Announcement

Collapse
No announcement yet.

heat sink fins

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

  • heat sink fins

    A design I'm considering for a heat sink would have a section of aluminum tubing being fitted with multiple circular fins. The fins would start out washer shaped, then have the inner hole formed to create a bit of a lip, which might be around 1/8 inch high or so. These would then get pressed onto the tubing, forming a stack of fins. An aluminum plug is pressed into one end of the tubing, and the heat-producing device is then mounted to that end. The other end, being open, allows for some electronics to be included inside the tubing, also taking advantage of the heat-dissipating capability of the structure.

    Question I have is whether I can use the typical aluminum sheet material that you'd find at a regular metal shop, or whether a specific alloy should be used to allow the forming of the lip to take place without cracking the aluminum.

    Thickness of the material- well I've considered using flashing material as it's readily available, easy to handle as a roll, and is about the right thickness for what I have in mind. That's fairly thin, which would make it easier to form the lip.

    I can set up my rotary shear to cut out the discs to begin with- cutting the inner hole would be more problematic, but perhaps that can become part of the lip-producing operation. I'd machine up a form on the lathe both to mount the disc on, and to roll the lip to the desired finished ID. Obviously I want this operation to go quickly since there would be a lot of these circular fins to be made.

    any ideas?
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

  • #2
    If you don't use a fairly soft ductile aluminium, you will have to anneal or cracks are guaranteed.
    Some experiments with the interference fit would be needed, also the fit will be looser, the further the fins have to be pushed along the tube.
    At the Helicopter museum, we have a bloody great 9 cylinder radial engine using stacked fins on the cylinder barrels exactly as you describe. I forget the make, might be a Wright Cyclone.
    Last edited by old mart; 09-02-2017, 05:47 PM.

    Comment


    • #3
      Assembled parts usually do not have the lowest thermal resistance... every interface has some airspace that does not conduct. If this is not critical stuff, it can work fine.

      usually assembled heatsinks are brazed, or in some cases use a thermal epoxy plus a press fit. Can be pretty good that way.

      A fair number of baseboard heaters I have seen use punched fins with a center hole that is "lipped-over" to form a short cylindrical contact area on the fin, to carry heat to the fin from the heater rod. They work fine too.


      I'd be surprised if the cylinder barrels are stacked fins, usually those are steel with the fins cut right in the metal, and the metal is pretty thin.
      1601

      Keep eye on ball.
      Hashim Khan

      Comment


      • #4
        The discharge tube on our Quincy compressor is built just like you propose. Unless you really want to make it, you might try sourcing some of that.

        Comment


        • #5
          You don't give any idea of scale, but Laminova make pure aluminium intercooler cores in the shape you mention.
          Peter - novice home machinist, modern motorcycle enthusiast.

          Denford Viceroy 280 Synchro (11 x 24)
          Herbert 0V adapted to R8 by 'Sir John'.
          Monarch 10EE 1942

          Comment


          • #6
            I'm looking at about 1-1/4 ID and 1-1/2 OD tubing to begin with. Length would end up being about the same as the OD. That probably means about 6 fin/discs each, with the last one being a removable end cover which would be thicker and not pressed on. I'd most likely be using JB weld as a lube during the assembly process, and of course it would be the bonding agent as well. Since there would be about 40 of these to be made, I'd be doing it in stages- prep the tubing, prep the fins, prep a fixture to use in the press, then mix up a quantity of JB and have a field day at the press.

            I looked at the idea of using square tubing and square fins. Easier to cut to shape, just squares of aluminum, but probably more difficult to remove the center portion. The lips could be formed easily in this case because they would be straight bends, and I can see how to make a custom bender to allow this. I don't think I would get as good a contact between the fin plate and the square tubing, but it may be good enough- particularly with the JB as a filler and bonder. And I also won't get as good a contact with a pressed-in insert that the heat would be coming from. Using round tubing makes these two things easier to achieve.

            Using square tubing there would be no need for a lathe operation, but the rest of the forming would probably take more time. Using round tubing- and doing these on the lathe-I'd set up cutter to remove the center section of the disc, then guide a roller bearing to spin the lip. I'd be making about 200 of these, and it would be nice if I could produce the center hole and lip in a few hours one evening. Tedious to do, but one per minute I could handle. That seems doable.
            I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

            Comment


            • #7
              Aluminum fins are installed on copper pipe by pulling an expander plug through the tube to expand it into the fins.

              Comment


              • #8
                Forget about jb weld - that will just be an "insulator". Compressor inter-coolers are built like you want and it's just a hard press fit of a partially formed punched center. I suspect the fins literally finish forming themselves over the tube when pressed. Make a simple punch and die to form the fins from say 16awg and use that in your press. You could do it for square also.

                You could just use heat on the fins (convection oven would be perfect) and drop them on the tube.


                As for "which alloy" read this https://www.clintonaluminum.com/whic...oy-bends-best/

                The exterior formed panels I use retail store front are typically 5052. This is one of the main sheet alloys used for marine work so look for those suppliers.


                You could just make it all out of copper and just solder loose-fit fins on.
                Last edited by lakeside53; 09-02-2017, 10:10 PM.

                Comment


                • #9
                  According to what little research I've done, the common extruded heat sinks you see are often 6061 or 6063. These don't transmit heat as well as other alloys but they're commonly available and machinable. I'm not an engineer though so don't take my word for it...

                  There's also a whole world of math involved in how to design a proper heat sink. I just erred to the biggest one I could find that fit the application and figured more surface area would dissipate more heat.

                  Sent from my SM-G950U using Tapatalk

                  Comment


                  • #10
                    Why not start with a cube of aluminum, and cut deep parallel fins on one side with a slotting saw? This would leave a mounting slab in perfect thermal contact with the fins.
                    Allan Ostling

                    Phoenix, Arizona

                    Comment


                    • #11
                      Originally posted by LibbyHillBrewer View Post
                      There's also a whole world of math involved in how to design a proper heat sink.
                      This wiki article shows the ordinary differential equations involved: https://en.wikipedia.org/wiki/Fin_(extended_surface). They have close-form solutions. All you have to do is plug in the numbers for your parameters.

                      Note the illustration, showing the difference in temperature distribution for low-efficiency and high-efficiency fins.
                      Allan Ostling

                      Phoenix, Arizona

                      Comment


                      • #12
                        use 6061
                        smooth the inside edges after the hole is punched, (scotch brite drum) prior to forming.

                        Comment


                        • #13
                          After my last post, I started thinking about punching the holes. I don't have a punch and die of that size, but I could make or obtain such a set. What puts me off punching is the fact that you either have to pull the punch back through the hole (which means more complex fixturing) or push it right through. Pushing it through means remounting the punch each time- and having a die that's deep enough to allow for that, and potentially an alignment problem.

                          As far as forming the lip, it could be pressed also. The piece that you shove through to bend the material over and form the size of the hole would probably also have to go right through and be reclaimed and remounted for the next hole to be done. The actual forming of the hole and lip would be fairly quick- the reprep for the next hole might be a little time-consuming.

                          I don't know that the use of JB would interfere with the heat-transfer capability. I feel that it would act as a filler for the microscopic pits and potentially improve the heat transfer. It's going to be mostly squeezed away during the process of mounting the fin anyway, and it would aid in setting the fin into place- not that you'd need that, but it seems worthwhile to ease the process.

                          As far as the heat transfer capability, all that math boggles my mind, but I seem to have some innate ability to 'see' what is going to be required in terms of surface area, air flow, heat flow and initial heat spreading. One thing I like to do when using a part that needs heat removed from it, like a power semiconductor, is to mount the device directly to a copper piece, then insulate that copper piece from the heatsink. As I see it the device benefits from having an initial 'high speed' of heat removal from the device, which is helped by not using an insulator between the device and the surface it's mounted on. Once the heat is able to spread over a larger area it is more able to dissipate through an insulator and into a normal heat sink, with a minimal thermal gradient. I don't think I can prove that it works, but it makes sense to me.

                          At any rate, these heat sinks I'm wanting to make serve three functions- one is to keep the surface where the device is mounted as cool as practical, another is to be visually appealing, and the last is to provide a sealed cavity where a bit of circuitry can reside. Otherwise I'd probably be far better off just buying heat sinks and fabricating a case.
                          Last edited by darryl; 09-03-2017, 12:28 AM.
                          I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

                          Comment


                          • #14
                            Originally posted by darryl View Post
                            After my last post, I started thinking about punching the holes. I don't have a punch and die of that size, but I could make or obtain such a set. What puts me off punching is the fact that you either have to pull the punch back through the hole (which means more complex fixturing) or push it right through. Pushing it through means remounting the punch each time- and having a die that's deep enough to allow for that, and potentially an alignment problem.

                            As far as forming the lip, it could be pressed also. The piece that you shove through to bend the material over and form the size of the hole would probably also have to go right through and be reclaimed and remounted for the next hole to be done. The actual forming of the hole and lip would be fairly quick- the reprep for the next hole might be a little time-consuming.

                            I don't know that the use of JB would interfere with the heat-transfer capability. I feel that it would act as a filler for the microscopic pits and potentially improve the heat transfer. It's going to be mostly squeezed away during the process of mounting the fin anyway, and it would aid in setting the fin into place- not that you'd need that, but it seems worthwhile to ease the process.
                            For crissake

                            Look into "dimpling dies".

                            It's not rocket surgery!

                            Comment


                            • #15
                              Dimpling die- that looks like a good way to go. It might be the best way to get full thermal contact as there's enough give in the final dimple that it should form an intimate contact with the tube as it's placed. I would avoid having to machine the OD of the tube, which I might have to do if I heat-shrink the fins on.

                              But I also want to give the heat shrink idea some thought. If you lower each heated disc using a jig that holds it and keeps it level, it should drop onto the tube easily. Use spacers between fins and they should stack up nicely. If you loaded all the fins and spacers fairly quickly you would get them all on before the tube heated up enough to matter. I could bore the stack of fins in maybe two batches, so not much time lost in this prep- though I agree the bore size would require some precision. Assembly would go quickly, though each disc would still have to be punched first before the boring operation.

                              Seems that any way I do this there's at least one operation that takes a sizable amount of time. Whether it's taking apart and reassembling the punch for each hole, or running the press up and down for each hole, or feeding the tubes one by one through the lathe for a smoothing and sizing operation-plus boring the stacks of fins-

                              Next working day I'm going to see if I can borrow a dimpling die that will work for me.
                              I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

                              Comment

                              Working...
                              X