Choosing heatsinks for lead-free soldering of transistors

Thread Starter

GreenPenInc

Joined May 13, 2007
12
Hi all,

I'm about to place an order with Digi-Key for a bunch of components for the power supply I'm building. (The project is from the TAB guide to understanding electricity and electronics; if you "Search Inside" for "assembly and testing", you can find the description of the project's various stages.)

Right now my order is about 20 bucks. There is a 5 dollar handling fee for orders below $25, so I'm planning to flesh out the order with a few extra goodies.

One potential problem I foresee is that the slightly higher melting temperature of the lead-free solder I use, coupled with my inexperience at soldering, gives a higher probability of damaging sensitive semiconducting components, like my 2SA/2SC transistor pair. I am already determined to use lead-free solder for a variety of reasons which I'd be happy to go into.

So my question is: which heatsinks on Digi-Key's catalog should I order? The specific transistors I'm ordering are the TIP31C, TIP32C, 2SA1943 (replacement for 2SA1302), and 2SC5200 (replacement for 2SC3281). Any pointers you can give me would be much appreciated!
 

beenthere

Joined Apr 20, 2004
15,819
Lead free is a real tempest in a teapot, but that's not the issue.

It sounds as if you are worried about frying your transistors whle soldering them in place. They appear to be all through-hole. There should not be any reason for concern. Using the right flux and a properly tinned tip will let you make a joint long before any harm comes to the transistor.

The heat sinks that would make a difference would have to clamp onto the individual leads of each transistor - but only for the duration of the soldering. This was an absolute necessity in years long past. The germanium transistors in TO-5 cans had thair internal connections hand soldered using indium. The melting point was so low that the transistor (think 2N404) would be ruined if it got soldered without heat sinking the leads - that's why I still have so many hemostats.

We were so amazed by the first silicon transistor to show up in TO-92 that we just had to test it. We jammed it into the end loop in a Weller soldering gun, and held it at high heat for a full 5 minutes. The junctions were still good after it cooled down.

The only time you worry about heat is while the transistor is operating.
 

mozikluv

Joined Jan 22, 2004
1,435
i don't know at what ambient temperature your transistors would be working.

before you calculate you must remember that the thermal resistance from the die to the outside air (junction to ambient. ja) is a combination of 3 thermal resistances, junction to case (jc), case to sink (cs) and sink to ambient (sa). for you to determine what is the (jc) of each transistor you mentioned you have to consult their respective datasheet.

because convection heat flow known as power dissipation is analogous to current flow, hence thermal resistance is analogous to electrical resistance and temperature drops are analogous to voltage drops, the power dissipation of each transistor found using this formula

Pdmax = (Tjmax - Tamb) / junction to ambient

for Tjmax you can either use 75degrees C or 150degrees C

heatsink to ambient (sa) is calculated in this manner:
sa = [(Tjmax - Tamb) - Pdmax(jc + cs)] / Pdmax

in case the working condiction is higher than 25degrees C then the thermal resistance for the heatsink would be smaller.

moz
 

bloguetronica

Joined Apr 27, 2007
1,541
I normally use heatsinks from Fisher. The have a wide selection of heatsinks. You should use a larger heatsink so the temperature of the transistor won't rise to that point. Use the formula to calculate the heatsink (in ºC/W or F/W):
Heatsink = (Tamb - Ttransistor) / Pdissipated.
I normally consider an ambient temperature of 40ºC and a device temperature of 70ºC for safety reasons and in order to make them last longer. This is a simple formula applicable if you use heat conductive paste (so heat resistance can be neglected).
 
Top