Results 1 to 9 of 9

Thread: Calculating whether we need heat sinks on the triacs

  1. #1
    Join Date
    Feb 2008
    Location
    Warminster, PA
    Posts
    1,755
    Post Thanks / Like

    Default Calculating whether we need heat sinks on the triacs

    So, Iím changing everything over to LEDís this year and it got me thinking, do I really need the heat sinks on the triacs? No matter watt (haha, power joke), Iím going to put them on just to be sure, but I figured itís good knowledge to have. A couple of google searches later, and this is how I understand it and my explanation for how to figure it out. Please let me know if Iím mistaken.

    So, for those of you who are interested, here it isÖ.

    First, Iím using all Renards and Iím switching all of the triacs to BTA06-600cwís.

    The first number we need is the maximum heat the device can take. In the datasheet in table 3, under operating junction temperature we see that the maximum is 125 degrees Celsius.

    Now, before we turn anything on, the temperature of the triac will already be at whatever the outside temperature is. For this argument, Iím going to set it at 40 degrees Celsius, which would be a bit over 100 degrees Fahrenheit. It better not ever get that hot in PA during Christmas, but itís better to be safe. ( or to just be prepared for July 4th).

    Anyway, that means that we can allow a temperature rise in the triac of the maximum temperature Ė the outside temperature. So, 125 Ė 40 = 85 degrees Celsius.

    Now, if we look at the datasheet again, we can see in Table 6 that thereís a 60 degree rise for every Watt of power on the device. Thatís the junction to ambient listing (basically, the triac without a heatsink).

    With the maximum rise in temperature the device can handle, and the rate of temperature rise in Watts, we can get the watts by dividing the allowed temperature change by the change rate of the device. So, 85/60 which equals 1.4 Watts.

    On the datasheet again, we can look at Figure 1 and see that when the power is at 1.4 Watts, Amps is at about 1.3 or 1.4 amps.

    So, the final result is about 1.3 amps. Keep in mind though, these are numbers would put the triac on that edge of overheating so I wouldnít suggest going to even close to that number. If you go the 80 % route, that would put you slightly over 1 amp per channel.

    In the end though, Iím putting on my heat sinks.

    So, thereís my (very wordy) thought for today. Thoughts?

    Giving credit where credit is due: Hereís where I got the this information from: https://www.electro-tech-online.com/...urrent.125320/
    Chris
    [URL="http://www.maloneylights.com"]http://www.maloneylights.com[/URL]
    [URL="http://halloween.maloneylights.com"]http://halloween.maloneylights.com[/URL]
    [URL="http://ctm.maloneylights.com"]http://ctm.maloneylights.com[/URL]

  2. #2
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,519
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    Hi, a great piece of work.

  3. #3
    Join Date
    May 2007
    Location
    Mooresville, NC
    Posts
    7,685
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    Thanks Chris, that is great work.

    When I was using incans, I always used them, no mater what. Now with LEDs, in my display, I only have one LED string per channel, so that's about .04 amps. I do not use a heat sink on any of the new controllers I have.
    I do however, use it to align the triacs so if I ever do need one, it can be installed easily.
    James Family Christmas - 1600 channels of SS Renard channels
    Facebook - James Family Christmas


  4. #4
    Join Date
    Dec 2020
    Posts
    102
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    Ah, I haven't soldered the triacs yet because I am wondering if I should try to align them using a big heatsink (that I then would take off again). Since I am just starting with this hobby I don't have a heatsink, though. In my case I will still have some icans, especially some x-mas trees and inflatables. We don't get snow here but the temperatures are still cool in the evening so I think I can get away without a headsink.

  5. #5
    Join Date
    Jan 2015
    Location
    Folsom, CA, USA
    Posts
    2,287
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    Quote Originally Posted by munu View Post
    Ah, I haven't soldered the triacs yet because I am wondering if I should try to align them using a big heatsink (that I then would take off again). Since I am just starting with this hobby I don't have a heatsink, though. In my case I will still have some icans, especially some x-mas trees and inflatables. We don't get snow here but the temperatures are still cool in the evening so I think I can get away without a headsink.
    Make the temp heat sink from a piece of pasteboard or long popsicle stick if just foe alignment for soldering. Then remove when soldered. No need for aluminum unless you are going to use and keep in place.
    In Lights Therapy

  6. #6
    Join Date
    Dec 2020
    Posts
    102
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    I have some pieces of wood that might actually work for alignment. I have to cut them to size, first. Shouldn't be a big deal. But thinking about this, if I go through that process then maybe I should just install a real heatsink. I checked Ebay but there are so many different aluminum alloys... I bought some copper before to build a terminal bar (completely different project) but it was so hard I couldn't drill through that thing.

  7. #7
    Join Date
    Dec 2012
    Location
    Newtown CT
    Posts
    4,953
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    Here is my opinion on heatsinks: It does not hurt to have them. Without them you run the risk of stressing your parts. The cost of a heatsink is trivial. Why waste time discussing and deciding? Just put them on and move on.


    2020 Full sized show reworked for the new location. Only adding (famous last words) 13 RBLs that I finally got converted to using pixels
    2019 - Just moved into a new home (yet another change of plans). Will be dim but not dark. Too much to do at the new place to leave time for a show. Dim show (3000 pixels) had regular visits most nights.
    https://www.youtube.com/channel/UCyX...ttrsZNARkUce0Q

  8. #8
    Join Date
    Dec 2020
    Posts
    102
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    @MartinMueller2003 I actually got some ALU from Lowe's today. I had no problems cutting it with a jigsaw and drilling holes with a regular hand drill. I used a metal file to get the sharp edges off. Put some thermal paste on the triacs, slightly screwed the triacs to the headsink (so I can still move them and align with the through holes), put everything on the board and then soldered them on with a bit of extra clearance to the PCB (first the very left and then the very right so they don't move). I only did one side (12 triacs) so far as I had to purchase some extra screws (only had 25 M3s 8mm and already used a bunch of them before so I only had some 8 left) but I have the other 'heatsink' ready. Thanks for the feedback.

  9. #9
    Join Date
    May 2007
    Posts
    5,484
    Post Thanks / Like

    Default Re: Calculating whether we need heat sinks on the triacs

    As far as it goes, I'll agree with ctmal.

    But just to drop a fly in the ointment, I'll add a few comments.

    1) Here in Minnesota it rarely (i.e. never) reaches +40C during the winter lighting season. So the numbers are quite conservative for where I live and for many regions of the world.

    2) Even in Australia, Africa or South America the thermal mass of the triacs (unless you're using TO-92 parts) will take at least few cycles to reach their maximum values. The concern, of course, is that the controller becomes stuck on (due to either software failure or some sort of other failure in the electronics). So I agree that heatsinks are a good idea.

    3) Now for the hard part - determining what is meant by ambient and how to do the calculations. If a triac is dissipating full-power in isolation outside in the natural environment, then the calculations given above are quite valid. But if the triac is in a (more or less) sealed enclosure, then the 'ambient' temperature is not the outside air temperature, but the (higher) temperature inside the enclosure determined by the amount of heat generated inside the enclosure and the thermal resistance between the interior and exterior of the enclosure. Also, if several triacs share one heatsink, the temperature rise due to the other triacs must also be taken into consideration. This means that the above calculations result in marginal safety in the Southern Hemisphere, as air temperatures might exceed 40C during the summer in at least portions of Australia.

    So, conclusions: it's likely that none of this affects the results presented by ctmal for the snow belt in the Northern hemisphere, as the difference between 5C (i.e. 41F) and 40C (104F) results plenty of margin. Those in the Southern hemisphere should take a closer look at these matters, although I suspect that most are already using LED lights (in string or pixel form), and will likely never have to deal with controlling that much power through triacs.
    Phil

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •