Announcement

Collapse
No announcement yet.

DIY Multi-reservoir Chiller

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • DIY Multi-reservoir Chiller

    Hello GK,
    I have been having some water temp issues in my systems. Especially in the cloner water has been recorded up to 90 degrees in there. NASTY NASTY.

    So I was given a stand up freezer a couple weeks ago. I have been thinking out a way to use the freezer to chill all my res' of which I have 4 including the cloner. I also still want to be able to use the freezer for food.

    Here is a general breakdown of what I was thinking. I have not started building yet but I like to get an on paper drawing before I begin. All opinions are welcome if you can think of an easier way I'm always listening.

    I will use a 10 gal tote that is 20.75 L x 14.5 W x 13 H. It will easily fit on the very bottom of the freezer leaving 2 shelves for food storage. The average temp range of a freezer is -15 to +15F but I will set it to 0F due to food safety guidelines. For my solution I will use common antifreeze which is stable down to -35F. I will use 1/2" black plastic tubing carrying the solution out to 3/8" Copper Tube wound in coils and directly in the res'.

    To control the temp of the water I have partially designed a NTC Thermistor circuit. Basically it will work with a 10K thermistor in parallel with a 10K 1% metal film resistor and will be on the resistive feedback line of an op amp. So when the temp changes so will the voltage due to resistance gain. I'll run the output to a Comparator set up to be a Schmitt trigger. When a specified voltage is achieved in correlation to a temp of ~62F the circuit will trigger. The output of the trigger will control a 120v relay which will have NO be empty and NC be 120v. The relay will control the pump which circulates the solution through the copper coils. I will have a pump for each res. There are going to be a couple more resistors and possibly a transistor to sync some current to the relay. But that's the general control circuit.

    So basically the pump will run until it hits ~62F then shut off to prevent freezing or super low water temps. When the temp warms up the trigger is gone and the pump kicks back on.

    I can get the supplies from Digikey and HD for around 150-200. I have a few 250gph pumps just sitting around so they will likely be used.
    The only concern I have is the thermal transfer rate of the coils to the water in the res'. Will a temp of 0- -5F be suitable to lower the temps of the res' enough to hit mid 60's ?

    This is as far as my brain has thought this through. I'm going to start ordering parts next week. I'll prolly have it built within 2-3 weeks. Stay tuned...Steelcity
    Technologically Declined

  • #2
    Fuckin Ninja....
    Lifes a garden, dig it....then..plant it, grow it, dry it....roll it up and smoke that shit!!!


    MOMS :
    Dabney Blue
    White Rhino
    Blue Dream
    Gorilla Glue #4

    Comment


    • #3
      Um...ok...don't have a fuckin clue what you just said but it sounds cool.
      “Many people die at twenty five and aren't buried until they are seventy five.” Benjamin Franklin

      "Take the risk of thinking for yourself. Much more happiness, truth, beauty and wisdom will come to you that way."Christopher Hitchens

      "When the world wearies and society ceases to satisfy, there is always the garden." Minnie Aumonier

      "In the early days of the Republic, it would have been unthinkable that Congress could prohibit the local cultivation, possession, and consumption of marijuana.” Supreme Court Justice Clarence Thomas, Dissenting Opinion Raich v Gonzales


      VCSMC

      Comment


      • #4
        WOW!!

        Ok, my input is:

        DO NOT USE ANTIFREEZE!!!!! Food grade propylene glycol will serve the same purpose and will not kill you if it gets into your food or kill your plants if you develop even a tiny leak in your ressy.

        The temps in the freezer will be more than sufficient to cool your ressys.

        Temps can be controlled by a ball valve far more simply. My chiller continuously pumps water as cold as 8°C through a 3/8" copper coil with about 4 "wraps" about 4" in diamater. I keep the temps from becoming too cold by closing the valve and lessening the flow of water through the coil. It is less sophisticated but effective. If (and I believe you can) keep the temp of your cooling solution at a fairly constant temp, then this is easily dialed it with some trial and error.

        One pump should be sufficient for all your cooling needs.
        reservoir->pump->distribution manifold w/valve for each output->tubing->copper cooling coil->tubing (of a larger diamater than supply)->reservoir.

        It is important that your supply lines be of a smaller diamater than that of the return lines. This way the supply side will be pressurized and your cooling fluid will be equally distributed to all of your cooling coils. The return side should have no pressure and route the fluid back to the reservoir.

        Those cooling coils will exchange heat faster than you think. One small pump should be plenty to cool your ressys.


        Retiree
        Who says you only grow when you're a kid?

        Comment


        • #5
          aaaaand another ninja..
          Lifes a garden, dig it....then..plant it, grow it, dry it....roll it up and smoke that shit!!!


          MOMS :
          Dabney Blue
          White Rhino
          Blue Dream
          Gorilla Glue #4

          Comment


          • #6
            Originally posted by SteelCity Smoker
            Hello GK,
            I have been having some water temp issues in my systems. Especially in the cloner water has been recorded up to 90 degrees in there. NASTY NASTY.

            So I was given a stand up freezer a couple weeks ago. I have been thinking out a way to use the freezer to chill all my res' of which I have 4 including the cloner. I also still want to be able to use the freezer for food.

            Here is a general breakdown of what I was thinking. I have not started building yet but I like to get an on paper drawing before I begin. All opinions are welcome if you can think of an easier way I'm always listening.

            I will use a 10 gal tote that is 20.75 L x 14.5 W x 13 H. It will easily fit on the very bottom of the freezer leaving 2 shelves for food storage. The average temp range of a freezer is -15 to +15F but I will set it to 0F due to food safety guidelines. For my solution I will use common antifreeze which is stable down to -35F. I will use 1/2" black plastic tubing carrying the solution out to 3/8" Copper Tube wound in coils and directly in the res'.

            To control the temp of the water I have partially designed a NTC Thermistor circuit. Basically it will work with a 10K thermistor in parallel with a 10K 1% metal film resistor and will be on the resistive feedback line of an op amp. So when the temp changes so will the voltage due to resistance gain. I'll run the output to a Comparator set up to be a Schmitt trigger. When a specified voltage is achieved in correlation to a temp of ~62F the circuit will trigger. The output of the trigger will control a 120v relay which will have NO be empty and NC be 120v. The relay will control the pump which circulates the solution through the copper coils. I will have a pump for each res. There are going to be a couple more resistors and possibly a transistor to sync some current to the relay. But that's the general control circuit.

            So basically the pump will run until it hits ~62F then shut off to prevent freezing or super low water temps. When the temp warms up the trigger is gone and the pump kicks back on.

            I can get the supplies from Digikey and HD for around 150-200. I have a few 250gph pumps just sitting around so they will likely be used.
            The only concern I have is the thermal transfer rate of the coils to the water in the res'. Will a temp of 0- -5F be suitable to lower the temps of the res' enough to hit mid 60's ?

            This is as far as my brain has thought this through. I'm going to start ordering parts next week. I'll prolly have it built within 2-3 weeks. Stay tuned...Steelcity

            Why are you using a schmitt trigger and a comparator? Hysteresis will cause the trigger to hold until the input voltage goes beyond your low and high thresholds. It's been a long, long time since I messed with digital controls so I may be mistaken but I don't think you need a comparator if you use the right chip for you schmitt trigger.

            Comment


            • #7
              Wow
              "you sing the low parts and brother i'll sing mine high"


              Currently growing:
              White Rhino-Nirvana
              Sage- THSeeds
              Bubba Kush- Humbolt
              Dabney blue- clone
              Cotton Candy-clone

              Comment


              • #8
                Retiree I like that idea but my res' are not the same size and as the plants suck down the water and lessen water volume and the coil would be too cold. Also at lights out my temps drop 8-15 degrees. In summer when the 1kwr's shut off the air conditioner finally get's it down to 70F lights on it's 85F due to a few thousand watt heaters.Winter when the floor is cool due to arctic weather outdoors it's not so bad. I want the system to be smart enough to keep the water at about 62F give or take a few degrees regardless of current water volume and ambient temp.

                I will look into a local supplier since it's liquid it's $ to ship I'm sure and for a 50/50 mix I'd need atleast 6 gal.

                Midnight I would use the comparator set up to be a schmitt trigger. I just have a small stock of standard ic's and already have everything to make one. LM311 with a couple 1k resistors and a 100k pot to make it adjustable. Here is the calculator I use. It does not show a pot but it goes in place of R3.

                http://hyperphysics.phy-astr.gsu.edu...c/schmitt.html

                I have designed a few phototransistor circuits using a 311 and it works really well espcially for a ttl pulse just use a pull up to 5v. I have the diagram if anyone wants to make a lights on/off sensor.The ones I have made use a usb connection to get power but if you had an old computer you could jack the power supply out of it.


                I like the critiques...Steelcity
                Last edited by SteelCity Smoker; 06-22-2011, 03:17 PM. Reason: amendment
                Technologically Declined

                Comment


                • #9
                  Originally posted by SteelCity Smoker
                  Retiree I like that idea but my res' are not the same size and as the plants suck down the water and lessen water volume and the coil would be too cold. Also at lights out my temps drop 8-15 degrees. In summer when the 1kwr's shut off the air conditioner finally get's it down to 70F lights on it's 85F due to a few thousand watt heaters.Winter when the floor is cool due to arctic weather outdoors it's not so bad. I want the system to be smart enough to keep the water at about 62F give or take a few degrees regardless of current water volume and ambient temp.

                  I will look into a local supplier since it's liquid it's $ to ship I'm sure and for a 50/50 mix I'd need atleast 6 gal.

                  Midnight I would use the comparator set up to be a schmitt trigger. I just have a small stock of standard ic's and already have everything to make one. LM311 with a couple 1k resistors and a 100k pot to make it adjustable. Here is the calculator I use. It does not show a pot but it goes in place of R3.

                  http://hyperphysics.phy-astr.gsu.edu...c/schmitt.html

                  I have designed a few phototransistor circuits using a 311 and it works really well espcially for a ttl pulse just use a pull up to 5v. I have the diagram if anyone wants to make a lights on/off sensor.The ones I have made use a usb connection to get power but if you had an old computer you could jack the power supply out of it.


                  I like the critiques...Steelcity
                  yep, that schematic will work for sure. Told you it's been a long time.

                  Comment


                  • #10
                    I think it'll be rockin' SC. I'm barely literate in low voltage speak. But you sound like you're more than competent. I do ask that you take a few pics. I can't wait to see it in action. I know you'll be pleased to have better control of your res temps. Big factor for sure in WC methods.



                    Retiree
                    Who says you only grow when you're a kid?

                    Comment


                    • #11
                      So I sat down today and drew up a few things. I think I will be fine using a Comparator and setting a reference voltage. I will get the reference voltage by putting the thermistor in water at my magic temp and metering the output. Once I have a reference voltage I'll tune the pot in on - Input (pin3). Tuning the pot in or trimming just means physically turning the Potentiometer (which is a variable resistor) ClockWise or CounterClockWise to increase or decrease the resistance which does the same to the voltage going to the - Input.

                      The + Input (pin 2) will be fed from the output of a resistive voltage divider formed by a 10k ohm 1% resistor and our NTC thermistor. Basically the thermistor changes resistance as it's temp changes. Resistance decreases with rising temp which is a character of the Negative Temperature Coefficient thermistor. As the resistance changes the output taken from between these 2 resistors changes as well. Since the thermistor acts as resistance between the voltage and it's ultimate destination of ground the voltage varies as the resistance of the thermistor varies. When the thermistor is at one extreme it's an easy path so the voltage is low seen at the input since it will bypass the input to go to ground. Apparently at the other end of the spectrum it's a very resistive path so more voltage will be seen at the input.

                      Voltage wants to go to ground as fast as possible. Instant example Lighting hits the highest object connected to the earth because air is resistive the fastest path is the highest object. The taller and more conductive the more lighting is "attracted".

                      The output of the comparator will go to a NPN transistor that can sync some current to flip the relay open and closed. The 470 ohm R2 creates enough current to open and close the transistor.
                      Depending on the state of the output being high or low from the comparator. Low is off.
                      The Diode at D1 is a spark arrestor of sorts it dampens the voltage spike so it doesn't lock the relay.






                      Here is a parts list from digikey dot com.
                      NTC Thermistor: 317-1382-ND
                      Transistor:2n3904TFCT-ND
                      Relay:OUAZ-SH-112D 405-ND * For some reason this would not come up as typed in had to drop the number into the search then click add to order.IDK
                      Comparator:497-1570-5-ND
                      Pot:490-2970-ND
                      Diode:1n4001-E3/54GICT-ND
                      R1 10K:PPC10.0KXCT-ND
                      R2 470:RNF14FTD470RCT-ND
                      Perfboard:V2010-ND aka circuit board or what you will mount all this on.
                      Enclosure:HM850-ND

                      Total is about $40-50 I had some of this stuff and am making 3 circuits so I require multiples so I'm not 100% sure but that's a good ballpark.

                      You will also need a cheap extension cord to cut up and use the female socket as the outlet and use the male portion to get power from an existing oulet rather than running a new circuit. I would put atleast a 2ft lead coming from the female socket. So buy an extension cord according to how far you need to go to get power. Inside the box make sure you tie the white wires together since they are Common ground or Neutral and we only want to switch the BLack or "hot" 120v wire. I did not label the relay contacts in the drawing since I have not tested the system yet but will do so when I have a working prototype.

                      Also you will need a power supply that is +12vdc which most ppl have laying around. Camera chargers and computer equipment usually uses 12v. If you have an old router or printer in the attic it probably has a 12v power supply. You can check by looking at the tag on the power supply it will give the output voltage and current limit. This circuit will draw about 70-80mA which most power supplies are atleast 500mA if not a few Amps.

                      I have everything on order and am rounding up the rest from HD this week (after the holiday chaos). So I'll probably update within a week or 2 with a working setup and pictures.
                      Technologically Declined

                      Comment


                      • #12
                        So is there an easy way to find where the cooling lines run through the case of this thing or do I need to do some disassembly ? I was thinking if I put a bowl of water in there while at room temp then turn it on since where the lines are will tend to run colder . Open it up and check the inside walls after about 20-30 mins to see if water is collecting/condensing in a discernible pattern. Any other ideas are welcome. I need to drill 3x 1/2" outgoing and 3x3/4" return and wiring holes in this thing and hitting an ac line somewhere is not an option. I'm sure they can be fixed but I don't want to learn how...Steelcity

                        PS
                        I got everything from HD today total was about $75 for:
                        20' "soft copper" tubing 3/8"<-- 1/2"od vinyl tubing fits over this cooling coil
                        40' of black vinyl 1/2" tubing <-- Sending lines
                        10' of " " 5/8" tubing<--Fits over 1/2" and into 3/4"
                        10 Pack of 3/8" hose clamps
                        10 Pack of 5/8" hose clamps<-- will open up to 7/8"
                        10 gal tote with lid
                        25' 3 prong extension cord

                        I have 3/4" black tubing already left over from previous endeavors. I also have 3/4" tub fittings as the termination at the chiller res. The 5/8" tubing makes a nice reducer between the 3/8" copper with 1/2" vinyl and 5/8" over that and 3/4" over that and siliconed and hose clamped. I'll take a pic when I put it together. Since this is closed loop I'm going to silicon and clamp all connections. I am going to be using antifreeze due to cost and availability but since the copper was "cheap" I"m going to make the connections outside the res for the cooling coil. In the case that it leaks it will be onto my floor not into the res.

                        I have decided since I am running three pumps I'm going to hardwire them into my box instead of having a female plug for all. It will keep the exit hole out of the fridge smaller if I can chop the end off and wire it directly to neutral and hot to the relay...
                        Last edited by SteelCity Smoker; 07-03-2011, 12:43 PM. Reason: PS
                        Technologically Declined

                        Comment


                        • #13
                          Nice work so far man! Just curious what's the power/refrigeration rating on that freezer? Wondering what the top end cooling capability is for that thing and how it handles a large volume of liquid to cool. Probably way more then you need for a couple resses.

                          Comment


                          • #14
                            It can go down to -20F cranked completely up. Since I am only trying to hit mid 60's I think it should do fine. I will let you know. The res is the only thing in there right now it's a full size 3 rack stand up freezer. I figure I will be circulating around 10-12 gallons. Only way to know is to follow this path till the bloody end...Me
                            Technologically Declined

                            Comment

                            Working...
                            X