Just a quick question
Just a quick question, when using the hot air gun when re-flowing chips, why doesn't all that heat damage the chip?
Made from silicon and or delrin, both are able to withstand high heat. Ever hear of Silicon Valley? It's where a lot of computer chips are made here in the USA.
Short answer is that applying heat externally to flow a joint is very different than the chip itself generating that equivalent thermal load from within. One can usually get away doing rework on chassis mount, through hole, and SMD components without damaging em.
Long answer is that external heat to a deenergized chip actually can shorten its life or outright kill it and thermal management is very important to high reliability rework. Much like steel becoming fatigued and eventually failing under normally in spec punishment, there is a trade off between getting a joint up to temp to complete the operation and exposing the parent chip package to the least heat for the minimal amount of time. Different chips have different thermal ceilings for in operation and while deenergized or in storage but in general Silicone based commercial IC's can only get to about 70C while under load. Yes CPU's and GPU's can exceed that depending on how they're physically engineered and the throttling is coded but in general BJT's, FET's, op amps, comparators, and control IC's start to fry around there. The storage and deenergized temps can get substantially higher though. That's what we're talking about when reworking a board. Also keep in mind that when reballing, reflowing, removing or installing a SMD chip they don't have much thermal mass so they heat quickly but also dissipate that heat quickly. Think about how often you've seen Alex test a GPU without heatsink, allow it to throttle/safely shut down, and within a short period it's back to ambient temp. That's fine in moderation and actually more damaging than rework processes. Do it too often or try to reflow the same chip too many times and you're playing with fire, but modern tech is fairly resilient. When using a ThermoJet (PACE name) pads will reach melt temp for even unleaded solders before the core, where the actual wire bonds and silicone are, comes close. Preheating a board and component with an IR or convective/conductive hot plate (with proper ramping!) allows one to minimize that gap between component and flow temp. This way you don't have to wait so long for the whole ground plane or even fiberglass board to reach proper temp, can apply lower heat for less time to get the same results, and have an added cushion to any potential degradation.
Know I jumped around a bit and probably got too technical but does that make sense?
Second Chance Repairs LLC