Close
0%
0%

Z80 computer

Z80 computer

Similar projects worth following

A Zilog Z80 based microcomputer of the style of the late 70s.

View all 2 project logs

Enjoy this project?

Share

Discussions

agp.cooper wrote 11/23/2019 at 05:02 point

Useful comments, I have no evidence of destroying a chip out of circuit for obvious reasons (how do you know it happened?). I have destroyed chips though carelessness with a loose power supply lead (obvious), mixing high power devices with "less" tolerance devices (think of a bus conflict between SRAM and an Adruino, the Arduino can supply 40 mA, enough to destroy a SRAM chip). And from static to a chip in circuit!

My understanding, until the decoupling capacitor is in place, the chip is vulnerable. The gate only needs about 300v for puncture and almost no current. Not hard to work out scenarios were static could flow though an unprotected chip (like picking one up!). I used to get an 80v 50Hz signal off my laptop it I toughed it (as measure by an oscilloscope).

It is hard to get a number but chips have a (RMS?) power limit that the protection diodes can handle even with the decoupling capacitor. (Thus the failure in circuit described above).

Long time ago I had a simple bipolar transistor radio controlled car. When the antenna toughed the TV screen (i.e. cathode ray tube), it died. So even bipolar transistors can die from static.

So I think it is very fair to say you will usually get away with ignoring the recommendations but you can be unlucky. In a hostile environment you need to take extra precautions.

But, I think I have lost more chips by soldering them either in the wrong place or upside down!

So, all I am saying is best not take the chips out of the protective foam before installation and to use an earth strap (around your wrist) to eliminate the possibility of a static discharge.

The other issues are "advanced" issues.

AlanX

  Are you sure? yes | no

agp.cooper wrote 11/21/2019 at 00:35 point

I don't know if your are aware or not, but (almost all) ICs are static electricity sensitive. Best not to handle them (i.e. remove them from the conductive foam) until you are really to install them. Even  then you should use an "earthing strap". Not doing so is a risk but usually you will get away with it.
Regards AlanX

  Are you sure? yes | no

Starhawk wrote 11/21/2019 at 03:10 point

While I normally refrain from attempting to contend with, contradict, or otherwise correct anyone who arguably has more knowledge and experience in a given field than I do... I'll be honest with you, that's almost complete crap.

The first person to really explain this to me was my tech shop friend, and he pretty well grew up on Commodore stuff, literally, and out of that came not only his intimate knowledge and considerable plethora of skills with regard to more modern computing, but a considerable and healthy regard for that which came before, both in terms of computing, and in terms of electronics before the two fields were reasonably separable... while I initially chalked his seemingly overconfident-bordering-on-smug disregard for antistatic bags and the like for nearly anything electronic or computer-y in nature, it became apparent over a few years of repeated patient explanations and of my own accumulating experiences, that he had things exactly right.

In a technical sort of sense, you are indeed somewhat correct that static electricity can in certain situations damage semiconductor-based components of any complexity, from discrete diodes on up to processors and FPGAs and the like. However, in actual practice, discrete semiconductors are themselves essentially immune, more by way of simplicity than anything else. TTL and early *MOS processes (PMOS, NMOS, HMOS, etc) that predated CMOS -- chips constructed with those technologies at heart, including most variants (LSTTL, etc) almost universally are also immune, because the way that the inputs and outputs are protected in circuitry, along with the robustness of the manufacturing process of the silicon die itself, results in a chip that is /extremely/ hardy to such things.

In fact, early NMOS 6502 CPUs are not only well known for being able to weather pretty much any sort of incidental static mistreatment (I can't speak to *purposeful* attempts, mind you, such as direct contact with the finger of an aspiring young engineer who's just scuffed himself across the entire living room carpet in the dead of winter because he gets a "charge" of his own (hyuck hyuck hyuck) out of putting a miniature bolt of lightning through Daddy's latest kitchen-table hobby project) but for being so robust indeed that, in many cases, the datasheet-specified pin loading limits are so far below actual capabilities that the applicable rule of thumb is "don't bother with buffer chips on the buses until it genuinely starts to act up".

Additionally, /modern/ CMOS-process chips (I'd say anything made after about 1990 as a general rule, although TBH that's very likely being quite unnecessarily conservative) are robust enough, due in this case mostly to improvements in the process that were made once it became apparent that all the various kinks and traps and whatnot of the initial methodology for making those chips had all been worked out, that, although they aren't nearly as resilient as the older chips from pre-CMOS processes, they're pretty dang close. Enough so that, at least for our purposes here, it's quite safe to say that they are likewise impenetrable to incidental static discharge of the sort that you might encounter on your garage workbench or kitchen table or whatever.

The chips that you specifically need to worry about are in fact the /early/ CMOS chips. Any 4000-series part, or CTTL/HCTTL/etc part, or other known-CMOS-process chip with a date code from 1970 clear through to about somewhere around probably 1985-1987... those need to be treated with extreme care in regards to static protections. Pretty much anything else... naaah. I don't even know where my antistatic wrist strap *is* any more... heck, I think I gave it away at some point a few years ago. Heck with it, you don't really need one anyways... not for this, not for computers. Don't bother. Really. Don't bother.

  Are you sure? yes | no

Ken Yap wrote 11/21/2019 at 04:28 point

Actually the greatest risk is not when the chips are out in the open like in the pictures but when they are fitted or being fitted to a circuit board in use.

Electrostatic damage occurs when the discharge flows through the chip. Out in the open there is no discharge path (unless the chips are resting on a conductive surface). When you pick up the chip its potential rises to that of your hand. You might be tempting Murphy though if you hold one end of the chip and bring the other end of the chip close to grounded metal. Otherwise no need for extreme measures.

But if you're say fitting a DIMM to a computer them it's wise to ensure your body potential is the same as the computer case with a grounding strap. Or at least touch the case before you fit the part. Otherwise imagine what would happen if a spark jumped from a part pin to the PCB. It may survive, parts are pretty tough these days.

Static discharges do weird things to equipment. I remember one Tektronix hard copy printer for their vector terminals in our Uni lab that could be made to print by zapping the corner of the case. No damage though, just wasted paper.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates