When/how to use Zener 5.1v diodes for protection

[Kasterborus]

I have a Atmel MPU input pin held high through a pull-up resistor, a
push button switch pulls it low.

If I want to extend the cable with the push button to about 6' should
I use a Zener 5.1v protection diode?

(This was recommended as a guard against static buildup - I just
wondered if this is the right way to do it)

Thanks

 

[john jardine]

"Kasterborus" <kasterborus@yahoo.com> wrote in message
news:f9f2378a-2d1a-4703-9adc-a465461662b5@c23g2000hsa.googlegroups.com...

I have a Atmel MPU input pin held high through a pull-up resistor, a
push button switch pulls it low.

If I want to extend the cable with the push button to about 6' should
I use a Zener 5.1v protection diode?

(This was recommended as a guard against static buildup - I just
wondered if this is the right way to do it)

Thanks

A Zener is worthwhile.
I've a Mega32 running at the moment, 2 pins on pushbuttons with pullups
enabled. The pullups measure out at 35K ohms.
These will cope with normal amounts of "static" discharges. This leaves the
lighning jolt to the pin when you walk across a nylon carpet and touch the
pin connected wire. The Zener will help, as also a 100nF cap connected (at
the chip) from the pin to 0V.

 

[skeptic]

On Jan 19, 4:53 pm, "john jardine" <john.jard...@idnet.co.uk> wrote:

"Kasterborus" <kasterbo...@yahoo.com> wrote in message

news:f9f2378a-2d1a-4703-9adc-a465461662b5@c23g2000hsa.googlegroups.com...

I have a Atmel MPU input pin held high through a pull-up resistor, a
push button switch pulls it low.

If I want to extend the cable with the push button to about 6' should
I use a Zener 5.1v protection diode?

(This was recommended as a guard against static buildup - I just
wondered if this is the right way to do it)

Thanks

A Zener is worthwhile.
I've a Mega32 running at the moment, 2 pins on pushbuttons with pullups
enabled. The pullups measure out at 35K ohms.
These will cope with normal amounts of "static" discharges. This leaves the
lighning jolt to the pin when you walk across a nylon carpet and touch the
pin connected wire. The Zener will help, as also a 100nF cap connected (at
the chip) from the pin to 0V.

A much better solution than a 5.1 V Zener is a pair of fast diodes.
They should be reversed biased from the input line to +5 V (cathode to
5 V) and from input to ground (cathode to input line). They have a
number of advantages over 5.1 V Zeners. There are no voltage
tolerance issues with diodes as there are with Zeners. The voltage
variation with temperature is far less for diodes than for Zeners.
The current through a positive going spike will dissipate far less
power through the 0.6 V of a diode than through the 5.1 V of a Zener
meaning you're less likely to blow the device with a large spike.
Lastly fast diodes are a lot faster than Zeners and will react to a
fast risetime spike sooner.

 

[Chris]

On Jan 19, 5:53 pm, "john jardine" <john.jard...@idnet.co.uk> wrote:

"Kasterborus" <kasterbo...@yahoo.com> wrote in message

news:f9f2378a-2d1a-4703-9adc-a465461662b5@c23g2000hsa.googlegroups.com...

I have a Atmel MPU input pin held high through a pull-up resistor, a
push button switch pulls it low.

If I want to extend the cable with the push button to about 6' should
I use a Zener 5.1v protection diode?

(This was recommended as a guard against static buildup - I just
wondered if this is the right way to do it)

Thanks

A Zener is worthwhile.
I've a Mega32 running at the moment, 2 pins on pushbuttons with pullups
enabled. The pullups measure out at 35K ohms.
These will cope with normal amounts of "static" discharges. This leaves the
lighning jolt to the pin when you walk across a nylon carpet and touch the
pin connected wire.  The Zener will help, as also a 100nF cap connected (at
the chip) from the pin to 0V.

Really good idea to guard against ESD. Also, you might want to add a
series resistor after the zener and the cap to limit the current
flowing into the pin in the event of overvoltage, like this (view in
fixed font or M$ Notepad):

| |
| |
| ___ |
| o----o-----o----|___|-----o
| | | 1K |
| /-/ --- |
| ^ --- | Atmel uC
|1N4733A | | 100nF |
| === === |
| GND GND |
| |
| |
| |
| |
| '----
|
|
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

A zener-based protection scheme has benefits over diodes to each
supply rail. The speed of the diode isn't significant -- in fact,
you're adding the 0.1uF cap in parallel with the diode. And if
there's a positive-going spike, the charge will flow to the +5V rail
with the two diode scheme, which may make it exceed maximum spec for
the uC as well as other components, depending on trace inductance, the
ESR of the caps from the +5V supply to GND, and other considerations.
Also, you may end up with a situation where your power supply voltage
regulator may have its output at a greater voltage than its input,
which can be a disaster for certain regulators.

Now, you have to attend to the layout here either way, so try to have
a good, low inductance, low resistance path for the ESD charge to
travel without creating a spike on the GND lines. But from a
beginner, hobbyist perspective, a zener/cap combination with a series
resistor is a great start to preventing your uC from getting smoked by
a 100pF cap namely the user) discharging a few KV of potential to GND
through your project.

Cheers
Chris