Power Dissipation in DIP Resistor Network

Thread Starter

iang

Joined Jan 15, 2008
3
I have a situation where I need to install a DIP pull up resistor network into a user configurable card. The issue I am having is that these networks seem to be rated at a maximum of 250mW per resistor and I need them to be more in the range of 500mW if at all possible. Does anyone know if something like this exists or are there anyways around this? Unfortunately, custom making any network would not work because they need to be commercially availabe if possible.
 

kubeek

Joined Sep 20, 2005
5,794
Don´t think that you can find DIP resistors with 500mW each (BTW don´t you mean SIP? like single in-line package, not dual usually).
The other point is that most probably a 12-resistor SIP network, even though rated 250mW each, will burn up if you try to use the 250mW at all of the resistors (3W total).
I think that 1W total is close the maximal power.

You should better use standard 0.5W resistors instead, they will have much better cooling than a small network, but they take a lot of place.
Maybe you could try mounting them vertically, like the body of the resistor is not parallel to the pcb, but rather pointing upwards.
 

Papabravo

Joined Feb 24, 2006
21,159
Would you care to explain why you need that much power dissipation in a pullup resistor? Are you trying to minimize rise time on an open collector(drain) bus? If so you might wnat to consider switching to a tri-state bus. It'll save you a number of headaches.
 

SgtWookie

Joined Jul 17, 2007
22,230
Resistor networks are available in DIP, SIP, and many other configurations as well.

Seems typical to find SIPs rated for 1/8W per resistor, DIPs 1/10W per resistor, and the package to have a limitation in the vicinity of 1W.

You might take a look at Bourns 4100R series of thick-film moulded DIPs. A 4108R-1-RC consists of 4 isolated resistors in an 8-pin DIP, 1.69W rating for the IC. Still not quite up to your specs, but you could use 3 of the 4 resistors and have some "breathing room", depending upon your cooling.

As has already been asked, why the dissipation? Unless you're doing PECL/MECL, that's a LOT of power to be burning up. It would make a pretty good room heater.
 

mrmeval

Joined Jun 30, 2006
833
Is this something you are designing or something that already exists? Are you trying to fix it or improve it?

If it's a design then consider using discrete resistors elevated off the board or finding a way to dissipate less power.

Surface mounted resistors or laser etched network on a ceramic substrate would probably be too expensive.
 
Top