Chrispy_ wrote:Oh god, this is so true.
This is why I hate antistatic straps. If you are working on hardware inside your PC, just touch the case so that you're matching the potential of the electronics you're working on. That is actually far more reliable that being grounded at preventing ESD damage.
Well... yes and no. In low humidity conditions static charge can build up quite rapidly, from something as simple as shifting around in your chair. So unless you're touching the chassis pretty darned near constantly, there's still the potential for ESD damage.
Chrispy_ wrote:For what it's worth though, I have loads of kit that is, according to ESD precautions, abused horrifically. RAM modules lying in a draw full of plastic-wrapped stuff, graphics cards and procesors just sitting out on a desk for weeks. Me picking this stuff up to work on PC's after just walking across a carpetted room with rubber-soled shoes without even grounding myself on something first.
To date, I can't think of a single instance where I've killed something. It's all scaremongering by component manufacturers to minimise their risk of damage and RMA rates. The actual hardware is pretty robust against static - let's face it, there are diodes capable of blocking ESD voltages all over the board before you get to sensitive electronics,
Not all ESD damage is immediately apparent. CMOS gates can sustain partial damage that doesn't kill the chip outright, but causes it to no longer meet spec and/or die prematurely at a later date. Yes, there are protection diodes on most PCBs; but these are typically only used on external signal pins. Individual chips have internal protection diodes too, but these provide only partial protection.
Chrispy_ wrote:and almost everything is covered in heatsink/heatspreader these days anyway.
Only if you're talking high-end GPUs with heatsinks on both sides of the PCB. Most other expansion cards, motherboards, HDDs/SDDs, etc. still have a lot of exposed components and traces.
Last winter, I killed a mechanical keyboard simply by sitting down in my chair and reaching for the keyboard. Just as my fingers were about to touch the keys, I felt a spark arc from my finger into the keyboard, somewhere near the Enter key. The discharge killed both of the Shift keys, and the USB port the keyboard was connected to. (It turns out the RK-9000 has a dedicated row in the switch matrix which is used just for the Shift keys, and the sense line for that row got fried...)