[1UP]

I just opened up an old, dead DVD-rom drive and tried to desolder a few capacitors from the PCB.

I'm using an 18w iron and overall wasn't very successful. The old joint just didn't seem to want to melt.

Do you need a more powerful iron to desolder?

 

[BlazerFRS]

I have found that the solder on many computer boards is tough stuff; I tried getting some caps off a mother board with my 40W radioshack iron with no luck. It just wouldn't melt, even with the additon of more solder, etc.

I ended up using my dad's old weller soldering gun; Dual heat 240/360 Watt. it worked but the thing is a beast.

 

[guzzler]

There's very little solder on professionally printed and constructed boards. Any solder that is left will likely be quite dry. Often, the easiest way is to add more fresh solder and a drop of flux onto a joint and then try again. The through hole stuff will also be quite tight in terms of dimensional tolerance, so you're not likely to be able to complete remove all the solder in one go. The best easiest way is to heat each joint individually and just rock the capacitor to the opposite side and eventually it will come out far enough for you to remove

g

 

[mono]

I agree with Guzzler's points. Additionally some caps for power filtering do have a leg (if not both) connected to some rather large areas of copper on the board, In such cases it may be prudent even if unnecessary to use a hotter iron. Ideally you want that solder to heat as quickly as possible without excessively heating the cap and other areas. This is accomplished by higher intensity but shorter duration heating.

 

[1UP]

How much heat can a component take before it dies?

i.e. how long should the iron be in touch with the joint to desolder?

 

[guzzler]

As long as you don't heat enough to start burning the solder (the flux really), and only heat it until it just goes molten, you should be okay. The fewer times you have to do it, the better, but it's not so much the cumalative effect as the instantaneous. Semiconductors will take less heat than capacitors. It may be worth noting that the life of an electrolytic could be shortened by desoldering. Film capacitors are more resilient in my experience, although it is possible to melt the dialectric. Proper testing afterwards is a VERY good idea

g

 

[JWFokker]

Guzzlers method is exactly what I do. I'd be more worried about the PCB than the component though. I've killed a couple small amps from lifting traces because I cooked the PCB. Damn surface mount crap. While adding more solder works, I think just using flux would be better for removing components.