ESD: Truths, myths, and flat out lies

Discussion in 'Electronics' started by SarverSystems, Sep 25, 2004.

  1. mikeblas

    mikeblas [H]ard|DCer of the Month - May 2006

    Messages:
    12,782
    Joined:
    Jun 26, 2004
    A two-prong outlet doesn't have an electrical ground. It's just hot and neutral. If the house is wired correctly, neutral is at ground potential, but it might not be. If there's knob-and-tube wiring, there's no ground at the receptacle box. If there's BX, there might be a bonded connection through the jacket of the BX connection, but that's bonding and not grounding.

    An upgrade that adds grounds to a water pipe isn't adding a ground; it's adding a connection to a water pipe. The pipe may or may not be at ground potential, depending on if it is continuous to an earthed connection. Tying grounds to a pipe is one of the ten worst grounding mistakes you can make. The NEC disallows the use of a pipe as a grounding point unless the attachment is within five feet of the grounding rod connection.
     
  2. workshop35

    workshop35 Limp Gawd

    Messages:
    488
    Joined:
    Nov 24, 2013
    well it definitely seems like more of a bad practice than i thought, so thanks for straightening me out. I'll be sure to drive a dedicated ground stake and run a proper ground if i ever have to do this. Thankfully my house doesnt have these issues.
     
  3. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    I know. I used to manufacture computers and servers. My point is that the component handling surfaces are grounded before the components are brought in. ...the components aren't attached to the conduit to ground before the conduit is actually attached to ground.
     
  4. workshop35

    workshop35 Limp Gawd

    Messages:
    488
    Joined:
    Nov 24, 2013
    Lol i think at this point we are actually in agreement and we're coming at it from different situations. i typically am dealing with propellant/rockets where i am not at the component level all the time but rather grounding the entire unit. In these situations equalizing your potential at the unit itself can cause quite a substantial esd event.
     
  5. AaylaSecura

    AaylaSecura n00b

    Messages:
    5
    Joined:
    Jul 27, 2015
    Greetings! I am a bit confused about something: say I touch a sensitive device (CPU, MoBo, etc) while it is not plugged in anywhere and is on an insulating surface - that charge will stay there (not causing any damage) until I ground the device (by connecting it to the plugged in power supply) and that is when the charge will flow through it into the ground potentially damaging it, correct? In case this happens (I do transfer charge onto an ungrounded device) is there a safe way to discharge it before plugging it in?
    Another thing I'm wondering is can simply covering the device with regular (transparent) bubble wrap induce any charge on it - it's not rubbing, so my guess is it's safe? I ask because I assembled the setup onto a worktop to test it but there was a delay in getting some of the parts needed to finish the build, so I simply covered the setup with bubble wrap (didn't have anything else around!) to keep it from dust.
     
  6. mikeblas

    mikeblas [H]ard|DCer of the Month - May 2006

    Messages:
    12,782
    Joined:
    Jun 26, 2004
    No, that's not correct.
     
  7. Nobu

    Nobu 2[H]4U

    Messages:
    2,530
    Joined:
    Jun 7, 2007
    If there is a great enough difference in charge to damage the device when you touch it, then once you touch the device the damage (if any) is done whether it was grounded or not--there's nothing more you can do at that point.

    The best thing you can do is reach the same potential as the device before contact (both you and device are on a common ground prior to touching)--the computer case is usually a good ground, but if we're talking about a new component in a sealed bag, then the surface of the bag (prior to opening the bag, and while removing the component) would be fine, though it'd be more ideal to be touching the case at the same time as the bag, since the case (if the psu is connected/installed) is connected to building ground, which should be at about the same potential as what the component was when it was packaged.
     
  8. AaylaSecura

    AaylaSecura n00b

    Messages:
    5
    Joined:
    Jul 27, 2015
    Thanks for the explanation! I did ground myself while working with the setup at all times, the reason I asked this question was about the bubble wrap cover - was it a bad idea, should I have used cotton or paper sheet (I mean, yes I've done it now but I'd like to know whether it's bad practice in general)?
     
  9. Nobu

    Nobu 2[H]4U

    Messages:
    2,530
    Joined:
    Jun 7, 2007
    ESD bags are specially designed to allow the flow of electrons so that a charge does not build on the surface of the bag (sort of like how teflon on a pot keeps things from sticking), whereas regular packing bubble wrap (the clear stuff) isn't. That doesn't mean it's inherently dangerous, but given the right conditions it could cause damage.

    However, even with an actual ESD bag you could cause damage if you don't use it properly--for instance, if you have the ESD bag sitting on a carpet that has a static charge, the ESD bag can allow that charge to easily flow to whatever it's touching. But whenever it's closed and sealed, its perfect for electronics, because a charge cannot build on the inside of the bag and it can be equalized with the outside by simply placing the bag on a grounded ESD mat.
     
  10. SarverSystems

    SarverSystems [H]ardness Supreme

    Messages:
    7,405
    Joined:
    Feb 9, 2002
    Also, the ESD bags themselves will not generate its own ESD.

    Bubble wrap creates its own charge just by friction of the bubble wrap against itself.
     
  11. AaylaSecura

    AaylaSecura n00b

    Messages:
    5
    Joined:
    Jul 27, 2015
    As I thought. Thank you for the explanation!
     
  12. SarverSystems

    SarverSystems [H]ardness Supreme

    Messages:
    7,405
    Joined:
    Feb 9, 2002
    The true ESD bags (silver in color) are actually made of 3 layers.

    Outside layer is a semiconductor material, made to disipate any outside ESD charge to the surface of the bag.

    Middle layer, I'm not sure if this is a conductive layer, or a non-conductive later.

    Inner layer is the same as the outer layer.

    ESD-resistant bags are NOT the same thing, and really not much good for anything. They can be easily identifiable because they are usually pink in color. They do not protect the contents from an ESD charge. The only ESD properties they have is that they do not produce their own ESD charge under normal cicrumstances.

    Both of the above statements can be proven with a simple test.

    Get one of each bag.
    Please a 9V battery on top of each bag, with both the + and - leads touching the bag.
    In about 1 hr, the ESD bag will be warm, and the 9V battery will be dead.
    The ESD-resistant bag will have no affect on the battery.
     
  13. mikeblas

    mikeblas [H]ard|DCer of the Month - May 2006

    Messages:
    12,782
    Joined:
    Jun 26, 2004
    I'm convinced that you don't know what "semiconductor" means.
     
  14. w1retap

    w1retap [H]ardForum Junkie

    Messages:
    11,768
    Joined:
    Jul 17, 2006
    Just got kicked out of a Facebook pc-related group for teaching people proper ESD safe techniques for PC repairs. (I'm ESDA certified, perform rework all the time) The 12 year olds running Facebook groups now seem to think it is a myth. :p
     
    DejaWiz likes this.
  15. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    Sad.
    Even sadder that a measurable percentage will start having all sorts of problems in the future from erratic operation to outright failures, but blame the brands instead of their own unsafe handling procedures.
    Ignorance is bliss? Not in this case.
     
    w1retap likes this.
  16. w1retap

    w1retap [H]ardForum Junkie

    Messages:
    11,768
    Joined:
    Jul 17, 2006
    Yea it's pretty crazy. I guess the 'newer' generation of kids are just extremely uninformed/uneducated. I was offering a service (for free minus shipping costs) to de-lid someone's 7700k for them I stated I'm EDSA and IPC rework certified, and I have an ESDA compliant workbench. Then two or three people jumped in saying how dumb I was, and they've been building PC's for years without problems on carpet. (mind you two of these were moderators/admins) I never got hostile, just explained to them that it isn't always a flat out failure from ESD, but rather degradation which causes failures down the road. The next thing I know, I can't get into the group anymore since I was banned. :woot: But, that's the level of people we're dealing with today. If they can't see it with their eyes, it must be fake despite science and documented evidence. Even helping people these days is hard.. haha.
     
    El Derpo and DejaWiz like this.
  17. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005

    That's exactly the problem as you said: if they can't see it and it hasn't instantly died completely, then in their minds, static discharge damage doesn't exist at all.

    I've tried, to great extent, to explain that the human body can generate a static charge from simply moving an arm in a sleeve, and that discharge won't be close to producing a zap/spark like grabbing a door handle after walking across the carpet in fleece jammies as a kid, but it's still more than enough to instantly course through an IC and degrade the integrity of internal interconnects and transistors.

    Many still don't seem to grasp the concept of:
    1. Producing a spark takes about 10K-30K volts of static charge in the human body.
    2. An IC can be damaged with as little as 10-25 volts of ESD, which will not ever be sentiently detected, like the spark jumping to a door handle would be.

    Again, it goes back to the original point that you and many others have stated: many people have the mindset of "if I can't see it or feel it and my PC component hasn't outright died, then it never happened, so the entire concept *must be* fictitious.". Then in two or three years when their component does finally shit the bed, they blame the brand.
     
    w1retap likes this.
  18. matt167

    matt167 Gawd

    Messages:
    674
    Joined:
    Dec 18, 2016
    I remember years ago when it was outright easy to just fry a motherboard by mishandling when first building. Hardware seems to have gotten more resistant to damage. don't hear about it as much. My dad always told me to put the Ram in very last, hold every card and mobo only from the edges, and always touch the case. That's how I learned so that is what I do.. I've built a lot of computers through the years and have yet to see any effects.
     
  19. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    It's just the opposite...as die processes shrink (nm gets smaller), then the IC is more susceptible to ESD, since normal operating voltage design has also decreased.

    You wouldn't pump 3-5 volts through a current gen processor or memory module that's designed to only run with a max of 1.1-1.7 volts, would you? So why would you think an instant jolt of 10-25 volts would be just fine?
     
  20. matt167

    matt167 Gawd

    Messages:
    674
    Joined:
    Dec 18, 2016
    No, but I still follow the old way of assembling computers to prevent ESD.. But I have not heard of flat out killing a mobo or other part with ESD in years. But experience tells me it did happen. Lucky when I did it, it was an 80386 system in about 1995. perhaps it's because there was a lot more to it in those days, with all the I/O cards, and then having to set jumpers and stuff. A lot more chances for ESD transfer
     
  21. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    You need to get it into your head that probably 90% of ESD occurrences won't outright instantly kill a component, but instead degrade the circuitry inside any of the surface mount ICs affixed to their respective PCB, causing reliability issues (eventual erratic operation, and then shortened life span leading to component failure/death down the road).

    Let my analogize it like so: If you run your car (the PCB) with a fourth to half its oil (ESD discharge), then it will likely seem to operate normally for a while, probably for a couple/few years, even. But the components making up the rotating assembly (interconnects and transistors) are deteriorating much faster than normal while operating, and eventually the engine (IC) will fail, again much faster than normal. Just because vehicles are manufactured with much tighter design tolerances today than they were 20 years ago doesn't mean they are more resistant to abuse.
     
    Last edited: Dec 30, 2018
    w1retap likes this.
  22. w1retap

    w1retap [H]ardForum Junkie

    Messages:
    11,768
    Joined:
    Jul 17, 2006
    A lot of modern designs for circuit boards do have shunts built into the most sensitive I/O circuits, but it usually only protects to a few hundred volts, and proper handling with ESD preventative techniques should still take place. It's very easy to generate 1000V+ and not even know. A lot of people say touching a metal object helps, but all you're really doing is transferring voltage potential from one object to another since they're likely a floating ground or not grounded at all.. plus you can generate thousands of volts again in the matter of seconds from just moving your body against your clothes.
     
    DejaWiz likes this.
  23. matt167

    matt167 Gawd

    Messages:
    674
    Joined:
    Dec 18, 2016
    I'm actually agreeing with you.. But I'm also saying there was a time that I remember it was pretty easy to zap a motherboard into complete nonfunctional status... I just take the proper precautions and have done so for 20 years.. I've never had RAM issues, and had exactly one video card die weeks after install. All my systems have been stable for years. So I'm either lucky or my precautions work
     
    DejaWiz likes this.
  24. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    Glad to hear it, since precautions are as simple as a $5 wrist strap, properly connected. After all, what's another $5 added to hundreds or even thousands of dollars worth of components?
     
    SticKx911 likes this.
  25. matt167

    matt167 Gawd

    Messages:
    674
    Joined:
    Dec 18, 2016
    Exactly. Although i would check continuity on a wrist strap before use. I had one with no continuity between clamp and wrist contact. I dont use one every time. But always touch the case.
     
    DejaWiz and mnewxcv like this.
  26. mnewxcv

    mnewxcv [H]ardness Supreme

    Messages:
    6,220
    Joined:
    Mar 4, 2007
    No problems here touching case frequently throughout the build process.
     
  27. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    That's a placebo effect.

    ...unless the power supply is installed in the case and plugged in to an outlet, then the case is an active ground. But continual contact with the case needs to be made.
     
  28. mnewxcv

    mnewxcv [H]ardness Supreme

    Messages:
    6,220
    Joined:
    Mar 4, 2007
    yes the case power supply is plugged in (switched off) aka grounded. As for continuous contact, that depends, but I usually rest my bare forearms on the edge of the case when doing stuff on the motherboard.
     
    DejaWiz likes this.
  29. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    *Thumbs up
     
  30. mikeblas

    mikeblas [H]ard|DCer of the Month - May 2006

    Messages:
    12,782
    Joined:
    Jun 26, 2004
    I don't think it's a placebo. I think it's a broad conclusion based on a very narrow sample.

    More inappropriate projection of small sample sets. It's funny how it's so hard to escape.
     
  31. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005

    Here are some things that ESD non-believers need to take to heart:
    How many brand new components are packaged WITHOUT static discharge prevention methods (many including a warning label)?

    How many component manufacturing/assembly plants HAVEN'T had millions of dollars worth of static discharge prevention methods spent on them?

    Going through EE schooling, ESD prevention was one of the FIRST things taught to us.

    As an OEM repair service provider, it's still taught in certification/recertification courses.

    [edited due to my misunderstanding]
     
    Last edited: Dec 30, 2018
  32. Nobu

    Nobu 2[H]4U

    Messages:
    2,530
    Joined:
    Jun 7, 2007
    I think he was agreeing with you, except that he doesn't think it's all placebo. There is a possibility (although small) that some could get away with little or no ESD damage despite not taking proper precautions, just like ESD damage is still possible even when taking all possible precautions. You won't know until something goes wrong or you put it under a microscope, either way.
     
    DejaWiz likes this.
  33. w1retap

    w1retap [H]ardForum Junkie

    Messages:
    11,768
    Joined:
    Jul 17, 2006
    It's like that almost every single time ESD is mentioned in any computer group. Go ahead and post elsewhere/anywhere and you'll see what I mean. I 100% guarantee you'll get jumped on by 3+ people within 20 minutes saying the same thing I quoted.
     
    DejaWiz likes this.
  34. MMitch

    MMitch Limp Gawd

    Messages:
    443
    Joined:
    Nov 29, 2016
    Wow, glad to read this thread. So many RCA (Root cause analysis) on damaged boards in very high end electronic ends up with ESD damage from user. Always [not] fun to teach those using these instruments that they're the cause of the ten's of thousand of dollars bill.
    I didn't read all the thread but if I may add, a simple wrist band grounded (with proper resistance to ensure charged device doesn't discharge too fast) may not be enough. A proper ESD safe area should be prepared. I mean, proper humidity level, nothing (when possible) that can generate ESD, usage of ESD-safe tools, ESD mat (also properly grounded with resistance), ESD smock (this may be overkill for PC builder and could be replaced with clothes that are more ESD friendly).

    You don't need to invest that much to be protected. Clean table (nothing else on it but ideally an ESD mat), ESD wrist wrap (I prefer ESD floor with shoes while standing, I use wrist wrap only while sitting even when using ESD chair), and humidity above 20%. This should be easy to do for everyone and ensure you don't need to use RMA after 1.5year for no apparent reason while blaming company XYZ for their crappy product ;) If I may add, also test ESD precautions now and then to ensure they still work and aren't intermittent.

    Be careful when using the true ESD bags, silver ones, as soon as you ply them or crush them per say, you must trash them as they're not ESD safe anymore. I avoid re-using those especially when they went thru a shipping department...
     
  35. DejaWiz

    DejaWiz Oracle of Unfortunate Truths

    Messages:
    18,392
    Joined:
    Apr 15, 2005
    Ah, then I misunderstood his post. Makes me out to be the one trying to argue...

    mikeblas sorry, dude!
     
  36. BlueFireIce

    BlueFireIce [H]ardness Supreme

    Messages:
    6,003
    Joined:
    May 29, 2008
    I think the problem with ESD is that in many cases you can get away with zero protection in a build and never anything goes wrong, and when they DO go wrong, it is so far beyond the tool set end users have to determine the problem was from ESD to start with that they just assume it's just a "POS part". Which stems from people assuming that an ESD event has to result in the intimidate "frying" of the part.