Ethernet line sugre protection

Thread Starter

bodhisatva

Joined May 20, 2004
23
>>>Click here for the schematic<<<

Hello people,
this is a circuit for metallic protection of Ethernet line pair.
1.5KE6.8A is a special clamping zener with extremely fast response <1ns and high surge capability, called also Transient Voltage Suppressor. Because of its relatively high capacitance it is not applicable for high speed data transfer lines. There is a trick to reduce it's capacitance by connecting it in series with two low capacitance diodes. See the schematic here>>Click for Schematic

My questions are:

The diodes HER107 are ultra fast - 70ns but not fast enough in comparison with the 1.5KE6.8A. Will this affect the performance of the circuit when an Electrostatic discharge occurs. The rise of the voltage is very fast - some picoseconds.

What happened when you connect in series two diodes? What is the resultant response time of the circuit?

Is this critical for the protection of sensitive elements? Should I use a faster schottky diode - 5ns?

Thank's for the time.

bodhisatva
 

mozikluv

Joined Jan 22, 2004
1,435
hi bod,

i would like to make a reply on your query but i could not see the posted schem. i don't know what's wrong.

anyway, you said that the zener clamp is used in that circuit and has a response time of 1ns. now you are contemplating of substituting with a lesser speed.

when that kind zener was used i believe the designer had in mind the kind of transient speed will be encountered thus making him decide to use that particular device. in other words the designer had put a certain elbow room. the question is how much leeway can it handle.

since you are the one to use it, it's only you who knows how much transient you will be encountering. if you think that a response time of 5ns is more than enough to handle the expected transient then you can safely do it.

for protection purposes, you can only increase response time not lower it.
 

Dave

Joined Nov 17, 2003
6,969
Originally posted by mozikluv@May 21 2004, 07:40 AM
hi bod,

i would like to make a reply on your query but i could not see the posted schem. i don't know what's wrong.
I'll second that :unsure:
 

Thread Starter

bodhisatva

Joined May 20, 2004
23
--------------------------
I put a link to schematic in the original post. I hope it will work.

You are wright, mozikluv that it's a mater of designer decision. I don't have experience or research on Electro Static Discharge(ESD). But according to the information I found in the Internet the peak of the current happens in the first few nanoseconds. According to some of the authors the first picoseconds(<1ns). The 1.5KE6.8A is designed for this purpose but because of it's capacitance (up to 1000pF) is not suitable for high speed data lines. The addition of the low capacitance diodes in series reduces the attenuation.

OK. But what is the behavior of a diode subjected to 1ns pulse of high current flow in forward direction?

There are such a diodes like BAT85 cheap and available. Schottky diode with 10 pF junction capacitance and 5ns reverse recovery time. From the datasheet :
"...Planar Schottky barrier diode with an integrated protection ring against static discharge..."
I'm curious what kind of protection ring is this...

thank's
 

Attachments

Top