storing fets from esd damage

Thread Starter

pager48

Joined Nov 25, 2018
161
does wrapping the leads - not tabs together of to220/247/264 mosfets with aluminum foil protect from esd damage?
 

dl324

Joined Mar 30, 2015
16,846
It's better than nothing. I used to get components wrapped in aluminum foil; think it was from PolyPaks.

These days, metalized mylar bags or pink antistatic bags are used by most distributors.
 

dl324

Joined Mar 30, 2015
16,846
So when the foil shorting all the leads together - a gate would not be punctured by esd if the foil is hit instead of just the gate pin?
If all of the pins are shorted together, there is no potential difference between the gate and source/drain voltage that could damage the gate oxide.

The problem with using foil for ESD protection is that if a lead makes a hole and isn't touching the foil, the device could get zapped.
 

DickCappels

Joined Aug 21, 2008
10,152
The preferred materials -pink foam/bags, Metalized bags, and that black foam are all dissipative meaning not very conductive. The reason they are not highly conductive is that a resistive surface is less likely to zap a junction as a charged component is put in the bag of plugged into the foam, but it has enough conductivity to keep a significant charge from building up when, for instance, a bunch of FETs slides around in a protective bag.

At times I have resorted to aluminum foil because it seems that it is better than nothing.
 

dl324

Joined Mar 30, 2015
16,846
So what if the drain tab were to be zapped would the foil prevent damage in that case too?
It would depend on what potential the other terminals were at.

Also, it's possible for a device that's been zapped to be somewhat functional. In the past year, I've zapped several 2N7000, even though I tried to handle them properly. They all exhibit abnormal leakage.
 

dl324

Joined Mar 30, 2015
16,846
Seems like completely wrapping the parts in foil is the safest and cheapest route rather than spraying the storage containers with antistatic spray.
I store unused parts in the original antistatic material. I store parts that have been used for prototyping in black antistatic foam.
 

Tonyr1084

Joined Sep 24, 2015
7,853
LOOK OUT - - - HERE COMES A LONG LONG WINDED POST. "(

Couple issues that get under my skin when it comes to ESD. First is "Pink Polly". Yeah, it's safe around sensitive components but it does not shield components from ESD potentials. Another issue that bothers me is foil. More so when people say making the leads all touch the foil.

Look up "Faraday Cage" and you'll understand what the metalized bags do. They present a conductive bubble around a part so that any ESD event never reaches the protected component(s). Another peeve is when they say "Never pierce a velostat (conductive metalized bag)". So many people want to throw them out once they've been pierced. True, they do present a means for static to contact a component lead - BUT - an ESD event is when static voltage finds a path to ground. Let me just say it also occurs when the potential between two items is different. That's why when you get out of your car on a winters day you get that huge snap when you touch the body. Certainly the body is not grounded. But the potential difference is the snap you feel. But in such cases the body of the car, being massive enough, can represent a ground. Earth is not the only ground. An airplane flying through the sky is a ground of sorts as well.

But lets stick with ESD and FETs. A FET wrapped in foil IS protected. Not all leads need to contact the foil. If one lead is touching and the other two are not - within the foil there is NO PATH TO GROUND. And a pin hole doesn't mean static is going to leak in. Electricity and foil are different than a balloon full of water and a pin hole. The closer the static charge comes to a conductive surface the more it is attracted to that surface.

What blows components during an ESD event is a rush of current finding a path to ground. (or in any other instance, a surface of much higher or lower potential). When static begins to move it is no longer static. It's dynamic. It's electrical current. And if the shortest path to ground is through the plastic casing of a chip, through the die, through the lead to a ground then that is an ESD event that likely damages a component. And damage doesn't always mean total failure. Think of a strong chain. One link has been exposed to acid. The chain is still strong, but over time the link corrodes, and eventually becomes too weak to handle the load it was expected to handle. So ESD control is important at all phases of construction. Once a component is in a circuit is LESS vulnerable, but it may still be vulnerable. I watched a satellite test go bad simply because someone wheeled a plastic cart past the satellite under test. Just so you know, components on satellites can be sensitive to static voltages of less than 50 volts.

The static field; imagine you have a conductive mat. You have a component laying on that mat. You put your hand (ungrounded) over the component but you don't touch it. You've created a field of static that can harm the chip, depending on how sensitive it may be. Touch it before you've become neutral in charge and you can send a BIG zap through the chip. Wrist straps are the #1 best way of preventing a chip from being damaged while on the work bench. But even when protected this way, sitting on the wrong material can generate huge spikes that the wrist strap and mat may not react quickly enough. My wife likes a plastic protector on the floor at her computer. I hate the darn thing. If I sit at her computer wearing just socks, and this is worst in the winter, if I slide my foot even a little bit I get a big snap right at the keyboard. ESD is far more than wrapping a sensitive component in a Faraday cage. It's preventing when possible, neutralizing any stray potentials, and draining any hazardous static voltages from reaching the work piece. Voltages as high as 3,000 volts can BARELY be felt when one knows what to feel for. Otherwise they're just absolutely un-noticed. And 2500 volts can be deadly to a lot of components.

But it's not the voltage, it's the current. Like that chain mentioned a few paragraphs earlier, an EOS event (Electrical Over Stress) can weaken a trace. Like half way burning out a fuse. It still conducts, but it can no longer handle the full amount of current it was rated for - so too are the traces inside the chip. They are microscopic. Melt half way through one trace and it may still work. But under normal every day load it may fail. Not so serious an issue when it's a child's toy. But in a jet airliner, medical equipment, anti-nuclear missile defense - IT HAS TO WORK! AND WORK EVERY TIME WITH 100% RELIABILITY. OK, not shouting, just making a point.

Foil? It works. But as long as the component is not damaged before you get it fully wrapped. I like the conductive foam best as it is resistive. A conductive mat is also resistive, so no huge currents are seen when plugging a component into foam or laying it on an ESD mat. So long as you were wearing your wrist strap and maintaining the same electrical potential as the mat and the chip.

Ionizers are nice because they shower the work field with positive and negative ions. Anything with a positive charge will attract negative ions and neutralize the potential difference. Same with negative charges. Surfaces that are not conductive can hold static charges. An ionizer neutralizes those charges.

OK, usually I keep quiet about many things here because there are many things I'm not well versed on - or even outrightly don't know what I'd be talking about if I tried. But ESD is something I know. 30 plus years in the industry. Automotive, medical, aerospace and defense. Some pool spa controllers as well; don't think I've ever worked on any "Toys".

Peace all. Happy holidays if I don't get back and respond again.
 
Last edited:

dl324

Joined Mar 30, 2015
16,846
But lets stick with ESD and FETs. A FET wrapped in foil IS protected. Not all leads need to contact the foil.
Have you considered the case where a lead has pierced the foil and it isn't touching the foil? In that case, there can be a ESD event that can cause damage.

If foil surface is intact and some leads happen to not be touching, they are indeed protected. But that's not the case that was being discussed.
 

Tonyr1084

Joined Sep 24, 2015
7,853
Have you considered the case where a lead has pierced the foil and it isn't touching the foil? In that case, there can be a ESD event that can cause damage.
Yes, that is considered. Guess I wasn't clear enough on that part.

If a lead is protruding the cage (lets call velostat bags and foil wrappings a Faraday cage), the lead is in fact exposed to potential voltages. Should something contact the protruding lead, and lets assume for some reason the protruding lead is not touching the foil; to where inside the cage is the current going to go? Keep in mind we're talking static voltage which is not current and current which is when static voltages no longer are static, but moving.

As mentioned about grounds (which I hesitate to use because even a small door knob can give you quite a snap when you walk across the carpet wearing leather shoes) the current is going into a significantly different potential. For sake of argument lets assume it's the doorknob that is charged to 50,000 volts (and yes, the human body can easily carry that high a voltage in static form), and you are, for all intents and purposes, at zero volts. You grab the door knob and you get that unpleasant snap. The VOLTAGE is there, but the mass is not. Just like a capacitor, the larger the storage reservoir, the greater the CURRENT potential when discharging. I've burned tips of screwdrivers away just discharging caps that came out of some industrial power control units. Tiny caps at that high a voltage (assume for a moment that they can handle that kind of voltage) will hardly give you a snap whereas a cap from an overhead crane motor control board can knock you on your arse. I get off the track easily, lets get back to the mass of a FET.

There's not much mass when concerned with a gate lead poking through aluminum foil. I'm sure on some nuclear level there WILL be some current leaking into the cage, but I wouldn't have those numbers. So a gate poking through the cage is unprotected, but the chances of harm are extremely low. There is LIKELY more potential harm during the wrapping of the FET in foil than AFTER it's wrapped. That's why the preferred method is to poke the leads of a FET into static foam (the black stuff that is carbon impregnated). All differences in potentials are equalized but at extremely low currents. Not so with foil. With foil you get the sudden and complete transference of the potential energy stored in the form of static; which is why I don't care for using foil. In a pinch I'd use it. It's far more protective than throwing it inside pink polly.

By the way: Pink polly is a plastic that does not produce static. However, it is not a cage. Velostat is a metalized plastic bag that creates a cage around the part. I'll look to see if I can find some illustrations that I have in mind and post them. If not, I think I can draw up an illustration or two on my own.
 

Tonyr1084

Joined Sep 24, 2015
7,853
Been a while and several computers ago, I can't find any images. So here's one I just banged out. Three scenarios play out. A hand over an ESD mat, A FET in a Faraday Cage and the same FET with the gate poking through the cage. The blue lines represent the presence of ion flow. When the potential is great enough current will flow through the air without the presence of a spark. Static IS discharging to the mat. In the drawing considerable space is drawn in for clarity, but as one reaches for a component the distance between the source of static and the device and ground grows ever closer until contact is made. YOU DON'T NEED TO TOUCH A COMPONENT TO BLOW IT OUT! Case in point - the satellite under test. If you've ever messed with extreme high voltage you may have experienced a hissing sound as the source and ground grew close in proximity. That's electron flow through the air. You don't have to TOUCH or MAKE contact in order to initiate a cascading electrical field. Don't believe me? Build this: A simple relaxation oscillator using a neon bulb. Protect all electrical wiring from incidental contact. Notice the rate at which the bulb is blinking. Generate a good static charge and put your finger NEAR the glass bulb. The blink rate goes up noticeably. Why? Because of the presence of the static field. It induces some voltage which causes the neon bulb to ignite quicker. THAT is static field.

In the bag / cage / aluminum foil is our hero (or villain if you prefer). In example B he's completely protected from the force. All electrical charge is conducted around him via the cage. Even if he's touching the foil on the inside, there is still no current flow through the body of the FET. In example C, the gate has pierced the cage. Some current may flow through the gate into the drain causing some potential damage. This may be a latent case where damage does not manifest itself for some time. Still, in the event the cage is pierced, changes are good there's still contact with the gate, and any subsequent static that may pass through the exposed leg of the gate will likely find its way to the cage. But the potential always exists. However, if the cage is pierced but the lead is not exposed then there will be no current leaking into the component.

ESD and Bubba FET.jpg
 

dl324

Joined Mar 30, 2015
16,846
Tony's detailed description in #13 matches my concerns about aluminum foil - that it's conducting the static electricity all too well.
What are you concerned about? It's a potential difference that causes any damage. A static charge contacting the foil would put everything inside, and touching, it at the same potential; anything not touching it is safe too.

As has been pointed out, it's not the "best" protection but it's better than nothing.
 

ebp

Joined Feb 8, 2018
2,332
What blows a FET is a voltage differential between the gate and the body that exceeds the breakdown voltage of the very thin insulation. If something prevents the potential difference it protects the FET. It doesn't matter in the slightest if the whole FET is suddenly raised to a million volts by a static discharge as long as the voltage difference between the terminals does not exceed the maximum allowable level. Of course inductance comes into play if there is an extremely rapid voltage rise, even if the resistance is very low.

The huge majority of discrete FETs and integrated circuits are supplied electrically bulk-conductive plastic tape that has a static-dissipative (very very low conductivity) transparent cover. The leads of the parts are sometimes sort of shorted to each other as they lie in the tape. This is sufficient to protect the devices form triboelectric charge generation and minor discharge from external sources.

When you discharge 50 kV, you have to have something that is extremely high resistance to in any way reduce the current to anything that might be safe. If you have a strip of conductive foam say 1 cm long with one end at zero volts and apply 50 kV to the other end, the voltage gradient between two FET leads stuck into that foam can still be high enough to kill the FET. The conductivity is neither high enough to prevent high current nor low enough to constitute a short circuit. Looked at another way, if the notion is reduction of current due to resistance - 1 mA for 1 µs is sufficient to charge 50 pF to 20 V - the specified absolute maximum gate-source voltage for many discrete power MOSFETs. From a 50 kV source you need 50 megohms to limit the current to 1 mA. This is several orders of magnitude higher than the resistance of conductive plastic foam.

The ONLY way to assure safety of the part in the event of a high-energy static discharge into any part of it is to have the leads short circuited with a very low resistance and inductance between them.

When discrete dual-gate MOSFETs in 4-lead metal cans were common (well, they existed, "common" is a stretch) they were supplied with a metal shorting clip around the leads. These devices had tiny structures so it took very little charge to produce a very high voltage on the gate capacitance. The clips were normally left in place until the parts were soldered into the circuit board, though they could be removed prior to that if great care was taken.

There was a time when some ICs were supplied in aluminum rails. I think I still have one that was originally used for Intel UV EPROMs. Bulk conductive plastic rails were moderately common and are still used a bit. Most rails now just have and antistatic surface treatment.
 
Top