pumps directly and any excess into a small grid tie inverter. Im not expecting crazy amounts of power, but it is definitely enough to make a difference in total power consumption. For example, 300watts of power goes into the led cobs, 150watts is given off as waste heat into the peltier plate, as the heat flows through the plate to the water block, it creates somewhere between 15watts and 45watts as long as the light is operating. This is a total efficiency gain of 5%-15%
It may not sound like much, but when you run the lights anywhere between 12-18hrs per day, it adds up quick. Worst case scenario a payback of 5.5yrs at 300watts of lighting 12hrs per day at 10% efficiency. best case though at 18hrs per day at 350watts maxing out a power supply (one I'm looking at is 545watts of power capacity) is about 1yr 8months. If lets say i had a larger water block and 1000w of power to the leds with the plate in between, the return on investment is then as short as 7months. Its just a thought, but something that i at least want to test out. They aren't expensive and have many uses even if i didnt get the power expected out of it. Another option would be to use them to cool the water for the trout tank and keep cobs cool as well. Just run power to have a side on the pad cold. Sure beats the hundreds it would cost to buy just a 1/4hp chiller.
You're talking about a thermocouple. You'd have to keep the cold side cool enough for meaningful current generation. I'm really not sure that it would be effective for such small delta T.

But fuck it, no reason not to give it a shot on a small scale and see if it works.
 

nfhiggs

Well-Known Member
The idea is that heating the one side of the plate directly with the cobs mounted to it and mounting the other side of the plate directly to the water block. As the heat flows through the plate, it creates a current recovering somewhere between 10%-30% of the heat that goes through it converting it into electricity. Of course the peltier plates i was looking at were 12v range so they would be best of use to power all the 12v pumps directly and any excess into a small grid tie inverter. Im not expecting crazy amounts of power, but it is definitely enough to make a difference in total power consumption. For example, 300watts of power goes into the led cobs, 150watts is given off as waste heat into the peltier plate, as the heat flows through the plate to the water block, it creates somewhere between 15watts and 45watts as long as the light is operating. This is a total efficiency gain of 5%-15%
It may not sound like much, but when you run the lights anywhere between 12-18hrs per day, it adds up quick. Worst case scenario a payback of 5.5yrs at 300watts of lighting 12hrs per day at 10% efficiency. best case though at 18hrs per day at 350watts maxing out a power supply (one I'm looking at is 545watts of power capacity) is about 1yr 8months. If lets say i had a larger water block and 1000w of power to the leds with the plate in between, the return on investment is then as short as 7months. Its just a thought, but something that i at least want to test out. They aren't expensive and have many uses even if i didnt get the power expected out of it. Another option would be to use them to cool the water for the trout tank and keep cobs cool as well. Just run power to have a side on the pad cold. Sure beats the hundreds it would cost to buy just a 1/4hp chiller.
I get where you are going. I think though, that you will lose more in COB efficiency than you would gain in power generated. To generate any useable current, you have to have a pretty high delta-T, which means really hot cobs.

The company I used to work for used them for cooling reagent storage areas in the machine - basically a small fridge. They would cool down to 5C but used a shit ton of power to do it. If I remember right the power supply was 36v @17 amps.
 

Ken Beck

Well-Known Member
I get where you are going. I think though, that you will lose more in COB efficiency than you would gain in power generated. To generate any useable current, you have to have a pretty high delta-T, which means really hot cobs.

The company I used to work for used them for cooling reagent storage areas in the machine - basically a small fridge. They would cool down to 5C but used a shit ton of power to do it. If I remember right the power supply was 36v @17 amps.
Yea. It basically slows the transfer of heat which may ruin cobs...better test with the $2 chinese ones lol
 

Ken Beck

Well-Known Member
You're talking about a thermocouple. You'd have to keep the cold side cool enough for meaningful current generation. I'm really not sure that it would be effective for such small delta T.

But fuck it, no reason not to give it a shot on a small scale and see if it works.
Its more experiment to see how much it does, im not expecting those numbers, just would be cool if it did.
 

ttystikk

Well-Known Member
:peace: Before using peltier plates I suggest you to build up a roof with PV panels to cool your trout tank in summer.
Each square meter can produce 100 kwh / year - and your trout tank stay much cooler
without direct sun.
People forget to factor in the direct cooling effects of the shade under roof mounted solar panels.

Designing a house with them from a clean sheet could maximize the gains, plus maximize surface area of the panels vs total square footage, plus minimize installation costs, plus make it aesthetically pleasing...

And then you'll be left wondering just what you'll do with allllllllll that power!
  1. HVAC
  2. Domestic hot water
  3. Kitchen and laundry appliances
  4. Badass A/V and computer network
  5. Charge electric car
  6. No more natural gas, ever
  7. Electrify shop, lawn and garden power equipment
  8. Fully HVAC integrated indoor gardens
...all while using the utility for free 'perfect' battery backup for the price of your connection (if allowed in your area, fuck Florida's utility lobby).

 

nfhiggs

Well-Known Member
People forget to factor in the direct cooling effects of the shade under roof mounted solar panels.

Designing a house with them from a clean sheet could maximize the gains, plus maximize surface area of the panels vs total square footage, plus minimize installation costs, plus make it aesthetically pleasing...

And then you'll be left wondering just what you'll do with allllllllll that power!
  1. HVAC
  2. Domestic hot water
  3. Kitchen and laundry appliances
  4. Badass A/V and computer network
  5. Charge electric car
  6. No more natural gas, ever
  7. Electrify shop, lawn and garden power equipment
  8. Fully HVAC integrated indoor gardens
...all while using the utility for free 'perfect' battery backup for the price of your connection (if allowed in your area, fuck Florida's utility lobby).

If I ever build a house from the ground up, it will definitely have integrated solar.
 

ttystikk

Well-Known Member
Solar PV along with Solar water heating, (if that's what you are getting at) yes.
As well as simply having south facing windows for home heating. To limit heat gain during the summer, a strategically placed eave can effectively shade them. No moving parts, no controls, no pumps, no computers needed.
 

mahiluana

Well-Known Member
Solar PV along with Solar water heating
:peace: Do you know, that with watercooled PV panels happens the same
than it does with watercooled led light ?

PV panels can get very hot in summer (130°C).
Each Kelvin you can bring down the panel temperature, it produces ~0,5% more electricity.

more info i.e. here:

http://www.engineering.com/ElectronicsDesign/ElectronicsDesignArticles/ArticleID/6123/Photovoltaic-Thermal-System-Achieves-86-Efficiency.aspx

Hybrid liquid cooled panels i.e. can produce ~ 265 Wp + 888 W heat power / 1,59 m²

Only good to have, if you need really a lot of hot water in your swimming pool or water tank.

(:
 
Top