resistor power dissipation 1

[zappedagain]

I have datasheets from two different manufacturers (Dale/Vishay and Panasonic) that claim their resistors in the 2512 package have a thermal resistance of 65-66 C/W. A colleague said 200C/W is more like it and just proved it by putting 1W through a 2512 resistor and watching the solder melt off the leads after about 30 seconds.

How do they spec this? The Dale data sheet says they use EN140401-802 for packages under 1206 but leaves the larger packages undetermined. This shows 220C/W for the 1206.

I can ask the manufacturers, but the tech support for passives isn't very strong these days so I thought I'd post it here first.

Z

 

[IRstuff]

This would seem to say that Vishay has something like 100°C/W:

http://www.vishay.com/docs/53047/pra-appn.pdf

TTFN

FAQ731-376

 

[Comcokid]

It depends upon what the 2512 is soldered to.

I recently used 2512 on a circuit board, and verified the heating by using a DC supply and thermocouple on one of my PCBs and found around 70 to 80 C/W rise.

BUT - my PCB uses 4 oz copper with 100 mil traces to the resistor! In the actual design I'm only using them for 1/2W.

 

[MikeHalloran]

It does indeed depend on what they're soldered to.

One time we had a high value resistor in series with a fairly high voltage supply (and its load), and it was melting the solder in its PCB.

The fix was heat sinks on the resistor leads; ~1/2" square x .02" thick fins, pierced in the center and soldered on the leads adjacent the resistor body. Brass would have been fine, but we made them out of copper because the dominant alleged EE couldn't/ wouldn't understand the basics of convective heat transfer.



Mike Halloran
Pembroke Pines, FL, USA

 

[zappedagain]

IR - That's a great link! I see that they take a lot of the PCB effects into account to get their low numbers.

Thanks for the replies, all!

John D