Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Heatsink area size

Status
Not open for further replies.

offanon

Mechanical
Nov 16, 2009
7
I have a circuit with four BTA312X triacs. The power three 1.1KWatt elements and one 2KWatt element. The Thermal Resistance is 1.15. The heater is powered from 240v mains and the circuit has an ambient of 65°C. How do I calculate the area of the heatsink required to keep the triacs cool? I want to use an extruded aluminium heatsink.
 
Replies continue below

Recommended for you

You know the thermal resistance of triac-to-metal (with grease), and you know how much power you need to dissipate as heat in the triacs... plug-and-play, for the most part. All you need to worry about is metal-to-air resistance (depends upon ambient conditions), then decide if you need to use forced cooling.

Dan - Owner
Footwell%20Animation%20Tiny.gif
 
I have to use natural convection so no forced cooling. The heatsink size seems to be very large so I was hoping someone could run the calculations to see if Ive got it wrong. Im running with a surface area of 310723mm^2 but that cant be right can it? With fin spacing of 7mm, fin depth of 25mm and fin thickness of 2mm the x*y is coming in at 178mm * 220mm???
 
The area will depend on the temperature you want the heat sink to get to. When the system has reached steady state, your heat sink will convecting all the input energy. This will be based on the convective heat transfer coefficient, the differential temperature, and the area of the heat exchanger.

Natural convection will probably get you to 1 or 2 BTU/hour/degF/ft^2. So at steady state, Q = h*A*dT. Once you choose the max allowable temperature, you can then find area.

EE
 
For higher levels of cooling, fin direction also makes a huge difference... are the pointed up where natural convection currents can speed the cooling process, or are they sideways? That type of stuff.

You can get some decent back-of-the-napkin numbers without just a few minutes of calcs, but as the system gets more complicated (i.e., more cooling is needed), it becomes an engineering field unto itself.

Dan - Owner
Footwell%20Animation%20Tiny.gif
 
I didn't do any calculations, but the size could be reasonable for 22A. As already noted, the size will vary depending on what you use as the delta temperature or how close to the maximum temperature you're willing to push the junctions.
 
You've stated power, but it's not clear to me you've stipulated the actual thermal load from the triacs, or the power they're switching.

TTFN
faq731-376
7ofakss
 
First must know power lose in triacs. Even if 90 degree angle switch is used, switching loses are very low at 50/60Hz, so main loses are done by Wt = Vt x It, where Vt are triac on voltage and It rms current. If all loads are resistive, currents are 4.58A respectivly 8.33A. Vt is about 1.5V, so Wt=6.87W, respectivly 12.5W.
If ambient temperature Ta is 25C and accepted maxim capsule temperatue Tmax is for exampele 75C, it need that maxim thermal resistance (jonction to air) to be Rthmax =(Tmax-Ta) / Wt = 7.27 C/W and 4C/W. If Rth capsule-sink is 1.15, result maxim allowed Rth for sinks (6.12 and 2.85). From various technical guide must choose apropiate heatsink that have thermal resistance below these values.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor