14 Replies Latest reply on Oct 27, 2015 8:28 AM by PeteB

    Plastic enclosures incorporating a heat sink for 25A SSR


      I am looking to use 25A SSR to switch 240vac to a heater element with a standard panel mount controller in a marine environment I want to use a plastic enclosure or box without any forced ventilation in UK climate say max ambient 35C.  Can anyone recommend a suitable enclosure box and what size it needs to be. Is there such as thing as a plastic box with an alloy heat sink moulded into the back for mounting the SSR inside? Cutting a slot for a standard heatsink and sealing around it with mastic would be a bit of a botch.

      Thanks for any advice.


        • Re: Plastic enclosures incorporating a heat sink for 25A SSR

          Hi M Oyster,

          There is no magic solution that I know of. The SSRs I have seen and used all require heat sinking to a metal sink to dissipate their heat build up to the environment. Isolation inside a plastic box would necessarily block this transfer. Even if you had a sink inside the box it would still probably over heat as the air and the plastic would insulate the sink from the environment. Can you redesign and use a conventional relay or use the SSR to drive a conventional relay? The conventional relay would be a small enough load on the SSR so that it would run in free air. I have seen small postage stamp sized relays easily handle 12 amp resistive loads in sterilizers and I am sure there are relays that can handle the 25 amps you mention.


          • Re: Plastic enclosures incorporating a heat sink for 25A SSR

            I have never seen something like that.

            Would a waterproof die-cast metal box work for your application? a heatsink could be attached to the exterior without punching holes...

            There are lots of reasons why plastic isn't used in hot applications - like thermal expansion, softening, warping, fatigue and even melting, However if your plastic can stand the heat you could bolt the external heatsink to an internal metal plate using enough screws to conduct the heat from the internal plate through the plastic wall to the external heatsink. This solution is fairly easy and cosmetically it doesn't look to bad.

            • Re: Plastic enclosures incorporating a heat sink for 25A SSR

              Thanks for your quick and helpful replies. The application is switching a 3kw immersion heater and because I want a small temperature variation +-2C the switching is rather frequent say once every 5minutes is 1 million cycles per year. Small relays tend to have an electrical life of 100k cycles and I tried mercury relays and contactors the latter last about 18 months. Open to suggestions on more reliable relays but basically thats why I want to try SSRs. Die cast aluminium boxes corrode something rotten in seawater once the paint starts bubbling off.  Plastic boxes work if big enough. Douglas' suggestion is interesting but might be difficult to get a good enough thermal bond between the metal plates and the plastic in the sandwich?

              Thanks for any further ideas?!


              • Re: Plastic enclosures incorporating a heat sink for 25A SSR

                Hi Mark.


                Thanks for the link to your original question and the ensuing discussion.  I see that you've already ruled out a solution of a heatsink directly attached to the SSR (anodized aluminum HS would provide pretty good corrosion resistance...), with the SSR portion of that assembly inserted through a suitably gasketed / sealed cutout into the box.  That would be the thermally most efficient solution, but as you note, it's somewhat of a clumsy looking thing. 


                The bottom line here is that with a 12.5 / 13 amp load the typical dual SCR based SSR will generate bout 13 Watts of internal heating when it's passing the load current.  Getting that heat out of the SSR, then out of your plastic box to the outside ambient environment, (as you noted would be 35 C max), is the key to your application.  All of that is based on keeping the maximum internal SCR die temperature within the SSR below its operational maximum of 125 C.   (and of course, cooler is always better !)        FYI, the Crydom D2425 mounted on an HS301 with the HS301 exposed to 35 C max ambient air and passing 13 amps would result is at most an SCR die temp of about 80 C with the heater on 100% all the time.  


                Unless the thermal conductivity of the plastic enclosure to the ambient air is known, (I see that Douglas Wong mentions the plastic box in contact with water... but  I would assume that the plastic box is not directly in contact with water, but with the ambient air...), then it may end up being a need to test to see what your results are with a few scenarios.  


                For a typical 25 amp SSR, (like that Crydom D2425....http://www.crydom.com/en/products/catalog/s_1.pdf  ) the thermal resistance SCR junction to baseplate is 0.8 C/W.   At 13 amps then, the SCR will be approx 10-11 C warmer than the hottest temperature on the SSR baseplate.    So the bottom line for your application will be that you'll want to keep that hottest temperature below 115 C under the worst case circumstances, and as noted above, that would be with the heater running 100% of the time, the box sealed, and the surrounding ambient at the maximum you stated of 35 C.


                There is a short article I wrote a number of years ago on the Crydom website in the Tech Library, entitled "How to Verify the Proper Heatsink"...




                .. as a companion article to "Selecting a Suitable Heatsink" when the heatsinking system is a bit "unusual".      This explains what I noted in a little more detail about testing the unknown heatsinking system to see if it is adequate.


                I hope this helps,


                Pete B.

                • Re: Plastic enclosures incorporating a heat sink for 25A SSR

                  Thanks for more posts,and especially Pete B who has come up trumps as advertised!

                  In time order, there is another Mark with some good physics but not relevant because it needs to cope with peak demand which on all the time (heating from cold).

                  It has to be waterproof and cant have ventilation and is only air cooled, there are no enclosures in the CPC catalogue with built in heat sink.

                  Now combining the simple rule of thumb for SSR 1W/amp and max temperatures with some information on disapation of electrical cabinets from this link (interestingly there is no difference between plastic and painted steel (because convection not conduction) http://www.hoffmanonline.com/stream_document.aspx?rRID=233309&pRID=162533

                  Shows that the temp difference inside the box of 20C (= 55C with ambient 35C but must be in shade) 8W/ft2 I need 1.6ft2 box surface area to dissipate 13W.

                  So with the standard 3C/W HS301 sink recommended by PeteB the box size could be a relatively modest 230 x 160 x 120 mm (or 6"x 9" or a C5 envelope 1/2 A4 paper).

                  I will give that a try with a temp logger and report back later to close this thread.