Heat exchanger design for 8ft^3 dehydration chamber & lots of electronics controls

Discussion in 'The Projects Forum' started by RogueRose, Jul 8, 2016.

  1. RogueRose

    Thread Starter Member

    Oct 10, 2014
    189
    4
    I appologize that a large part of this project isn't electronics related but I need some input on a design and I know there are a lot of smart people here who may have some insight into this.

    Project is an 8ft^3 (2' x 2' x 2') plastic box that will be used to remove moisture for things like fruits, veggies, meats, etc. There is going to be some kind of adjustable heating element (20 - 600 watts would be ideal). Once the air is heated it moves through the chamber absorbing moisture. The air then needs to have the moisture removed and the cycle repeated.

    I have come up with a few options, but I'll explain my most preferred/ideal.

    The moist hot air passes through heat absorber A (side A) - as much heat as possible. air then passes through something like a dehumidifier/condenser which is a 1.5" ID PVC pipe surrounding a 1" OD thin walled aluminum tube/pipe (cold water/methanol/ethanol running inside/between the 1.5" ID wall and 1" OD wall - which chills the aluminum pipe) - inside the 1" pipe will be another chilled pipe (copper probably) of diameter yet to be determined, maybe .5"ID. So the air passing through the 1" pipe (which has less volume b/c of the inner chilled pipe) is chilled on the outside and inside of the pipe. Water drips down the 1" aluminum pipe as the temp drops and it can't hold as much moisture. Cooling the fluids is an issue to tackle later.

    After the condenser tube, air passes up or down a PVC pipe (3-4" ID) filled with a strong dessicant like CaCl2 chunks/pellets) and the water will drip down into a drain. This is the last moisture removal stage. Next the air enters side B of the heat exchanger (the first step) and extracts whatever heat is possible from this before entering the heating element (tube or whatever can be made) where whatever wattage is used to bring it to the target temp. Choice of heating element is of concern and suggestions are welcome (I thought nichrome or an old hair dryer..?)

    There will be a fan(s) for air circulation in the chamber but there also need to be some kind of way to blow air through the water removal steps.

    Now my biggest dillema as to what to do about a heat exchanger (or if it is needed) is the design. For those who have worked with Intel CPU's since the 775's most use a round heatsink but the thickness can vary from 15mm to 46mm, so there is quite a difference. I was thinking of stacking many sinks on top of each other (without fans) inside a tight fitting PVC pipe. A hole would be drilled through the center and a copper tubing "nub" would link each sink basically creating one continuous pipe/tube through the sinks drilling a hole through the center of the sink and soldering/linking with copper tubing. Maybe 8 sinks will be stacked on the "side A" and 8 sinks on "side B" and a pump will circulate water through these (IDK what flow rate, will be determined through trial).

    Here are some images of the heatsinks. Some of the fins radiate out in a straight line, others in somewhat of a spiral - any thoughts on if that matters or if alternating them may cause more contact with the fins?

    The problem with using these is the price at $10-15 per heatsink (have to get fans as well). I just need the metal. I'm wondering if there are any other products that use heatsinks like this where I could buy them in 2-12" sections (I've looked on Alibaba and found some but quantites of like 500 were needed to order)

    Stock Intel 775 heatsinks: Top
    [​IMG]

    Stock Intel heatsinks: Bottom
    [​IMG]

    [​IMG]
    Side View
    [​IMG]

    Similar sink:
    [​IMG]

    Intel LGA1156 Sink
    [​IMG]

    Ivy Bridge CPU
    [​IMG]

    Server Sinks
    [​IMG] [​IMG]


    Here are some other options I thought of for heat exchangers - these are usually a lot larger, but I can't put a pipe in the middle so it is a complete redesign.

    Another option - from Dell Poweredges - about $10 each
    [​IMG]

    HP Proliant Heatsink - $9.5 - 11 each (just the unit on the right)
    [​IMG]

    HP DL560 Server - similar toabove, just different angle.
    [​IMG]

    HP DL380- ~ $10 each used
    [​IMG]

    If I had to use the heat sinks for servers (4 above) then I would have to make a rectangular tube of PVC (can be done in oven and then putting in a jig/vice to make a round pipe have straight sides). The problem is with how to get fluids to contact the thick base-plates of the sinks. I could do this by cutting a hole out of the bottom of the square PVC tube/pipe, passing the thick plate down through & sealing in place with epoxy or JB weld type stuff. Under the base plates there would need to be a "U channel" of PVCmaybe 1/4" deep but as long and as wide as the tubing that encloses the fins 0 then seal the ends and add some hose-barb to pump water through it. So the shallow "U channel" will be filled with water and in contact with the base-plate of the sinks - the sinks being sealed water-tight into the "squared" PVC pipe that houses the fins. The same would be done as the round ones, a "Tube A" and "Tube B" and water circulates in a circle moving the heat.

    Wow, this got longer that I thought it would be, but this has been a major dillema b/c this can save a lot of heat energy. It would cost about $120-160 retail for these sinks (used for the servers) . I think the round ones would be easiest to make but IDK if they would be most efficient. If anyone has a source for these round sinks (from old dead computers maybe??) I'd love to hear from you.

    What do you think is the better choice for the heat exchanger - the round ones or the square-ish ones?

    As always any comments of questoins are welcome and I'm very open to suggestions.

    On the electrical side I am going to need to cobble together a lot of electronics and don't know where to start as far as using something like an Ardunio, Raspberry Pi, or other.
    Temp & humidity - probably 4 different locations
    control the heating element
    control the fans (in heat exchanger(s) and within chamber
    control the circulation pump for heat exchanger
    control the condensor pump.
    possibly a "bilge pump"
    possibly some cooling mechanism for the condensing unit.

    Thoughts and comments are greatly welcome!
     
  2. wayneh

    Expert

    Sep 9, 2010
    12,088
    3,027
    How is your project different than a commercial dehydrator like this?
    [​IMG]
    It sounds to me like you are hugely over-engineering this. Why bother to extract the water or recycle the heat? Just exhaust them and be done. Put it outside of you don't want the heat and moisture in you house, but the amounts involved are trivial. Do the math. And by the way, 600W would make it an oven, not a dehydrator.

    Those heat exchangers are all wrong for this job. They're meant to take away concentrated heat from something that would otherwise produce a large ∆T in a tiny area. And they're not meant to condense water. You want something more like a car radiator, or a heat exchanger from an old dehumidifier (I have several!). You need to move small amounts of heat but with small ∆T, requiring large surface area and low air velocity. And you need to handle condensate.

    The first place your moist air sees cooling is where most of the water will drop out, so you need to design for that. You can't cool first and then dehumidify.
     
    RogueRose likes this.
  3. RogueRose

    Thread Starter Member

    Oct 10, 2014
    189
    4
    There are specific reasons for the project build, commercial units won't suit.

    I figured a 3 stages will extract water and that is all accounted for (the first "heat absorption" stage will drop temp & cause minimal water condenstation), the condensor tube will extract a great deal of H2O while also cooling further. The final, desiccant step should reduce to near zero % humidity. Each step has a drainage mechanism.

    There is a HUGE difference in using air that has 20% humidity vs air that has 0-5% to dehydrate/desiccate something. Even though it doesn't sound like much difference, it's kind of the same when using acids for reactions where a 95% won't work but a 98% will work for the reaction - it seems like both would work, but, no. Here, one will just work much faster, some say 2-4x faster getting to the 0-5% starting humidity.

    As far as what type of "radiator" type thing to use for the heat exchanger, that was a bit of a dilemma as there are space and budget constraints. I do have 2 radiators from a dehumidifier, each are about 11" x 11" x ~1.25". I would love to use those for the heat exchangers but I'm not sure how I would design the "cage" to force air through them. If the fins faced the opposite direction the air could pass through 11" of radiator vs just eh 1.25. Any suggestions on how to make this work with these radiators would be greatly appreciated!!!!!!

    I have not determined the appropriate rate of air flow as that may change with temp and humidity levels. I have figured 1CFM may be a good starting point as that would allow the air to circulate about 8x an hour.

    This chamber may also be used for drying things which have solvents I want to recover, so with this setup, that is possible and any mixing with water can be handled later (the desiccant doesn't absorb the solvents I'm looking at using).
     
  4. wayneh

    Expert

    Sep 9, 2010
    12,088
    3,027
    Thanks for elaborating.

    I think you're going to need to tackle this in phases. Phase one is bulk volatiles removal. During this phase it can operate more-or-less like the commercial dehydrator, with the added condenser step. The input and circulating air does not need to be at unusually low humidity, and the chemical desiccant should be protected from the high moisture in this phase. This phase may last a few hours depending on the particle size.

    Phase two is desiccation. The air flow rate will be very much lower, near to zero. You won't be able to pull off any condensate and only the chemical desiccant will be acting. (In a lab, this phase is done under vacuum.) This phase may last half a day or more.

    The rate-limiting step in drying is the boundary layer that forms and tends to slow the water from reaching the surface where it can be pulled away. You can add heat, but that may degrade the product.

    Do you need the condenser to fit within your chamber? If not, you could have it as an attachment. When you're done and condensate is no longer forming, switch it out with the desiccant chamber and adjust the airflow and chamber conditions. No ambient air input, maybe a higher temp, much lower air circulation, etc.

    By the way, the desiccant chamber doesn't need a drain. There should never be free liquid appearing in there. You regenerate it by cooking it in an oven to drive off the collected moisture.
     
    RogueRose likes this.
  5. RogueRose

    Thread Starter Member

    Oct 10, 2014
    189
    4
    As for the dessicany, this depends on what is being used. Some will expand to 2-20x dry volume, others generate heat and change color and some form moisture droplets on their surface and slowly erode (as a saturates solution of dessicant and water). I am going to try the latter and in my 3 previous attempts, they have all formed pools of "solution" under the solid dessicant. There are a number of dessicants that do this and one of the more common is CaCl2 (anhydrous). The behavior of the anhydrous compound is much different than all of they hydrates from di-hexa hydrate in how it handles water vapor.

    The entire moisture extraction mechanism will be located outside of the chamber with the chamber having an air inlet and exhaust (leading to the dehydration steps). There will be a fan inside the chamber for air circulation or some other means to agitate the air.

    As you mentioned radiators before, I looked into water cooled CPU radiators that range from 120x120 to 160x160 and I think those are perfect for what I want to do with the stages, but again pricing can be an issue
     
  6. wayneh

    Expert

    Sep 9, 2010
    12,088
    3,027
    You learn something everyday around here. I worked with desiccants for years and never saw such a thing. Cool.
     
Loading...