Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Will Salt help cool a cars radiator?

  1. Sep 30, 2005 #1
    Today while sitting parked in my FIL's van, the one I just put a new thermostat in, but failed to really test to see if it would overheat? A couple miles down the road, it pegs the temp guage. I put the OEM of 192* instead of 160* because it wasnt passing smog, they said the computer this/that.

    So anyhow I had time to sit and think about car cooling systems, and it brings this question,

    I recall an electricity experiment in Chemistry class where a pyrex jar of water, a light bulb, and electricity, the water makes the connection but no light. Add salt and it conducts and the light comes on. (but People get shocked instantly because of something on the skin) but distilled water wont conduct that well.

    Okay, if salt makes the water a better CONDUCTOR, and the idea is to conduct heat, because its a heat exchanger, could adding salt increase the efficiency of the water and make it cool better because it appears to increase the waters conductivity?

    The salt will do what I suppose ocean water does to marine craft, but then dont they use the (salt) water to help cool the engine?
     
  2. jcsd
  3. Sep 30, 2005 #2

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Consider, that the ability to conduct electricity, also implies some level of chemical activity.


    Have you ever lived in an area that salts the roads in the winter?

    I see nothing but corrosion and troubles in the future of that cooling system. You do realize that what we refer to as "anti freeze" is more properly called "coolant", it is composed of a large, heavy molecules which are very good at doing just what you want to be done.
     
  4. Sep 30, 2005 #3

    Pengwuino

    User Avatar
    Gold Member

    On... MYTHBUSTERS!!!!!!!!!!!... they actually did something that has some relevancy here. They compared cooling a beer with different methods and it turned out that adding salt to ice made the beer cool down MUCH faster then ice alone!

    But salt water is corrosive. Put it in your cooling system and you'll quickly spring leaks and other things that involve metal corrosion.
     
  5. Sep 30, 2005 #4

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Anyone who has ever made ice cream is familiar with this, you add rock salt to the ice used to make ice cream. This is because salt water has a lower freezing point then fresh water. A brine/ ice mixture will be colder then a water/ice mixture.
     
  6. Oct 1, 2005 #5

    Pengwuino

    User Avatar
    Gold Member

    meh, never made ice cream lol.
     
  7. Oct 1, 2005 #6
    salt

    No, Im on the west coast, only heard of salting the roads and it ruins metal. Ive worked on marine craft, boats that sail in the ocean? One thing you find out is "dont never use no copper for nothing." emphasis on dont use copper. How can salt spray penetrate so much? dont know but it does.

    I suppose putting this into a radiator is trouble and I wouldnt do it or say it to anyone without studying the details.



    There is a product at the auto parts store Ive heard of thats supposed to improve the cooling coefficient by ? 20% ?

    But Im wondering if its electrical conductivity that makes the difference here, what if there was some suspension of some metal? Silver Nitrate?

    Now Ive got to search for the product or see if its still on the market
     
    Last edited: Oct 1, 2005
  8. Oct 1, 2005 #7

    russ_watters

    User Avatar

    Staff: Mentor

    Also, the ability to conduct heat and the ability to conduct electricity aren't necessarily related.
     
  9. Oct 1, 2005 #8

    Danger

    User Avatar
    Gold Member

    I would assume that there are probably dialectric effects at work as well as normal chemical interactions. A small electric current might be generated which causes electron erosion of the copper. (Not sure about the proper terminology.)
     
  10. Oct 1, 2005 #9

    Cliff_J

    User Avatar
    Science Advisor

    Salt = very bad idea.

    thermal conductivity != electrical conductivity

    Glass is a decent thermal conductor, doesn't handle electricity so well but has good light conductivity too. One physical property doesn't necessarily translate into other properties.

    Water wetter products at the autoparts stores work (in theory) by lowering the surface tension on the water to allow it to make better contact with the metal surfaces. My anecdotal experience matches some of the claims.

    You are approaching the problem incorrectly. If you are experiencing an overheating problem, the coolant is not the issue unless it is low. It may be as simple as a bad radiator cap that doesn't hold pressure and lets the coolant boil (under pressure boiling point rises) or it could be an bad EGR valve or O2 sensor or other item making the engine run lean or even something as bad as a blown head gasket.

    If its not passing smog, my first guess is EGR if the vehicle has higher mileage. But a decent mechanic would be able to diagnose the problem and not make wild guesses.
     
  11. Oct 1, 2005 #10

    The first problem is failing smog. The first check and it was super rich, too much gas. Since the motor had recently been worked on to replace an exhaust manifold that cracked, and to change valve cover gaskets on a Ford with 351W and the fuel injection that curves over, you got to un bolt the injection manifold to get to valve cover bolts, its in a Van so hard to get to some stuff... anyway long story and maybe a cable got missed? no, so changed the Ox sensor and second smog check it was close, passed at 2500 rpm but failed at idle.

    The EGR is tested by using a vacuum pump to open it at idle, it has to stall the engine; Its too rich at idle, sometimes that means the Catalytic converter has worn out, but we hooked up the computer thing the OBD-II and it said the coolant temp was 172* but it should be over 180* or the computer thinks its cold and makes it richer.

    The shop said put the OEM thermostat in and re test it. Okay, then it overheats. I called the parts store where I bought the thing to ask if it could be defective? The guy explains how the car "needs" to run at 160* now because thats how its been running, the motor got used to that condition and it adjusted itself, does that make sense? I said no, Im from more of an engineering side of things, Im not from Missouri but you would have to 'show me' because I dont attribute emotions to mechanical systems.

    The radiator cap could be bad, he said. Or the guage isnt reading right. Well I said when it got hot, it steamed out the cap and was loud. OH, the cap is bad, they cannot let steam out. I said well dont you mean they have a pressure limit? no, ever, they cant let any pressure out. Uhh I think youre mistaken there, they have a pounds rating... 18, thats for most Fords, 18 pounds (he says). So the cap has to hold until 18 lbs? yes

    I put the old thermostat back in, 160* and it runs fine. What bugs me is that Ford specified a certain thermostat for that system, so either a component of the system fails or it should work. If 192* doesnt work? something failed.

    It was a hot 100* day yesterday but that thermostat should have worked.

    My questions here are to prove or disprove the concept of thermal conductivity is proportional to electrical conductivity.

    Silicon can be used in Peltier cooling to suggest the two are equal, and I sure couldnt recall what was what; space shuttle tiles.... ? do those conduct heat or insulate heat? There was some material that could be heated with a torch red hot, then seconds later you could pick it up with your fingers

    I need a thermocouple, have copper tubing, maybe I could put copper tubing into a bucket, put a thermocouple at the far end put some regular water, then put some object heated up into the bucket, measure the water in the bucket and at the outer end of the tubing, do it a few times and note the rate of change, then do the same with adding salt to the water.

    Need 2 thermocouples. since Im a poor college student, maybe glass diodes will have to suffice on some ohms scale or something...
     
  12. Oct 3, 2005 #11

    Cliff_J

    User Avatar
    Science Advisor

    The newer computers do have a 'learning' ability and memory, but it is pretty short lived. A thermostat should just simply do its job, the computer should jet the engine within reason.

    EGR can stick or have a buildup on the diaphram and not close all the way. That is why in GA they do 2500RPM on a chassis roller in gear at 25-35MPH to load the engine and force the EGR into play. My old car failed NOx at idle when the EGR was stuck slightly open.

    Did you drill a tiny hole in the thermostat (if it didn't come with one) to ensure all the air was let out? And make sure the water pump is providing enough pressue? Getting all the air out can really be a pain, and a bad pump is a bad day unless its something built before the 70s.

    As far as the salt is concerned, electrolysis is plenty enough reason to avoid it, especially if you're using the older green antifreeze where the corrosion inhibitors break down after 2 years.

    Vans are not fun to work on at all.

    The shuttle tiles insulate heat. I've seen footage on a TV show where they are heated and then picked up by hand because they do not transfer heat back out to the hand.

    The new dex-cool antifreeze are silica free and seem to offer similar cooling performance but longer lifespans. Either antifreeze offers less cooling abiliy than plain water.
     
  13. Oct 3, 2005 #12

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    It looks to me that this thread has taken a detour down the engineering asile. Off to engineering with you!
     
  14. Oct 4, 2005 #13
    Hi Brad

    There is a strong relation between Electrical conductivity and Thermal Conductivity but, There are fluxuations in the order, for instance.

    Top Nine Order of Electrical conductivity
    Silver
    Copper
    Gold
    Aluminum
    Beryllium
    Calcium
    Magnesium
    Rhodium
    Sodium

    Top Nine Order of Thermal conductivity
    Silver
    Copper
    Gold
    Aluminum
    Calcium
    Berllyium
    Tungsten
    Magnesium
    Rhodium

    Although the first part of the two charts follow a curve pattern the pattern begins to shift in order of decending periodical but stay pretty close, There isn't any other property I know that comes closer for assuming electrical or thermal conductivity in relation to the mentioned properties with only a small margine of difference.
     
    Last edited: Oct 4, 2005
  15. Oct 5, 2005 #14
    Another way to increase engine cooling is to eliminate the Clutch Fan, You can get adapters to hook your Engine Fan directly to the pully, Clutch fans allow the Engine to heat up faster by dis-engaging the fan so its free spinning, Removing the Clutch Fan will allow the Fan cooling to be a full efficiency. Only do this if you do not have an on board computer that adjust fuel/air mixtures because cooling sensers rely on adjusting these accourding to how hot or cold the engine is.

    Adding Alumiseal which is powdered Aluminum to your radiator can help lower therms by carrying away some heat better and will seal up any potential leaks in the cooling system by filling the gaps and converting itself to Aluminum oxide which swells in molecule size in the gaps and plugs it up.

    Also, If you're still interested in the Silver powder you might make your own, Silvers cheap, Just need a Grinder, You can use mine.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Will Salt help cool a cars radiator?
  1. Cooling Towers (Replies: 2)

  2. Cooling Processors (Replies: 14)

Loading...