18
\$\begingroup\$

My understanding is that ESD safety things (mats, wrist straps, specially marked soldering irons) are designed to bring everything that can touch a component to the same electrical potential energy – ground.

But it seems unreasonable to expect that there's no voltage between my desk and the factory where my components were produced. After all, the factory is likely halfway across the world, and the resistance between here and there is significant.

So, say a component is carefully packaged and shipped to me in one of those little ESD-safe bags. Before opening the bag, I carefully ground myself and my workstation. Despite this, the component is destroyed as soon as I touch it because the ground that I tied myself to is much different from the ground that the component was tied to when it was produced.

What precautions are taken against this? Is it just something that can happen in theory but that isn't an issue in practice?

\$\endgroup\$
  • 7
    \$\begingroup\$ Earth ground is basically the same everywhere, except like right after a lightning strike... \$\endgroup\$ – vicatcu Jun 1 '17 at 3:03
  • \$\begingroup\$ Your component was detached from the factory, therefore floating in relation from it. As soon as it is attached to your ground, it's grounded in relation to you. Voltage is relative. \$\endgroup\$ – Wesley Lee Jun 1 '17 at 3:47
  • 7
    \$\begingroup\$ @WesleyLee Obviously. But presumably, while the component is floating, it remains at potential it was left at the last time it was connected to something. The difference between that potential and my potential is the harmful voltage I'm talking about. \$\endgroup\$ – Maxpm Jun 1 '17 at 4:52
  • 11
    \$\begingroup\$ @Maxpm I think the element you are missing is that ESD bags are not insulating, they are conducting to various degrees, and designed to dissipate and prevent buildup of static charge. The component is touching the bag so they are at the same potential, if you are at a different potential when you touch the bag, the bag will dissipate the charge harmlessly. en.wikipedia.org/wiki/Antistatic_bag \$\endgroup\$ – Sam Jun 1 '17 at 8:48
  • 2
    \$\begingroup\$ If you purchase from a reputable supplier, they will include a small bag of local earth which you must stand on while handling the components. \$\endgroup\$ – Ian Bland Jun 10 '17 at 19:49
32
\$\begingroup\$

Components are damaged by two or more of their pins being at a large enough potential difference. If the component has a conductive case, or pad, then that counts as a 'pin' too.

It's possible to break them by trying to charge them up to a new potential through one sensitive pin, while the voltage of the other pins is held more or less constant through capacitance to ground. That can be the situation where you, perhaps charged to 15kV with respect to ground, pick up a component that's at ground potential by (say) the gate lead.

Conductive packaging shorts all the pins together. What you do is to bring the conductive bag to your potential first. Any charging current that has to flow into the component does so through all pins, so does not damage the component.

Let's say an insulated carton of components in conductive bags charged to 100kV arrives at your workstation. You and the workstation are grounded. You open the carton, and as soon as you touch a component bag, a current flows between you and the bag to discharge it down to ground potential. Meanwhile, the bag has maintained all the component pins at the same potential, so no damaging voltage is applied across the component. Now you and component are at the same potential, you can open and touch.

Why did the component arrive at 100kV? Surely the other factory ground is not that different to yours? No, but the last bit of the trip might have been carried by a guy with nylon shoes. When stuff is properly packed, it doesn't matter if intermediate stages of the journey take it to potential way different from ground.

\$\endgroup\$
  • 2
    \$\begingroup\$ It strikes me as sometimes the conductive packaging shorts the pins together. If they were all embedded in a static dissipative foam block, obviously we can expect them to all move roughly in unison. However, consider the case of a harddrive wrapped in an anti-static bag. Its pins are likely not touching the bag. In such cases, I would assume there's slightly more design to be considered. \$\endgroup\$ – Cort Ammon - Reinstate Monica Jun 1 '17 at 20:13
  • \$\begingroup\$ @CortAmmon Yes, to keep complication down, I deliberately didn't go into the Faraday shielding effect of the conductive enclosure, which maintains zero field inside. So the pins don't need to touch it. \$\endgroup\$ – Neil_UK Jun 2 '17 at 14:22
  • 3
    \$\begingroup\$ It seems to me that this answer is, at best, only partly correct. In a typical IC, not all parts of the chip are directly connected to the pins. Even if you ground all the pins simultaneously, there will be some transient current between the points where the pins connect and the rest of the chip. Probably not (nearly) as much as you'd have if you only grounded one pin and let the others discharge through the chip, but still some. AFAICT, the real reason why anti-static bags work is given in The Photon's answer: they have a high enough resistance to slow down the discharge to a safe rate. \$\endgroup\$ – Ilmari Karonen Jun 2 '17 at 15:30
  • \$\begingroup\$ Components are also damaged when the case has a different potential than the pins (or other parts). Shorting the pins together does not guard against static damage -- a deadly lesson learned while handling solid-fuel rocket motors. \$\endgroup\$ – Brock Adams Jun 2 '17 at 21:43
  • \$\begingroup\$ Also, anything capacitively coupled through the component case will be likely aggravated by a real low-impedance grounding of the pins... \$\endgroup\$ – rackandboneman Jun 2 '17 at 22:40
31
\$\begingroup\$

Hopefully your parts are packaged in an ESD-dissipative tray or bag. Then when you set them down on your ESD mat in your lab, any charge that's built up on them can drain away through the packaging and the mat. They won't discharge quickly enough to damage the components because both the bag and the mat have substantial resistance (1 megohm to ground is common for ESD mats and wrist straps).

\$\endgroup\$
  • \$\begingroup\$ Static sensitive parts should never be shipped in dissipative bags or trays, as they do nothing to prevent a static shock from damaging the parts. The parts can still be destroyed by a static shock right through dissapative materials. All parts should be sent in anti-static materials. \$\endgroup\$ – Keltari Jun 3 '17 at 4:53
12
\$\begingroup\$

My understanding is that ESD safety things are designed to bring everything that can touch a component to the same electrical potential energy

That's when you're right.

ground.

And here's when you're wrong. There is no such thing as "universal ground". Not even Earth is. You just pick a point of a circuit and say "Hereby, by the power vested in me by Science of Electrical Engineering, I declare you as The Ground and all other grounds as null and void." and instead of a sword, you touch it with an ESD safety thing. That's it.

It doesn't matter if your ground is same or 1000V different than ground they've used at the factory. As you've said, the ESD safe equipment is designed to bring the part safely (read: slowly) to your ground.

\$\endgroup\$
2
\$\begingroup\$

A logical extension of this would be sending a replacement PCB to the ISS or a satellite where "ground" in the sense of what you stand on is a long way away and separated by vacuum. Large charges can build up in space, so there will be a significant potential difference. In an electrical sense there's no problem so long as you bring the two grounds together properly (slowly through a significant resistance, and without causing a large potential difference across the circuit -- see Neil_UK's answer)

\$\endgroup\$
0
\$\begingroup\$

Basic electrostatics: If a bag or metal box or some other conductive container completely surrounds some object, the potential of that object is effectively zero, regardless of whether it's connected to earth ground or a 10KV power source. (I suspect there's an exception for high frequency AC, but we can ignore that.)

When the object is bagged in the factory, the bag is grounded to the factory ground for the transfer. It may go through several thousand volts fluctuation between there and where you are, but the potential between any two points inside the bag is still zero.

When you receive the bag you first ground it to your desk ground, then open it. As your hand reaches in it is a zero relative to desk ground because your wrist strap assures that, and the component is at desk ground because the bag (and everything in it) is also grounded to your desk.

\$\endgroup\$
  • \$\begingroup\$ This is only true for objects inside the bag/box that are electrically connected to it. As @CortAmmon points out, if some parts are touching the bag but others are insulated, charging/discharging the bag can create a potential difference between different things inside the bag. (No insulator is perfect, but it could take a very long time to charge to equalize.) Anyway, this is why there are conductive trays, not just bags. \$\endgroup\$ – Peter Cordes Jun 2 '17 at 2:27
  • \$\begingroup\$ If you place an object in the bag, but insulated from it, then raise the potential of the sealed bag, the potential of the object is raised by a similar amount. \$\endgroup\$ – Hot Licks Jun 2 '17 at 2:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.