General ESD Questions

Discussion in 'General Electronics Chat' started by elec_system_design, Aug 18, 2017.

  1. elec_system_design

    Thread Starter Member

    Jun 23, 2017
    49
    0
    Hi there,

    I've been wanting to set up my electronics lab for a while now, but it turns out that my power outlet doesn't have proper grounding.

    I have an antistatic mat, but without proper grounding, I'm not sure how viable of a lab I have.

    What circuit components are most susceptible to damage by static?

    Also, someone once told me that antistatic mats didn't need to be properly grounded to work, is that true?

    Thanks
     
  2. BR-549

    AAC Fanatic!

    Sep 22, 2013
    4,065
    1,059
    What country are you in? What's your line supply? What kind of structure is the lab in?
     
  3. elec_system_design

    Thread Starter Member

    Jun 23, 2017
    49
    0
    I'm in the US. It's just that my house is really old, so the wiring is outdated.

    My lab is just part of my desk, but with an antistatic mat covering part of the table.
     
  4. Sensacell

    Moderator

    Jun 19, 2012
    1,980
    613
    Humidity is a major factor, if you live in a humid area, you can all but ignore ESD as an issue.

    But... If you live in the mountains with dry air and a cold winter- ESD can become a major issue requiring carful adherence to protocols.

    Remember that all grounds are relative, your goal is to keep yourself, your bench and the thing you are working on at the same voltage potential, a perfect "ground" is not required to achieve this.
     
    DickCappels likes this.
  5. phranzdan

    Member

    Aug 4, 2017
    40
    3
    Hi: Mos devices are most susceptible to static. Power Mos, CMOS IC's having a gate structure are easily damaged by static and unsafe handling. If no other source of grounding is available you should look for a cold water pipe preferably copper. Whatever the material, clean the surface thoroughly. use #14 or larger bare copper conductor to bring the ground to your bench. This should provide a low enough impedance ground for safe handling of mos devices. In general, most other devices such as diodes, junction transistors, etc. have a high enough static sensitivity rating to not need special handling.
     
  6. dl324

    AAC Fanatic!

    Mar 30, 2015
    7,215
    1,673
    What is preventing you from correcting the problem? If you have 3 prong outlets without proper earth ground, they need to be labeled as such.
    I don't use an antistatic mat at my bench. I ground myself before handling any static sensitive components.
    Anything with MOSFETs and some white LEDs.
    I most often use my antistatic mat without it connected to earth ground. If I bother wearing a grounding strap, I connect it to the grounding strap from the mat. Otherwise, I touch the grounding strap from the mat before touching components.

    Even when the mat and I are grounded, I refrain from touching component leads whenever possible.
     
  7. Reloadron

    Distinguished Member

    Jan 15, 2015
    3,612
    1,620
    OK, purely in the interest of keeping things simple here in the US. Start with a trip to Home Depot, Lowe's or the home improvement store of your choice. Buy a standard everyday off the shelf Galvan 0.625-in x 8-ft Copper Grounding Rod. Drive that rod into the ground and also get a ground rod clamp, also a off the shelf item. That becomes your ground tie point for your ESD work station. Depending on construction you can also use any entry cold water pipe. The idea is to be able to have someplace at ground potential to dissipate any Electro Static to.

    As to ESD sensitive components? They should be labeled as such from the manufacturer.

    Ron
     
  8. DickCappels

    Moderator

    Aug 21, 2008
    4,958
    1,508
    You can damage a whole range of semiconductors with ESD. That includes simple diodes and bipolar transistors -ESD events can increase leakage currents which might or might not be a problem.

    As Sensacell said, if you live in a high humidity environment you don't have to worry as much as you would when you live in a place with dry air.

    Besides grounding, it is a good idea to use shorting wires on MOSFETs while soldering them. If you ground your soldering iron tip then it is less of an issue. If you use a reflow oven you have even less to worry about.
     
  9. ian field

    AAC Fanatic!

    Oct 27, 2012
    6,205
    1,117
    MOS devices have very high impedance inputs - some pretty impressive static charges can develop on the pins.

    The damage is most often done when you short out that charge. The first order of the day is a conductive rubber mat for your work area. Its has a fair bit of resistance and discharges things more gently.

    If your mat is earthed - you could zap any part you touch after walking across the carpet. The important bit is to make you the same potential as the mat - the usual way is an anti-static wrist strap grounded to the mat.

    Actually earthing the mat is somewhere down the list. For one thing - with wrist strap hooked to an earthed mat and touch something live; you're a gonner! Usually the mat is earthed via about 1M, if you find any current trying to get home - its unlikely to kill you.

    As long as you stick to those rules - an earth of some sort ties up loose ends. If you don't have mains earth, use a copper water pipe - failing that; drive a copper earth rod into the ground under a convenient window.

    A trick shown to me early in my career is to pick up the device only by the plastic package and breath on it - the moisture in your breath condenses on the device and helps leak away static charges. That isn't a substitute for correct handling - but I get away with it quite a lot..........
     
  10. Tonyr1084

    Well-Known Member

    Sep 24, 2015
    2,221
    492
    A lot of opinions stated here. Well, maybe mine is just one more, but I'd consider my knowledge and experience with ESD to be pretty reliable. While many use common practices and don't experience any problems, the truest state of ESD control isn't always about preventing a device from being blown out, it's more about latent damage. Let me elaborate: Imagine a highway along a hill side. A rain storm washes part of the hill down and takes out a lane or two of the highway, but you can still get past. This is analogous to EOS, or Electrical Over Stress. It doesn't result in a failed component but it DOES reduce it's life expectancy and it's capacity. Just like that highway that you can still get through, electrons and the signals they carry can get through. But just like that highway, traffic will have to move a little slower to get through. In electrical terms that's the weak spot and it will generate more heat than the rest of the component. So EOS is among some of the truest concerns with ESD. Actually a full fledged failure due to ESD is better than an EOS event. Suppose you're flying on a plane and some technician failed to follow proper ESD protocol. His circuit didn't fail at bench test, but now it's your butt at 35,000 feet and that device finally fails. So hopefully you now get the point.

    In companies that specialize in military, medical, and even now automotive, ESD practices are strictly adhered to. Electrically conductive flooring and all technicians and other workers wearing heel straps to ground themselves while they walk about to dissipate any charges that can build up on the human body just from moving around. Sitting at my work station, my wife brought home one of those plastic mats for my chair to roll around on. I absolutely refused to use it. So she put it at her computer. Wearing socks while typing at her keyboard she shuffled her feet and BANG - she got a rather big static shock. By the way, I'm in a desert climate where humidity can be as low as 5% at times. 55% is considered a humid day. Whereas when I lived in Houston, 95% humidity was normal.

    When you're sitting at your bench with a static mat, and with your arm laying on the mat - you're essentially electrically bonded to the mat, and any electronic device on the bench is likely to be at your potential. So that when you touch it - there's no high energy discharge (ESD) to blow out the component. But if you shuffle your feet - even on carpet, especially with hard sole shoes, both you and the component can go to much higher potential, and often not at the same rate. So touching a static sensitive device can result in an EOS event. Wearing a wrist strap is advisable, and I have one at my bench. To be blatantly honest, I seldom put it on when putsing around with something. But when I work on a computer or other expensive device I generally ground up just to be sure my equipment is safe. After all, just sitting on a chair and shifting your position - depending on humidity - can generate static.

    The BEST way to prevent ESD and EOS is to wear your wrist strap whenever you're at your work bench. Your work mat is grounded to "The Ground". If you wanted to you could get an ESD mat for the floor and use heel straps too, but to be practical, the home hobbiest isn't going to have conductive flooring, so that when you get up from your work bench to turn and grab a box of resistors you'd need to be grounded. Of course resistors (generally) are not ESD sensitive.

    What would work best for the home hobbiest is to have a grounded mat and a wrist strap. Whenever you're working with ESD sensitive stuff - wearing the wrist strap and having you and the mat grounded is the best way to prevent ESD and EOS from plaguing your projects. Those who ignore that and never see any problems with their electronics - well, there are very many of us who don't strap in and never see a problem. But for the sake of doing it right - at least wear the wrist strap and have the mat grounded. Both you and the mat should be grounded through a resistor of AT LEAST 1 MEG ohm. The mat to ground through 1 meg ohm and the wrist strap grounded through 1 meg ohm. Simply clamping to the mat with an alligator clip is better than not wearing the wrist strap but having a dedicated banana jack (socket) available to plug into is best.

    So, a conductive mat grounded through a 1 meg ohm resistor and a wrist strap grounded TO GROUND through a 1 meg ohm resistor is as much protection as is practical for the home hobbiest. Keeping static sensitive devices in a conductive foam will help protect them as well. Pink Polly (plastic bags) don't generate static but they don't guard against it either. Velostat bags are static shielding.

    If you had a component in a Pink Polly bag, sitting on your grounded mat and you placed your hand above the bag (close to the bag) you can blow out the component because the static field can be strong enough to affect the component. I was working for Northrop Grumman in their Space Technology building. While running a test (not me) on a satellite someone wheeled a cart past the satellite under test. Got within 3 feet of the satellite and the sat immediately failed. The cause was ESD because the cart was supposed to drag a chain on the conductive flooring. But because the chain was rusty it wasn't making good contact and the static field generated by the cart was sufficient to blow out the device under test. YOU DON'T HAVE TO TOUCH A DEVICE TO BLOW IT OUT. Or to give it an EOS event.

    Satellites have to operate on EXTREMELY low power, hence they're far more sensitive to ESD than the stuff you're going to have on your bench. Unless you take something from a satellite. But by the time you get it home it'll probably be dead.

    30 plus years working in ESD areas. Sat through MANY ESD lectures and have given several as well.
     
  11. phranzdan

    Member

    Aug 4, 2017
    40
    3
    Hi Tony: Nice article which I am sure put the fear of god into all who read it. I think there is a huge misunderstanding about ESD relative to bipolar devices, diodes, and most junction devices. After 35 years in the Reliability Engineering Laboratory for major power device manufacturer, I have never seen the analysis of a failed device(MOS not withstanding) related to ESD. BJT' typically withstand somewhere around 4 KV in Human Body Model testing with no parametric degradation. We went so far as to conduct complete reliability evaluations on devices subjected to the testing to confirm that there were no "walking wounded". The main reason is energy. The human body not properly grounded can certainly crank up enough voltage to but a transistor into breakdown but there is not enough energy available to damage the structure. Most bipolar devices can conduct some current in the breakdown mode before damage occurs. Mos devices are a different matter. The gate is an oxide layer which normally has a voltage applied to cause conduction. This voltage is usually 5 or 10 volts for most power mos and parametric testing can be done at 20 volts or so. The current measured for Igss is very low usually nano or pico amps. A 1000 angstrom gate will tolerate possibly 80 volts but taken past this point the oxide ruptures and the device is destroyed. In a multi cell structure it requires only one cell be ruptured to render the device useless. Failure analysis on this device will show EOS with the failure mode to be a gate source short and the failure mechanism being a rupture of the gate oxide. I have been retired for a number of years so I don't know the oxide thickness of the latest devices but was aware of oxides in the 200 angstrom area. I believe this is typical for CMOS devices which would make them very vulnerable to damage. By all means, take every precaution you can relative to ESD. I hope this puts some perspective on the issue.
     
    Tonyr1084 likes this.
  12. KeepItSimpleStupid

    Distinguished Member

    Mar 4, 2014
    2,870
    505
    I think it's more like, label the ones (GFCI) that are not correct. It is permissible to have a 3-prong GFCI outlet non-grounded if properly labeled. e.g. Lacks equipment ground. Tags are usually in the box.

    So, lack of humidity, carpeting, moving around. The key is "same potential". A wrist strap can do that. make you the same potential as the chassis. e.g. touch the chassis first before handling components.

    I think the time I was really worried. I was replacing one of those huge DIP processors in a Fluke meter and sat the PCB on aluminum foil and powered it up. Nothing real bad happened.

    What you might do is to get an isolation transformer for your bench and create the ground for your bench. e.g. connect G and N together on the secondary side. That's permissible. Then use an outlet strip for your equipment. Use a GFCI outlet labeled properly at your bench location.

    In some old houses, the grounds are run to the outlets, but not the receptacles. They were 2 prong polarized. it's important to bond the box and the receptacle. and not rely on the screw.
     
  13. bertus

    Administrator

    Apr 5, 2008
    18,587
    3,600
  14. Tonyr1084

    Well-Known Member

    Sep 24, 2015
    2,221
    492
    While information may vary on the subject, the human body can store up to and even more than 50KVS (Static) charge. It's on those annoyingly dry and cold days that you can turn to get out of your car and when you turn to close it SNAP-BANG! You feel the effects of a good 50KVS charge. It can hurt. But like phranzdan said it's the available energy - the current that does the damage. 50KVS may seem like a lot of energy but it's not enough to kill you when you grab the doorknob of your front door on the way out in the morning.

    One demonstration I did was to take two volunteers, both wearing sweats, both sitting in plastic lawn chairs. Each holding a brass rod and a common ground point for them both to discharge to. With both men sitting still, they each discharged their static to ground, then touched the brass rods together. Then I had one stand up and hold out his brass rod toward the other. When they got within a half inch there was a visible spark between them. Then after they had equalized their charges the standing man sat down and the sitting man rose. They repeated the experiment and this time they got within an inch of each other when the spark flew. I wasn't able to measure accurately the charge they had built up, but I'd estimate it to be over 100,000 volts. It takes about 1200 volts to jump a gap of 0.01 inches in the right conditions. Neither volunteer was hospitalized. But I did get burned. Someone complained I was shocking students and OSHA came down on me pretty hard. That was the last lecture I gave for Baker Hughes INTEQ, a company that made off shore oil rig measurement tooling. Both high adherences to ESD AND to proper wiring and protection against stray sparks. Stray sparks on an oil rig can burn one down rather quickly. And that was where I got the bulk of my experience with ESD.

    I did all the ESD audits of every work station, and I had the power to fire people. One young man was dismissed after I asked him to put his wrist strap on. "Now come on", you're thinking. "Everybody forgets to put their wrist strap on from time to time." Yes, that's true. But after asking him to put his wrist strap on I turned the corner and then came back to check and it was off again. Less than 5 minutes, and he hadn't gotten up. So I asked him a second time and he put it on. I went through the door and then turned to look back through the window and I saw him remove his wrist strap and THROW it to the floor. It was then he lost his job.

    He was wearing heel straps, but he had his feet up on a wooden box under his bench, so he wasn't grounded. But not so much that he was putting components at risk - it was his whole attitude toward ESD and control. He needed to go. At the time it was costing $2,000 per hour to run an oil rig drill. Every time a part failed and had to be retrieved it took hours to remove and replace a defective tool. Trust me, when an oil company complains about failure - you (the company) loose(s) big big bucks.

    That's not the case on most hobby benches. ESD can be controlled, as I said, simply by resting your arm on the mat while you work with things. But how often do we lift our arm off to grab the soldering iron or to reach for something else? Often enough, and coupled with the fact that we're shuffling around in our chairs, and you can generate enough static to cause some harm. How likely is that to happen? Well, here's where I have to estimate, because habit is different from person to person; but I'd estimate that it's about 5 to 10% odds that you're going to damage something. Maybe lower. Maybe MUCH lower, like I said, this is an estimate. Still, the fact that touching the chassis of something before you work with it - a reasonable practice - it's likely that at some point you're going to let go of it. It's during that time, depending on what you're wearing, the surface you're shuffling your feet on, the humidity and a host of other factors, and you CAN damage something even with the habit of touching the chassis first. HEY! I've done that! Grab the chassis or rested my arm on it to maintain a common potential between myself and the frame. It works. Mostly. And I've never experienced a failure from that habit. However, I can NOT attest to any EOS (or walking dead as phranzdan called it) having any effect on any of my circuits. I HAVE had parts fail early for some unknown, unverifiable reason. Could it have been EOS? Maybe. Maybe not. I still support the idea of having a mat and a wrist strap. While wearing the strap you and the mat and everything on it is at the same potential. Well, sort of. Things that conduct are. Things that insulate don't surrender their static charge so easily. So keeping your mat grounded helps. Can your projects survive without being on a grounded mat? Likely they will. But if you want to know accurate information about ESD then I'd suggest grounding your mat and wearing a wrist strap whenever you're working with MOS or CMOS components. They're just too sensitive. OR you could put them at risk. The choice is yours.
     
  15. Tonyr1084

    Well-Known Member

    Sep 24, 2015
    2,221
    492
    One thing that hasn't been mentioned: An Ionizer. An Ionizer produces a high voltage charge on sharp tips with a fan blowing across them. Ion's are stripped away, both positive and negative charges. Since like charges repel, positive ions are repelled from things that have a positive charge and will attract the negative ions, and thus, become neutralized. Same is true of a negatively charged things, they attract the positive charged ions and neutralize their potential charges. I think Ionizers are not very expensive. Good ones may be. But they have to be blowing their ion cloud over your project, which often means blowing over you. The breeze can be nice and cooling in the summer time, but outright cold in the winter. A good Ionizer will have a heater as well. Not to keep you warm but to keep you from freezing to death. Do I have one? Nope. I only have an exhaust fan to strip away smoke from soldering. Blowing toward me when soldering is impractical because I want the smoke to go away from me. But solder smoke is not the big concern here on my bench and likely not on yours either. It's static. And with modern equipment working on ever lower voltages, they're also ever more susceptible to static damage and/or failure.
     
  16. Tonyr1084

    Well-Known Member

    Sep 24, 2015
    2,221
    492
    Are you aware that the human body can barely feel a static discharge at 3,000 volts? And in order to notice such a charge you have to be quite attentive to it. Mostly it can go unnoticed. Another culprit of static can be the old style computer monitors - the heat beasts that eat up valuable real-estate on your workbench. Those big bulky things with um - you remember them? CRT's?

    I mentioned a satellite earlier. Many of their components are susceptible to damage at voltages as low as 50 volts. As for your comment about never seeing a failure analysis, I have. I've stripped the tops off of microchips before in an effort to find the damaged point. Working as a microsection technician (for 4 years) I was hard pressed to discover the failure mode for most of them. But on one I did find the burnt out trace. My microscope was capable of viewing the gold plating on the fingers of an edge connector on a PCB. I successfully measured 25 microns (0.000025 inches). At the time my boss thought it was not possible to take such a small measurement. He snatched my sample away and sent it off to an electron scanning microscope. When he returned he proudly announced that I was wrong. It wasn't 25 microns. When hard pressed for the answer he reluctantly admitted it was 24.6 microns. "OK" I said "For FOUR TEN-MILLIONTHS of an inch I'm wrong." To this day I'm unaware of anyone else making such a measurement optically.

    I think I've gotten off topic. OK, from here out, I'll mostly read. I'll reply to questions put to me, and maybe I'll argue a point or two, but mostly I'll try and stay on topic.
     
  17. Reloadron

    Distinguished Member

    Jan 15, 2015
    3,612
    1,620
    During some of my earlier years, before my involvement with Navy Nuclear Propulsion I was involved with Ordinance from missiles to torpedoes and other fun stuff. Among my responsibilities was working in the "Convair Room" assorted circuit cards were manufactured for the Patriot Missile as well as the Tomahawk Cruise Missile (both air launched and sub launched flavors). If it was ESD related it was part of my life. Everything from floor wax, booties, and the regular testing of ESD work stations. We used the 5 Lb weights, spaced, with a megger to test the conductivity of the mats. The 3M corporation marketed the stuff making a fortune in the process. All of this compliance also applied to the torpedo programs including MK 46, MK 48, MK 48 ADCAP and MK 50. There were also the "Clean Room" standards to comply with.

    While all of this is well and fine I do not see much need for the home assembler, hobby enthusiast building a home workstation. A simple mat with a 10 Meg Ohm resistor to ground and a decent wrist strap with 1 Meg Ohm to ground is about all that is needed. Unless the mat or strap are beat to death they should last many years. The average hobby enthusiast is not building advanced ordinance systems or satellite communications in their garage or basement.

    You want a "ground" just drive a ground rod into the ground and call it good. Get a conductive approved mat and call it good and then a simple approved wrist strap and call it good. I doubt compliance to ISO 9001 or ISO 17025 is required or any other DoE or DoD standards is required.

    Ron
     
    Tonyr1084 likes this.
  18. Tonyr1084

    Well-Known Member

    Sep 24, 2015
    2,221
    492
    @Reloadron Nicely said. Yes, a mat and wrist strap are all that's needed. And the resistors to make sure nobody takes a shocking from mains by accident. Was it 10 MEG for the mat? It's been a while. 120 microamps (or 240 microamps on a 240 mains system) isn't going to hurt anyone. 12 microamps (through a 10 MEG ohm resistor) is even less current. And no home hobbiest should be using anything any higher than that for sure. (the voltages)
     
  19. Reloadron

    Distinguished Member

    Jan 15, 2015
    3,612
    1,620
    <Totally Off Topic But...>
    Yeah Tony, been awhile but I want to recall 10 Meg on the mats and 1 Meg on the wrist straps. In addition each worker who worked the areas had to test their straps after putting them on and if they flunked there was the special ESD lotion. The list of ESD precautions went on and on... :)

    This was about 30 years ago and I was paying $1,000 a roll for the 3M mat material which I want to say was something like 2 feet wide and maybe 50 foot long. We could custom cut our mat material to what we wanted. There was some humor to all of it and not just the ESD parts. People wonder why the US government pays so much for hardware? You want a circuit card? OK, that will be $100 thank you. Oh wait, you want a circuit card built to a MIL-STD and comply with what specifications? Oh, that will be the $1,000 version of the $100 circuit card. You can have whatever you want. :)

    Have a Good One
    Ron
     
  20. GopherT

    AAC Fanatic!

    Nov 23, 2012
    8,025
    6,786
    Well, if you want to learn the answers to your questions, do an experimemt. Buy three MOSFET transistors (at least) then leave one "as is" and put it in your pocket, solder a 10 k resistor between Source and Gate to a second transistor amd put it in your other pocket. Leave the third in the antistatic package.

    After a few days, do an off-state current leakage test on all three.

    Buy a few extras and think of other stress tests for them

    Post your results.

    PS, i have never had froubles with BJT, only MOSFETs (but rarely). I do not use a grounding strap but i humdify my home in the winter months.
     
Loading...