[Patchworks]

Ok all you DIY'ers, I'm having trouble coming up with that most efficient way to exchange heat from my ballasts to flexible tubing. I want to use the 8 hot ballasts under my tank to heat the tank! They are typical square ballasts.

A few of the ideas I thought about are below:

Idea #1

Light weight square metal tubing! If I could find some small 1" by 3/8th tubing, it would be the perfect flat surface to glue to the ballasts. My problems, first how to connect the tubing to the square metal tubing, second what type of square metal tubing to use and where to get it.​

Idea #2

Use some sort of copper tubing and flatten it so it would fit flat on the ballasts. I think this would be hard to get it nice and flat so it would maximize surface area for heat transfer.​

Idea #3

Use flexible tubing and wind it back and forth under the ballasts, then screw the ballasts down tight enough to hold the tubing in place. I'm not sure you would be able to get a good tight twist on the tubing.​

Any of you guys have ideas?

g

 

[Zapins]

You could just buy a 20 foot coil of the metal tubing and keep winding the metal around the ballast until you have a nice coil with no gaps. That way you wouldn't have to flatten the tubing.

Also, I would be careful using copper tubing since copper is toxic to aquatic organisms, particularly shrimp, snails, invertebrates and some plants.

If you could find bronze tubing that would be better. Otherwise just use a thicker silicone tubing (sold at home depot in 20 feet roles). The heat transfer won't be as good, but it will be good enough especially if you make a nice thick coil around the ballast (overlapping several times) the heat should all be captured. Then just run water through it.

 

[TAB]

I think I would look into coolent rather then tank water... Then have another coil that heats the water. JIMO.

 

[andyh]

I've been working on a similar project, but easier because I'm not trying to transfer heat from a solid - that's pretty hard. I'd be careful about flattening the copper tubing - the metal thins out and you don't know if you'll get leaks a month or a year from now.

Take a look at CPVC tubing - it handles high-temperature situations really well and doesn't crush easily.

 

[Patchworks]

You could just buy a 20 foot coil of the metal tubing and keep winding the metal around the ballast until you have a nice coil with no gaps. That way you wouldn't have to flatten the tubing.

Also, I would be careful using copper tubing since copper is toxic to aquatic organisms, particularly shrimp, snails, invertebrates and some plants.

Zapins, Wow, I didn't realize copper would cause a problem. I'm using it in a 1 liter bottle for fertilizers. I wonder if this will cause a problem?

Also, I'm wondering how hard it is going to be to get a good tight fit around the ballast with metal tubing?

I think I would look into coolent rather then tank water... Then have another coil that heats the water. JIMO.

Tab, Can you give more detail on your thoughts? I'm having a hard time following this idea?

I've been working on a similar project, but easier because I'm not trying to transfer heat from a solid - that's pretty hard. I'd be careful about flattening the copper tubing - the metal thins out and you don't know if you'll get leaks a month or a year from now.

Take a look at CPVC tubing - it handles high-temperature situations really well and doesn't crush easily.

Andyh, yeah, I think it was your LED project that gave me the idea. Now, I'm stuck with how to get mine to work. I'll look into the CPVC tubing. This is similar to what I what I was thinking lastnight. Look at the post after this one and give my your ideas.

 

[Patchworks]

Ok, here is another idea I came up with lastnight when I had quite a few beers in me and my creative mind was working.

I've never opened up a ballast so I'm assuming that everything would be attached to the bottom and the rest is just a cover. Based on that assumption, I again assume that the bottom would be the hottest part of the ballast and therefore the best part to attach to.

I was thinking about taking some black flexible vinyl tubing they sell at HD and sandwich the black tubing between 2 of the ballasts and use the screw holes on the ballasts to tighten them down with wing nuts so it would squash and flatten the tubbing a little.

When that is complete, I would then have 4 individual pieces, each consisting of 2 ballasts and tubing each that I could screw down to some soft of board.

To help out even more, I could wrap each of the 4 units in some sort of straping to help keep the heat in...

IDEAS?

g

 

[intothenew]

Since you mentioned beer, I started listening.

Let's look at it this way. We use a computer heat sink and thermal glue. Some how the heat sink needs to be enclosed on all but the base. Remove the paint from the ballast in some particular location and glue with this, high thermal conductivity glue. Silver, I know, but there is no contact with the water.

We then just need to come up with a way to encase the heat sink with something we can plumb to. High temp silicon and lexan?

Do you have room for such a contraption?

I'm going to go open another beer, that first one was pretty good.

 

[intothenew]

Check this out. If you keep the system closed I would bet there are plenty of options. As has been mentioned, you may should add another sink in the tank for heat dissipation instead of recirculation of tank water.

 

[TAB]

Tab, Can you give more detail on your thoughts? I'm having a hard time following this idea?

A pic is worth a thousand words... forgive my crapy paint skill.

basicly you have tow closed systems, One tht pumps a "coolent"( Which could be water) from a bucket, around the ballest, then back to bucket. The 2nd pumps water from the tank, into a a coil indside of the 5 gal bucket and back to the tank. That way neither system every contacts each other.

Once agan sorry for the very crapy paint drawing... using a touch pad on a laptop sucks.

 

[Patchworks]

Since you mentioned beer, I started listening.

Let's look at it this way. We use a computer heat sink and thermal glue. Some how the heat sink needs to be enclosed on all but the base. Remove the paint from the ballast in some particular location and glue with this, high thermal conductivity glue. Silver, I know, but there is no contact with the water.

We then just need to come up with a way to encase the heat sink with something we can plumb to. High temp silicon and lexan?QUOTE]

Funny, you mention this. I'm am a 18 years self-employed computer tech... I thought of this idea, but couldn't figure out a way to get the heat sink to water transfer.

I don't know why I didn't think of Liquid Cooling computer systems before. I'll look into that.

g

 

[Patchworks]

A pic is worth a thousand words... forgive my crapy paint skill.

basicly you have tow closed systems, One tht pumps a "coolent"( Which could be water) from a bucket, around the ballest, then back to bucket. The 2nd pumps water from the tank, into a a coil indside of the 5 gal bucket and back to the tank. That way neither system every contacts each other.

Ok, I understand now. Why are you concerned about having the aquarium water go around the ballast? I don't see that the water would ever be contaminated? Unless your talking about using copper around the ballasts?

Also, another issue is this requires 2 pumps. I was hoping to use Thermal Siphoning to automatically cycle the water without any pump... Basically, the warm water rises which pulls in water into the intake to replace the exiting hot water!

g

 

[TAB]

The only cost effective tubing I can see you using for heat transfer that is safe and won't melt( should the pump fail. the last thing you want its to come home to an empy tank) is Ti pipe which Canot be bet buy hand and costs $$$$.


Copper on the other had is cheap,

 

[Patchworks]

Ok guys, I got to thinking about the whole water cooled computer systems idea and found some interesting things on the net.

Why these guys do is put what they call a "Water Block" on the CPU and water runs thru it.

Here is an article about making custom water blocks out of copper or aluminum! Take a look!

http://www.madshrimps.be/?action=gethowto&number=1&howtopage=78&howtoID=34

g

 

[Patchworks]

basicly you have tow closed systems, One tht pumps a "coolent"( Which could be water) from a bucket, around the ballest, then back to bucket. The 2nd pumps water from the tank, into a a coil indside of the 5 gal bucket and back to the tank. That way neither system every contacts each other.

Tad,

I got to thinking about your concept. Concidentally, the ballasts are right next to my my Fluval 303 filter. I'm wondering if I could get the coolant tubing to wrap around my Fluval. They way I could simply heat the Filter which is already flowing from the tank!!! This would comply with your closed system concept!!!

g

 

[TAB]

I don't think that plastic is a very good conductor of heat. I also don't think there is enough dwell time in the fliter to heat the water effectivly. Now if you were get an extra long imput hose and coil it inside of a 5 gal bucket that the "coolent" that would work, but The very long length of pipe would slow down your flow rate alot, which may or may not be a prob.

 

[Patchworks]

Ok, another idea!! Could I simply take the ballasts and strap them to the Fluval Filter? Do you think that would melt it?

g

 

[Patchworks]

Ok, another idea, could I simply sandwich the input or output tube between 2 of the ballasts and bolt it together?

g

 

[Zapins]

Not to throw a spanner in the works, but assuming you do manage to get the heat transfer problem ironed out, how do you plan to regulate the temperature of the tank water? If the lights are on for 8-10 hours a day and they run at 60 C or so, even if the heat transfer suffers a 50% loss of heat you would still heat the tank water to 30 C, which would kill everything in the tank.

Even if this problem is solved, how are you going to heat the tank for the other half of the day when the lights are off? If the temperature is constantly swinging each day and night it could cause fish to stress and get ich.

I think this project might be difficult to get to work without addressing these problems. On the other hand, a simple submersible heater for 30$ will last you forever and only works for about 15 minutes out of an hour (6 hours a day max), which will probably be cheaper in the long run.

 

[Patchworks]

Not to throw a spanner in the works, but assuming you do manage to get the heat transfer problem ironed out, how do you plan to regulate the temperature of the tank water? If the lights are on for 8-10 hours a day and they run at 60 C or so, even if the heat transfer suffers a 50% loss of heat you would still heat the tank water to 30 C, which would kill everything in the tank.

Even if this problem is solved, how are you going to heat the tank for the other half of the day when the lights are off? If the temperature is constantly swinging each day and night it could cause fish to stress and get ich.

I think this project might be difficult to get to work without addressing these problems. On the other hand, a simple submersible heater for 30$ will last you forever and only works for about 15 minutes out of an hour (6 hours a day max), which will probably be cheaper in the long run.

This is very true... I have a heater controller that will prevent the other heater from turning on, but I was considering putting in a bypass switch on my heater controller. I just like the DIY part of the project, plus it is "GREEN" and from my readings, about 35% of the cost of running an aquarium is from the heater. And in a 150, that can be significant.

g

 

[Patchworks]

Ok, my current idea is to run a coil thru the side of the Fluval and coil it around inside and then back out... That would surely heat the water. I would have to seal the side of the filter where the coil went thru with JBWeld or something.

After doing some reading, it is much less energy to run a small pump versus running a heater. Also, since Light is abundant in an aquarium, I thought about one of those small solar ones on ebay for $15. Put the solar unit in the canopy on the sider and the pump would only run when the lights are one.

To control the temp, I was considering a bypass solenoid that would simply turn on redirecting the water to by pass the filter coil if it got above a certain temp.

g