DIY Chiller

[cheeseboy]

Hi, I want to DIY an aquarium chiller for a small aquarium (30 gallons). I remember seeing someone that actually drilled 2 holes and put special flexible PVC tubing in the fridge, submerged in water. This setup is much bigger than any nanosized chiller that is commercially available and is inefficient.

Now what I want to do is use the cooling cooling coils directly. I think I would like to make a box out of acrylic with the cooling coils inside. This box would be filled with anti-freeze or possibly tap water moving with a power head to stop freezing. Then I would have a flexible PVC tubing containing aquarium water snaked through the cooling coils. Water would constantly be pumped through this tube. The cooling coils could be controlled with a one stage temperature controller. The only thing I am concerned about is the strength of flexible PVC because if anti-freeze gets in to my aquarium it could be bad.

Any Ideas on how I can improve my design?


Example:

[Criminal#58369]

Sounds interesting, yes it would be bad if it into the tank

[VINH_XUAN]

Hi Cheeseboy, I'm working on one using the flex pvc tube. Visit the USplastic website, there'r some tubes rated to handle -70F to 140 F.
My 18X18X18 fridge can housing 60ft coil of 1/2" dia, 1/16" wall, run by a Mag 5.
Good luck with your DIY.

[kgross]

The problem is that your small fridge does not have enough horsepower to remove much heat. A fridge relies on insulation to keep the stuff cold so it only has to cool it down when the door is opened.

Now this will work, either setup will work about the same, the problem is that you will burn up the fridge soon, (since the compressor will be running non-stop) and it might not be able to cool as much as you need since it does not have a large enough compressor to begin with.

Kim

[cheeseboy]

Originally posted by kgross
The problem is that your small fridge does not have enough horsepower to remove much heat. A fridge relies on insulation to keep the stuff cold so it only has to cool it down when the door is opened.

Now this will work, either setup will work about the same, the problem is that you will burn up the fridge soon, (since the compressor will be running non-stop) and it might not be able to cool as much as you need since it does not have a large enough compressor to begin with.

Kim

No, The compressor would not be running non-stop. That's why I have the compressor rigged to a temp controller so when the tank temp gets too high the compressor will kick on.

Plus I hardly need any cooling power since I have a small tank. Look at the nano chillers HP they have like 1/15 horsepower.

[BeanAnimal]

Cheesboy, I think your missing the key point here.

A 30G tank is NOT an nano. A "dorm" fridge is capable of somewhere between 100-200 BTU/h of cooling. Lets assume you got a fairly robust size dorm fridge that can do 150 BTU/h. Your tank has 30 gallons of water, or ~250 lbs of water. So, neglecting ANY heat input. It will take the dorm fridge somewhere between 1.25 and 2.5 hours to cool the tank down 1 degree. WITH THE COMPRESSOER RUNNING NONSTOP and assuming that NONE of the heat it produces finds it's way back into the tank.

Now if the heat load of your tank can cause the temperature to rise 1 degree in ~2 hours, then the dorm fridge will not be able to keep up. The compressor will run ALL OF THE TIME (100 percent duty cycle)

Lets dig a little deeper at what looks to be a LOT better scenario. Lets say during PEAK heat times, it takes your tank 4 hours to rise 1 degree (instead of 2). This mean that compressor will have to run ~50% of the time (50 percent duty cycle) to maintain your tanks temperature during the day. This will KILL that compressor in very short order. These things are designed to run a 10%-30% duty cycle. Pushed harder thatn that they quickly overheat and die.

You are also going to have a very severe heat exchange problem and will lose efficiency in your heat exchanger and due to the fact that the typical dorm fridges not really that well insulated to begin with. YOur overloaded unit will also spew a lot of heat back into the room. Moving it to a remote location and pumpung water to it will make matters even worse as you will have heat loss in your lines and heat addition due to the the pump. Lets just assume you lose 10%. In reality your looking at 100 BUT/h of cooling and no more.

Pleae also not that buckets of water, coils of hose or any other nonsense will do nothing to improve the efficiency.

I hope this makes sense.... people want to deny the reality of the situation, but the basic operating parameters tell a different story.

[VINH_XUAN]

I looked at few different mini fridges and all have the compressor of 1hp. That sounds good to me. This morning I tried to put things together. Unfort. I destroyed a $70 fridge right at the first step is to drill 2 holes through the fridge wall start from outer walll. The hole saw cut through the hot pipe. Gone, an expensive lesson about inspecting the object before trying to do anything with it. Tommorrow I'll get another one and make holes from inside out using knife until I reach the aluminum wall to avoid the hot pipe.

[BeanAnimal]

VINH...

I will just be frank. I have spelled out exactly why the mini fridges will not work. I am not sure what in the world you are looking at, but a dorm fridge is certainly not 1HP.... that would be in the neighborhood of 3000 BTU/h (a very small window shaker.) You may be looking at the THERMAL PROTECTION RATING on the side of the compressor. That is 100% meaningless.

I have a MID size refrigerator (Much larger than a dorm model) that has a SAMSUNG MK152C-L1U compressor. The Label says "1HP thermally protected" However the nameplate on the unit says 1.1 AMPS at 115V. That is the FULL LOAD current draw of my fridge. That is 138 Watts at FULL LOAD. A FAR CRY FROM 1 HP. My MID SIZE fridge is about maybe 1/10 HP, 1/7 AT BEST!!!!

If I can find the datasheet on it I will post the specs here.

Let me say this again, the $70 mini fridge you purchases is not capable of being a chiller for anything more than a nano tank.

Honestly, I am not sure why people find the reality of this so hard to swallow. It's your money... but you are wasting it.

[theatrus]

Agree with Bean 100% here.

If you're going to do it, get a small window AC unit. They can be had for $100 easily, but will require someone with experience in servicing refrigeration systems to extend the evaporator coil to a titanium heat exchanger. A dorm fridge isn't going to cut it.

But I'd say just buy the AC unit and cool your room with the tank

[VINH_XUAN]

Thank you BeanAnimal, my bad, gezzz.... 1 hp on a mini fridge?...must be out of my mind. it's great and helpful to have you join the thread BeanAnimal. Taking your analyis into account and remembering of reading some posts with success in DIY chiller like this, I still can't convince myself not to go ahead making one to prove myself wrong. My new 55 gal reef reached 88.9 F in just 2 days running even no light but the two external pumps, creatures and sponges on the rocks either died off or whitening, and this DIY chiller is the only way I can think of to preserve some lives still left on the rock (hopefully) before getting a better way of cooling it down.
What is your suggestion for this situation BeanAnimal and all ?
Thank you in advance

[kgross]

Here is what you can do. Go purchase another dorm fridge and give it a try, luckly summer is almost over. Then when you find out it will not work, it just does not have a large enough heat pump, you can go get a much larger window ac unit and cool the room the tank is in or find a much larger heat pump on either an window ac unit, large commercial fridge or something else and build a much large working chiller, but this might take a little more work than the simple dorm fridge or purchase a chiller. Depending on the size of the room your tank is in a 6000 to 10,000 btu window ac unit could do enough to keep the tank cool, or if you can handle the humidity a few fans could do it also, but then you will have to deal with the lose of water do to evaporation. But atleast you will be able to keep stuff alive.

Kim

[BeanAnimal]

Well lets look at this way...

88.9 (lets call it 89 with no lights) is about 5 degrees above safe. Lets just assume that the lights will add another 2 degrees. You need 7 degress of pulldown on a 55 gallon tank.

Lets just say there is 50 gallons of total water. At 8.33 LBS a gallon, you have 416 punds of water. So it will take you 416*7 BTUs to pull the tank down to the safe temperature. That is 2912 BTUs. For giggles lets assume you use the dorm fridge and it is capable of 150 BTU/h. It will take that fridge 19.5 hours of NONSTOP operation to pull that tank down to the safe temp. This is WITHOUT considering the heat INPUT into the tank. You said it took 2 days to get to that temperature. So lets just assume that the heat the heat input is somewhere in the neighborhood of 2912 BTUs / 24 hours. That is 121.33 BTU/h. (honestly, it is likely much more than that because your starting temperature was likely a LOT less than 84 degrees F).

In any case, this means that the mini fridge can remove 150 BTU/h and the envronment is putting 121.33 BTU/h back into the tank. Your NET cooling is about 29 BTU/h. That means it will take 100 hours (4 days) to get that tank down to the safe temperature. In reality the tank will never cool down at all (as I mentioned your heat input is likely a LOT more than we used for our example). The dorm fridge will work it's guts out and you will not see much of anything.

As Kim mentioned... you need to look into cooling the room and evaporation. Evaporation is the most efficient form of cooling you will find. Get some fans on that tank to get the air moving over the water. The window AC will help remove the moisture from the air and facilitate more evap.

[Karl K]

Alright! Here we go with the scientific explanaiton of why these compressors die put in a way that no one should really argue!

There are 3 general types of hermetic (sealed in a can) compressors; High, medium & low temperature.

All of these compressors have no way to cool the motors other than the cool suction gas returning to the compressor which flows through the motor windings to cool them. The windings are built expecting what is called "design conditions". If the suction gas temperature is above "design conditions" the motor will not be cooled enough and it will die.

Now for the tricky part. When you run a dorm fridge (or even a full size fidge) as a chiller, it has a high heat load that is higher than "design conditions". This heats up the suction gas more than is allowed for, causing warm gas to return (too much superheat). These are medium temperature comressors (in dorm fridges, fridges, water coolers, and other equipment meant to cool items to or below freezing) and expect suction gas in the area of 0 deg F or so. Worse yet would be using a deep freezer with a low temp comressor expecting suction gas as low as -20 deg F!

The reason air conditioners work is they use high temp compressors that are designed for up to 60 deg F suction gas (about the maximum safe level for sustained operation IMO). I know the urge to build stuff is strong, but save your money and buy a small chiller.

You can figure out exactly how small of a chiller you can get away with by measuring the temp of the tank with everything off in the hottest part of the day then turn on everything that adds heat to the system (all pumps, lights, etc...) for one hour. Take the temperature again. Once you have that temp rise, multiply that by the number of gallons of water it will be cooling. Now multiply that by 8.33 .

The number you get will be the BTU/h input into your tank that needs to be removed. Multipy that by 1.25 to allow yourself an 80% duty cycle and then buy your chiller.

Numbers close enough to pick your chiller
1/15 hp = 800 btu/h
1/10 hp = 1200 btu/h
1/4 hp = 3000 btu/h
1/3 hp = 4000 btu/h
1/2 hp = 6000 btu/h
1 hp = 12000 btu/h

The compressors in these chillers are high temperature, just like an a/c unit.

Hope that clears this up once and for all.

If there was a way to make this sticky, it would sure stop a lot of these threads, eh?

[VINH_XUAN]

Thank you Kark K and Beananimal and all again. You all convinced me not try another one. This morning I turned off the pump for the skimmer then within 4-5 hrs, the temp cooled down to 85.5 F, still far from safe point. My room read 78 F, I added a fan above the tank as Kgross recommended. Now the tank temp. read 84-85 F. So most the heat come from the ruturn pump. I wonder if that much amount of heat added by the pump(s) is normal ?!!! my pumps are Blueline 40x and Japan made Iwaki 55, both are about 1yr old, pretty quiet. What is you recommendation for low heat external pump?
Thank you all

[krafty]

VINH,

I am running a MAG18 on my 120gal tank. Granted, that is a lot more water to heat up but, my tank stays (with pump only) near the ambient temp of the room.

The MAG18 would most likely be more than you need though. The pump is fairly simplistic. There is just a magnetic rod with an impeller on it. That is all the moving parts. It stays fairly cool. I believe the MAG pumps are made by a company called Danner. I found it at my LFS.

Josh

[wet thumb]

Great thread and lots of good info. I have a small air condition unit that was used in a wine cellar. We moved in and did not keep the wine cellar running but it is about the size of my fish room. I am running a 120 with a 55 gal sump and keep it cool with fans but this necessitates running a dehumidifier. Would it be less expensive to just run the air conditioner as this would work as a dehumidfier as well?

[Karl K]

It couldn't hurt to try. It is hard for me to give you a definite answer as there are so many variables to consider.

[matthewdmueller]

What about scrapping an old A/C unit and turning it into a useable drop-in or inline chiller. Is that possible, and if it is does anyone have any idea of how to do it? I am thinking that you would remove the evap coil and hook that up to some titanium tubing and a couple of other lines to get it to where you want it to go and then recharge the unit. Would that work for a drop in. I think I saw a thread that Karl K did about how to make a little canister chiller. That looked pretty niffty, how efficient is that?

[kgross]

That can work very well, the thread that you are talking about is one of the best threads on creating a chiller.

Yes you could do a drop in, but the flow through that Karl K did is better because you can control the heat transfer a little better to make sure the chiller is running at max effiency.

Kim

[Cuby2k]

I agree about the thread on using a window AC unit as the condenser section for the chiller. I plan on buying the chiller barrel portion for mine but honestly I am starting to understand why chillers for aquariums are so expensive. The hassle and time involved in trying to make something DIY work like this are probably more that t's worth.

I guess it all depends on the other demands in your life.

I thought about the dorm 'frig idea as well but when you look into the capacities as stated above it is just not a good idea. Bummer.

[perpetual98]

I remember back in the good old days where Bean would just make me so angry. Now I look forward to his posts. lol

Eric