1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Car radiator as a heat exchanger

  1. Jun 20, 2007 #1
    I am working as an intern at a biodiesel plant and we are looking into setting up solar heating to reduce the amount of electricity we are using to heat up our oil during the various stages of the process. Of course, this is only one of many projects going on at the college, so the budget is naturally tight and saving money is always a plus.

    To make a long story short, I was wondering if it would be a good idea to use a used car radiator as the heat exchanger?
    Basically the radiator would be in a sealed/insulated metal tank filled with water with a hot water outlet and a cold water inlet that are connected to a hot water heater, which would be used to collect heat for later use, and the radiator would be circulating the medium (probably a glycol/water mix) that is running through the solar collectors.

    Another question related question: would the thermosyphon principle work in a situation like this (ignoring whether or not the radiator is used as a heat exchanger) or would a pump be necessary to circulate water between the hot water heater and the heat exchanger tank? If it's not needed would it be helpful?

    That was a little long, but thanks ahead of time for any help.

    D'son BioD

    Attached Files:

  2. jcsd
  3. Jun 20, 2007 #2


    User Avatar
    Gold Member

    It sounds like a great way to get a heat exchanger going at a reasonable cost. Car radiators have quite a bit of finned surface area because the thermal conductivity of air is not that great. If the radiator is immersed in water it should transfer heat quite efficiently. I would mount the radiator horizontally, so the water would pass vertically through the gaps between the coils/fins. No idea on whether convection would provide adequate circulation, but a pump between the outlet of the insulated tank and the water heater would certainly bring more cool water in contact with the radiator and improve heat transfer. It's a long time since I did any heat/mass balances on water systems, so I can't help you much there.

    Suggestion: Is there any chance you could use the glycol/water from the solar collectors in conjunction with the radiator to directly heat the biodiesel and then use the water heater to make up the shortfall? That would eliminate one source of losses from energy transfer, and should save you some energy overall.
    Last edited: Jun 20, 2007
  4. Jun 20, 2007 #3


    User Avatar
    Gold Member

    Welcome to PF, D'son. I'm a bit puzzled by the second tank. Mainly, I don't see a reason for it. I know that you mentioned storing heat, but that won't work if you're circulating cold water through the bottom of it as your diagram implies.
    Anyhow, a car radiator should work fine as long as it has enough capacity for you. I suspect that a pump will be necessary, but I'm not sure.

    edit: Dang! Turbo beat me to it again. Good idea about the horizontal mounting.
    Last edited: Jun 20, 2007
  5. Jun 20, 2007 #4


    User Avatar

    Staff: Mentor

    You absolutely will need pumps. Regarding the second tank - you want it, but you only want it on the discharge side of the heat exchanger.
  6. Jun 21, 2007 #5
    Hello All,
    Thanks for the responses. The main reason for the hot water heater in the design is for those cloudy days that we would otherwise need to use one electricity. Also, this will allow the system to be working on weekends and other days that no one is running the equipment in the shop. Thus we will have to weigh the energy loss from the extra transfer against the utility of having that back up.
    I appreciate the tip about mounting the radiator horizontally. Would it be beneficial to perhaps place the radiator lower or higher in the tank or would it not matter?
    I rearranged the cool water return in the diagram so that if we are heating the oil the cool water return won't run throught the hot water heater at all. Instead it will go straight to the exchanger tank. I figure that way our storage tank/heat source won't be directly cooled down. Likewise if it's cloudy, and the heat exchanger isn't running, that stage can be bypassed. Of course, now that I think about it, that may mean I need to move the pump, but luckily we are still in the planning phase.
    Any other suggestions/comments?

    Attached Files:

  7. Jun 21, 2007 #6


    User Avatar

    Staff: Mentor

    Good point about recirculation, but that brings up a host of other potential issues. Do you know what temperatures and heat transfer rates are involved here? It would help a lot. You don't, for example, want to be circulating water through the collector on a cloudy day, so you'll probably want that 3-way isolation valve for it. But now your pump is in the wrong place... You probably want it on the return line, before the 3-way valve.
  8. Jun 21, 2007 #7


    User Avatar
    Science Advisor
    Homework Helper

    Regarding the original question: It depends on how good the heat exchange needs to be, how much flow you want to handle, and how much the radiator will cost compared to alternatives like soft copper pipe.

    You're probably already aware of these, but:

    Whether it's worthwhile obviously depends on geometry and so on, but it's probably going to be more efficient to have the radiator near the bottom of the cold water tank since the water will form a thermal gradient on its own.

    It's possible to squeeze more heat out of the radiator if hot and cold fluids are flowing in opposite directions, and by having more water flow close to it.

    Regarding the thermosiphon: solar panels are effective radiators at night, and, if they're raised, they can dump heat very quickly. You'll probably want a pump for the solar panels.

    Insulating the pipes running from the panels to the tank is likely to be worthwhile.
  9. Jun 26, 2007 #8
    Two Solutions! (hopefully)

    Hello All,
    Thank you for your help. I have done a little outside research and looked around at different models for systems similiar to the one we are planning to build, so for anyone who is following this thread I have attached two new possible designs. One with a separate heat exchanger and one with a combo heat exchanger/water storage tank. Which one is used will probably come down to whether or not we can find a large enough storage tank that is safe for the temperatures we are working at and able to have a radiator or other heat exchanger installed inside. I have to say the lack of extra plumbing by putting the exchanger inside the storage tank looks significant!

    Also, if for some reason you are planning to replicate either of these layouts, please note that expansion tanks maybe/definitely are needed on the different loops to deal with the heat inside of the closed systems.

    Brief explanation of the diagrams: The black loop will be insulated water pipes; the orange loop will be insulated pipes with a mix of water and gylcol antifreeze; the dashed gray lines represent wires; everything else should be labeled clearly enough. The PV panel next to the solar collectors will insure that the pump is running only when the sun is shining on the collectors to prevent unnecessary use. In the first diagram the wire to the second pump will need a switch to let it run in the same manner as the first pump on days when no one is monitoring the system. If PV panels are not readily available, the system could be run off of regular AC-grid power by replacing the PV cells with photo-resistor circuit used as an on/off switch. The thermostat control will ensure that the pump only runs when the collectors are hotter than the exchanger; otherwise, running the pump will have the opposite effect as desired.

    Thanks again!

    Attached Files:

  10. Jun 26, 2007 #9


    User Avatar
    Gold Member

    They're both nice designs, but I definitely like the second one best. In fact, I was going to suggest using the heater as the storage tank. Specifically, though, I can't see a reason for the oil tank bypass circuit in the first one. Why would you need that if you can just shut the pumps off?Using a PV loop is a great idea, either as the power source or as a comparitor switch. The only things that I can suggest are to keep your lines as short as possible to minimize losses, and to leave the parts of the pipes that are inside the tanks uninsulated to maximize heat transfer to the fluid. I notice that your diagrams show them insulated right up to the radiator (and presumable to the heat exchanger in the oil tank).
  11. Jun 26, 2007 #10


    User Avatar

    Staff: Mentor

    Without standard sybols, it is difficult for me to tell what is going on in those drawings. In places, you have flow arrows opposing each other and it also looks like you have your oil and water mixing.
  12. Jun 26, 2007 #11
    responses and clarifications

    Sorry about the confusion in the first diagram. The "flow arrows" on the white block arrows aren't really necessary. I am fairly certain that we are going to use the second diagram rather than the first. Also, just to clarify we will never (knock on wood) have oil and water mixing.

    One last thing: the diagrams are a little easier to read if you open them up in a program rather than just clicking on the link (at least on my screen).

    Thanks russ...

    To Danger:
    Once again my diagram is a little misleading, but take it only as a diagram and not a set of blue-prints. I agree the pipes in the tank should definitely be uninsulated. Thanks for making sure though.

    To All:
    After doing some price checking...if we can swing it we will probably use car radiators as all of the exchangers rather than coils of copper. It seems like it will be cheaper, especially if we can "recover" the radiators from junked cars.
  13. Jun 26, 2007 #12


    User Avatar
    Gold Member

    One other thought here, but I think that it's irrelevant in this case. For performance vehicles, a cross-flow rad is preferable to a top/bottom tank design. I can't see that it would apply to an immersed heat exchanger, though.
    If you're snapping up parts at a scrap yard, I'd concentrate on getting heavy-duty units from trucks or motorhomes. They have more core area.
  14. Jul 10, 2007 #13

    Without being an expert, I just want to mention this:

    As mentioned previously, the reason for all those fins on a car radiator is the low thermal conductivity of air, which then requires a much greater surface area to transfer the same amount of heat as the water does inside the pipes. In most fluid-to-fluid exchangers fins aren't used, since the thermal conductivity of fluids (notably water) are typically much higher than gases (such as air).

    Since you have apparently found that car radiators are cheaper than plain copper pipes, I would suggest that you consider removing some, if not all, the fins from the radiators - Assuming this is at all possible without compromising the pipes running between them. Since you are transferring heat between two fluids, this should provide a better exchange rate, since the outside fluid should then be able to pass through the radiator more easily and thus quicker, and without trapping the water near the pipes, where the most efficient exchange of heat occurs. Some experimentation might be necessary, to decide how much of the fins to remove.

    Have you looked into the possibility of using domestic radiators instead of car radiators? Since these are designed to exchange only part of the heat in the circulating water to the room air (so as to save some heat for the other radiators in the house), they might work better in a fluid-to-fluid situation. The idea would be to get hold of used or scrapped radiators, just like you consider with the car radiators.

    Good luck with your project.


  15. Nov 16, 2011 #14
    Hi guys I set up up a radiator in a insulated box.double glazed with glass. vacuumed between the glazing.sealed the box pressurized the box 1 bar. I figured if heat won't transfer in a vacuum then it will transfer better with pressure where I need it most. Around the radiator. any info or ideas on this would be apreciated thanks
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook