ESD: Truths, myths, and flat out lies

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
That's interesting. I certainly believe that an LED module, as shown in the picture, would become damaged, since a CMOS chip is driving the LED. What's the failure mode of the actual LED itself under ESD?
 

BrainEater

Gawd
Joined
Jul 21, 2004
Messages
1,017
Even the led's are super sensitive.

Quite a few LED manufacturing companies actually build in ESD protection right onto the led chip.....The last set of luxeon III's I bought are rated 16kV esd protection , and even that isnt enough sometimes. ;)

Failure mode ? .... does "zapped" count ?

:D
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
No, "zapped" doesn't count; sounds like you'd be better off with "I don't know".

In my experience, LEDs aren't sensitive at all. I've ruined plenty of CMOS parts, but never had a problem with any LED. Of course, I'm referring to plain old LEDs, not parts that have integrated controllers, logic, regulators, or modules with LEDs on them. I think it's the latter that the Osram document is referring to.
 

BrainEater

Gawd
Joined
Jul 21, 2004
Messages
1,017
aw.

:D

Ok the failure mode would be "reverse-bias electrical over-stress"

However , without the ability to actually test the LED's I've 'zapped' (I've killed several) , it would be hard to know forsure....but please take my word for it , static will kill LED's dead....laser diodes too.Every single high-end LED I have ever bought , has come in full ESD packaging.

There is a lot more documentation online about the subject , that osram doc was just the first I saw......

edit :

Here's a doc from where I get my LED's (phillips lumileds , (Luxeon brand)) :
http://www.philipslumileds.com/pdfs/RD25.pdf

I'll direct your attention to page 11.

:D
 

Adrenaline

[H]ard|Gawd
Joined
Oct 11, 2000
Messages
1,073
One thing to note about all these LED precautions, is in the end current kills the device not voltage. I have ran LEDs with a current limitted 15KVAC supply and they last no problem, as long as the current is kept low.
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
Right; since it's current that's the problem, the device has to be in the path of idscharge--not just subject to a discharge event like MOS parts.

Are those Luxeon high-power LEDs really just LEDs, or is there controller or limiter circuitry in the package? These are very special parts, producing 30 lumens from 350 mA and including lenses and reflectors. What are you using them for?

Thanks for the link!
 

BrainEater

Gawd
Joined
Jul 21, 2004
Messages
1,017
You are welcome !

-----

It depends what luxeon you buy.Some of them are modules , most of them are just emitters tho.

These are the last emitters I bought.I use em for project lighting like This....

:D
 

kookoo102

Limp Gawd
Joined
Mar 4, 2009
Messages
455
Good info!
i have been having ESD problems in my house. i could walk accross the house and touch the front of my case and zap it (computer shuts off). i have even just walked in the room and zapped the computer... without even touching it!!
come to find out there was a USB cord plugged in but laying on the carpet.
lets just say that motherboard does not work anymore.
 

bluc

n00b
Joined
Aug 8, 2009
Messages
4
Hello all first post here!! I am setting up an esd workbench and need some advice.I have 4 banana plugs wired up to an earth wire which goes out to seperate earth stake(straight into ground) this is a straight through cable no resister. I then conect my anti static desk mat to a banana plug with an anti static wrist strap which has the needed resistance. I also connect myself to the earth/banana plug with a wrist strap same resistance as desk mat. I need to know if I also am supposed to conect a wire between the mat and myself to equalize the static between the me and the mat or if all static simply gets drained out down the earth cable. I also plan to get a anti static floor mat as soon as I can afford it. Any help much appreciated. cheers:confused:
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
You're probably best off, bluc, posting your own new thread.

Your arrangement is fine; you and your surface will bleed to the same potential through the resistors.
 

thaltek

Limp Gawd
Joined
Sep 15, 2009
Messages
206
i guess i have been lucky so far in that i havn't used any esd protection without problems....and i have done at least 6 full system builds from scratch...... :eek: i would like to thank the poster of this thread for making me more paranoid than ever about esd..... yha yesterday i broke down and bought a wrist strap...... :cool:

i guess esd isn't an issue if you also happen to love tropical fish and have 70+ gallons of fresh water humidifying the room..... :D
 

ATITek

Limp Gawd
Joined
Aug 6, 2004
Messages
138
This is in a A+ Cert book (CompTIA A+ Certification Workbook):

~~4. What is the purpose of an antistatic wrist strap?

An antistatic wrist strap equalizes the electrical potential between you and the PC. It doesn’t “ground” you per se, unless you use it in conjunction with a grounding mat; it simply keeps the potential the same between you and the parts that you don’t want to accidentally zap.


~~5. Where does an antistatic wrist strap attach at each end?

The wrist strap attaches to your wrist, of course. The alligator clip at the other end attaches to the PC’s frame or to its power supply — anything that’s metal and not a circuit board.

~~8. When working inside a PC without an antistatic wrist strap, experts recommend that you frequently touch the metal frame of the PC case or the power supply box. What does that do?

Touching the metal frame equalizes the charge between you and the PC’s frame, keeping excess charge from building up. If you touch the metal frame, the electrical potential in your body flows harmlessly into that nonelectronic source. If you touched a sensitive circuit board instead with excess charge built up in your body, it would get zapped.




Source: http://www.computerbooksonline.com/content/viewer.asp?a=3266&z=5


I also have a PDF copy of the chapter. Long story short, you are safe not grounding the PC with power cord. A anti-static strap connect to the case of the PC the eliminates POTENTIAL of ESD.
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
Touching the frame (or strapping to it) equalizes the potential between you and the frame. Does it equalize the potential between you and your tools, your work surface, or the next location you move the parts?
 

Elledan

[H]ard|DCer of the Month - April 2010
Joined
Oct 18, 2001
Messages
15,885
Touching the frame (or strapping to it) equalizes the potential between you and the frame. Does it equalize the potential between you and your tools, your work surface, or the next location you move the parts?

Each individual ground plane has to be equalized separately.
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
I'd agree, which means that the A+ certification is wrong. (Again.)
 

DejaWiz

Fully [H]
Joined
Apr 15, 2005
Messages
20,384
Long story short, you are safe not grounding the PC with power cord. A anti-static strap connect to the case of the PC the eliminates POTENTIAL of ESD.

I disagree.

To elaborate (simplify?) what mikeblas and Elledan are stating: Just because a wrist strap is present to prevent you from damaging components doesn't mean another entity can't, like a carpeted floor that the chassis happens to be laying on. It's still quite possible to discharge static voltage build-up thus damage the components if the circuit path as a whole isn't properly grounded. The preferable way to maximize ESD protection/prevention is to ground the chassis (or surface that all the components are in/on) while utilizing a device to equalize the potential between all ESD capable entities (humans, workbench surface, tools, etc.).
 

Elledan

[H]ard|DCer of the Month - April 2010
Joined
Oct 18, 2001
Messages
15,885
I disagree.

To elaborate (simplify?) what mikeblas and Elledan are stating: Just because a wrist strap is present to prevent you from damaging components doesn't mean another entity can't, like a carpeted floor that the chassis happens to be laying on. It's still quite possible to discharge static voltage build-up thus damage the components if the circuit path as a whole isn't properly grounded. The preferable way to maximize ESD protection/prevention is to ground the chassis (or surface that all the components are in/on) while utilizing a device to equalize the potential between all ESD capable entities (humans, workbench surface, tools, etc.).

Which is for example why my soldering station has an input for a ground wire which I can connect to whatever project I'm working on lest I fry some ICs with the soldering iron :)
 

lasserith

Limp Gawd
Joined
Sep 16, 2011
Messages
163
I'm going to be doing a build around Christmas (my GF is gifting me some parts to complete it) and I am definitely concerned about ESD (I've zapped ram before being an idiot about ESD). I have hardwood floors to work on and I'll put down foil first to set everything on and connect the case to the foil with a copper cable (pickup some cheap 14 gauge speaker wire from home depot) and I'll ground myself to the case via either a ghetto (aluminum foil bracelet + more speaker wire) or legit arm band. Is there any way to safely connect my foil base to the ground of my apartment? I can of course install the PSU first and plug that in to ground the case, but is it possible to just hook my foil base up to the third prong in an outlet? I guess what I'm really asking is can I get some sort of adapter that would plug into the socket and just provide a screw on point for the ground so I don't have my GF worrying I'm going to kill myself putting a copper cable into the socket?
 

DejaWiz

Fully [H]
Joined
Apr 15, 2005
Messages
20,384
Yes, a wire connector with a banana plug end would fit right in to the outlet ground plug.
 

Elledan

[H]ard|DCer of the Month - April 2010
Joined
Oct 18, 2001
Messages
15,885
Why would you want to ground on outlet ground, though? The only thing you're trying to prevent is to get a voltage potential building between you and the item you're working on. Creating a lower potential by connecting the whole setup to the outlet's ground just seems redundant and possibly dangerous.
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
It isn't dangerous to ground to an outlet (provided the outlet is correctly grounded).

Your also trying to create a low potential between the parts, the item you're working on, and yourself. You pick up a part out of a container (or package, or bin, or desk) and install it in the device you're servicing. If there is a potential difference between you and the device, you and the part, or the part and the device, then you've got a chance at ESD damage.
 

PedroDaGr8

Limp Gawd
Joined
Feb 17, 2004
Messages
489
THis is not quite true. If a device somehow gets a high voltage fault (say a flyback starts faulting to the chasis or maybe lightning strike) which travels through your ground you will get a momentary point where your ground is no longer at earth. It can be quite high though for a brief moment. The closer you are/the larger the gauge of your ground stake, the less of an issue this can be but still it could theoretically cause some issues.
 

SarverSystems

Supreme [H]ardness
Joined
Feb 9, 2002
Messages
7,408
That's is why a; ESD writestraps have a resistor inside them. They are also made of semiconductore material, not metal.
 

lasserith

Limp Gawd
Joined
Sep 16, 2011
Messages
163
So then it would be worth it to actually get a legit strap instead of just having an aluminum bracelet with a copper wire attached eh? And don't hook up to an outlet's ground?
 

Elledan

[H]ard|DCer of the Month - April 2010
Joined
Oct 18, 2001
Messages
15,885
It isn't dangerous to ground to an outlet (provided the outlet is correctly grounded).

Your also trying to create a low potential between the parts, the item you're working on, and yourself. You pick up a part out of a container (or package, or bin, or desk) and install it in the device you're servicing. If there is a potential difference between you and the device, you and the part, or the part and the device, then you've got a chance at ESD damage.

That's a good point. Hadn't thought about it like that yet :) I can indeed see how you'd rather have the high potential for the IC you're picking up rather than for your hand ;)

@lasserith - Yes, it's most definitely worth it to get a real ESD strap rather than DIY something :)
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
So then it would be worth it to actually get a legit strap instead of just having an aluminum bracelet with a copper wire attached eh? And don't hook up to an outlet's ground?

A proper strap will have a resistor in it that prevents a lot of current from flowing, as well as the very sudden discharge of current. Using a bracelet will avoid ESD, but it puts the technician at risk should they come in contact with an energized circuit.

That's is why a; ESD writestraps have a resistor inside them. They are also made of semiconductore material, not metal.

LOL, no. They're made of metal; the drain wire usually goes to an anchor on the strap which contacts your wrist directly. Semiconductors are generally crystalline and very brittle so making a flexible strap out of a semiconductor wouldn't work.

Further, semiconductors have polarity -- they wouldn't drain or source potential if you were charged one way or another, and still pose the threat going the other way.

You've been posting misinformation like this here for more than seven years; the lack of active moderation in this forum has left this useless thread a sticky. Really, static electricity is the stuff of elementary school science class.

That's a good point. Hadn't thought about it like that yet :) I can indeed see how you'd rather have the high potential for the IC you're picking up rather than for your hand ;)
Looking through a catalog that sells professional ESD gear can offer a great education on ESD setups. In a lab handling sensitive equipment, a technician might use a mat, a wrist or a shoe strap, a monitor, and an air ionizer.

ESD tools -- from slow discharge pliers to ESD-safe soldering irons -- help keep charges from building up on other objects. The mat keeps components and tools on the bench safe, and the air ionizer makes the air itself conductive enough to dissipate charge without adding humidity.

Charge monitors can help audit for quality assurance, but they also check that charges aren't building up -- indicating a faulty connection or detached plug - before damage can occur.
 
Last edited:

DejaWiz

Fully [H]
Joined
Apr 15, 2005
Messages
20,384
So then it would be worth it to actually get a legit strap instead of just having an aluminum bracelet with a copper wire attached eh? And don't hook up to an outlet's ground?

Well, getting a legit strap is always a good idea since they're cheap as dirt, but the simplest way to prevent sudden discharge is to ensure that there is a grounded circuit path so ESD doesn't occur.


Wrist Strap -> Chassis or ESD Mat -> Ground Point.
 

lasserith

Limp Gawd
Joined
Sep 16, 2011
Messages
163
So if the parts are so sensitive to esd how is it possible to even safely remove the parts from the anti static bags? If I am at the same potential as the mat I'm working on I'm not at the same potential as the mobo/graphics card/ what have you. Thus when I open the bag/whatever it ships in and grab the part can't I zap it then?
 

Mohonri

Supreme [H]ardness
Joined
Jul 29, 2005
Messages
5,751
To be honest, the risk from ESD is much smaller than people make it out to be. Certainly grounding yourself to the computer chassis is good. Once you touch the anti-static bag, it's safe to take stuff out of it--it's anti-static because it's conductive. The danger is really if you have a static charge built up on your body and you touch something inside the computer without grounding yourself to the chassis first.
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
Vague appeals to the extreme, like "smaller than people make it out to be" or "so sensitive" don't really teach us anything.

It's an undeniable fact that metal-oxide semiconductor parts can be damaged by static discharge. It's a lot easier to spend a few bucks for a wrist strap -- and never worry about ESD -- than it is to try to quantify the size of the risk or the sensitivity of a family of devices. Even if one manages to do the later, they're left either measuring static fields and taking action to avoid them, or using the same precautions they meant to avoid, anyhow.

Built-in defenses to static discharge within devices have gotten better over the years, but damage to discharge is still cumulative, and still does occur.

Touching an anti-static bag doesn't mean it's safe to remove the device. Say the handler has a very high static potential. When they touch the bag, they may dissipate some of that static potential to the bag (and the device inside), but they may not fully dissipate their static potential.

The handler and the bag may be left at a high potential, though not as high as the handler had before touching the bag. When removing a part from the bag, the high potential of the bag, the part, and the handler might be relieved when the part comes in contact with the work surface or the assembly of interest. If that charge is released through the device, the device is at risk of damage.

One of the points brought up repeatedly in this thread is that static damage isn't always detectable. Maybe the device ends up operating well enough and remains in spec; maybe it doesn't; maybe it completely fails. Analysis of a functional problem requires expensive equipment and lots (and lots!) of time; it's far easier and cheaper for end users to just RMA the part and get a new one.

The fact that static damage is difficult to reliably detect and diagnose ends up meaning that casual empirical analysis of the risks is also very likely to be incorrect. Assertions like "I don't wear a wrist strap and I have never damaged anything in 20 years" are ignorant, and the belief that if one person doesn't have a problem over a long period of time that nobody else will -- or even that other people are highly unlikely to do so -- is foolish.
 

MrAgmoore

[H]ard|Gawd
Joined
Jun 5, 2008
Messages
2,000
1)

We've known about electrostatics, in the field of Physics for a while now...

Rub two surfaces together and there's an electron transfer between the materials:

http://en.wikipedia.org/wiki/Electrostatics#Charge_neutralization


2) Electronic devices operate on the scale of micro volts ( 0.000001 Volts ). If your electronic device fails, you won't see an atomic bomb going off.

3) You are right to demand proof of the existance and effects of ESD. When I studied Advanced Physics, almost everything was proven experimentally. People have a hard time believing or visualizing things that they cannot physically see. It takes years of training to understand esoteric stuff in the field of Physics or Electronics. This thread is just an informative shortcut.

Somebody mentioned that they have seen pictures, that reveal the effects of ESD, through an electron microscope so that might be something to search for on Google?
 

mtrupi

Gawd
Joined
Mar 26, 2007
Messages
756
1)
Somebody mentioned that they have seen pictures, that reveal the effects of ESD, through an electron microscope so that might be something to search for on Google?

Yes, a failed part can be sent to a failure analysis lab. An electron microscope is one tool used to find signs that it could be damage due to ESD.
 

Mohonri

Supreme [H]ardness
Joined
Jul 29, 2005
Messages
5,751
2) Electronic devices operate on the scale of micro volts ( 0.000001 Volts ). If your electronic device fails, you won't see an atomic bomb going off.
I do believe you're confusing Volts with Amps there--0.000001 Volts isn't nearly enough to turn on a transistor, let alone damage it. :)

//electrical engineer
 

Elledan

[H]ard|DCer of the Month - April 2010
Joined
Oct 18, 2001
Messages
15,885
I do believe you're confusing Volts with Amps there--0.000001 Volts isn't nearly enough to turn on a transistor, let alone damage it. :)

//electrical engineer

Not yet, anyway ;)
 

Veeb0rg

2[H]4U
Joined
Dec 31, 2000
Messages
3,278
I do believe you're confusing Volts with Amps there--0.000001 Volts isn't nearly enough to turn on a transistor, let alone damage it. :)

//electrical engineer

Last time I tried to turn on a transistor, it demanded dinner AND a movie first. :p
 
Joined
Feb 24, 2013
Messages
13
So when it comes to ESD, as far as I understand, you have to have a mat that connects to the wall through a plug. But is this only valid in the UK, which has a third line that I believe grounds the current? What about in Europe where there are only two rods in the plug? Neither of those rods go to ground in the wall, they both belong to the current, so will the ESD mattress still work and provide the current with a way to escape?
 

Mohonri

Supreme [H]ardness
Joined
Jul 29, 2005
Messages
5,751
The mat just needs to be grounded to something. If your electrical system doesn't have a ground, the next easiest thing is a copper water line or gas line.
 

mikeblas

[H]ard|DCer of the Month - May 2006
Joined
Jun 26, 2004
Messages
12,775
If your electrical system doesn't have a ground, then the gas and water lines probably aren't grounded, either.
 
Top