Re: [CAD_CAM_EDM_DRO] Re: help with g201 needed

mariss92705 wrote:

> Hi,
>
> I have to make a comment here. Electrostatic discharge is NOT a
> concern with opto-isolated inputs, particularly in the GXXX drives.
>
> The opto diode has a 1K resistor in parallel with it and that
> combination is in series with a 200 ohm resistor. The impedance is so
> low as to make ESD damage impossible in any practical scenerio.
>
> One of three things could have happened:
>
> (1) Excessive forward or "on" current. When that has happened the 200
> ohm resistor has fried. Not the case here.
>
> (2) Excessive reverse voltage. LEDs have a reverse or breakdown
> voltage of about 5 to 7 volts. Voltages in excess of that will
> quickly destroy the LED.
>
> (3) The part was flawed from the manufacturer and failed after a
> period of time. That ocassionaly happens (<< 0.1%).
>
> For this particular drive (1) was ruled out by inspection. Though
> possible, (3) is unlikely; I have seen that only twice in over 10,000
> opto-isolators used. That leaves (2) as the most likely cause, though
> it remains undetermined.

Well, that seems a reasonable assumption. But, you can't say it wasn't
ESD! Yes, if the opto device with the 2 resistors is isolated in the middle

of a table, it should be impossible to pop it. But, when connected to a
system, and an ESD event ocurrs to one of the wires, all you need is
5 mA to exceed the reverse standoff of the LED. You can easily get
hundreds of mA is a low-voltage ESD event, and amps in a worst-case
situation.

Also, in a completed device, if an ESD pulse is applied to one terminal
of the opto circuit, current can jump from the other opto terminal to
some other component in the device. The opto-isolators are only good
to 5 KV or so. In a completed SYSTEM, exceeding those potentials is
pretty unlikely, but a charged hand touching the opto terminals could
deliver voltage above the standoff rating, and discharge across the
isolation barrier.

>
> About ESD in general:
>
> Small-signal MOSFETs without intrinsic gate protection zeners are the
> most sensitive and must be handled with care. Bipolar devices (like
> LEDs) are insensitive to common ESD and require no special handling
> in my experience. In between are unterminated CMOS logic inputs of
> which I have seen less than a dozen damaged by ESD in the last 20
> years.

ESD is fairly rare, and since I live in humid St. Louis, it is practically
nonexistant in the warmer months. But, I have certainly had ESD
events. I have blown out chips on 2 or 3 occasions when touching
sensitive circuits after picking up a charge. I have also had circuits
survive AMAZING ESD events, of the sort where there are visible
sparks, audible snaps, and strong muscle response to the current
through the arm. I have also had a few induced currents from nearby
lightning blow out gear all over the house, or just scramble operating
equipment so that it needs to be reset.

Jon