Audio inout protection

Are the 'ESD' protection diodes across the input line or from the input line to the power rails?

It's mostly a matter of how much power the protection diodes can dissipate. In general you add some series resistance in front of the diodes such that at maximum over voltage the current through the diodes is no more than they can tolerate.

George H.

Thanks for quick response.

1) ESD diodes are placed from protected line to power rail. They are in a 4-lead package (PRTR5V0U2) with 1 to GND, 4 to +5V VCC and the other 2 to protected lines.

2) There is no power rating for PRTR5V0U2 in the datasheet but they are specified for 8KV ESD protectiom and similar individual bi- directional diodes (V5.5MLA0603) are rated for 30A non repetitve surge current. In case of V5.5MLA0603 they are connected from protected line to chasis ground.

Thanks bhav

As this is an audio system, you can limit the energy under the curve seen by the diodes by using a coupling cap in front of the diodes to the input. If you direct couple, with a diode to the +rail, the rail will rise to the input voltage level other than for short pulses, unless limited elsewhere...

Regards,

Chris

A car battery will vaporize the ESD diodes, and likely other stuff as well. The usual fix is some series current limiter, like a polyfuse or a resistor, with a serious zener-type device, a transzorb maybe, on the inboard side of that. The combination has to stand continuous, maybe bipolar, +24 volts.

There are some ESD "diodes" that latch, essentially SCRs. They would be handy behind a polyfuse.

The surface-mount polyfuses are generally terrible. The radial lead ones are better.

John

Well now, use a bit of sense and put the highest value resistor the app allows from the source to those "protected" inputs (and outputs as well).

Assume a max of a few mA at most...use SERIES resistors as mentioned.

ut

e 4V e

ed.- Hide quoted text -

Thanks for all the responses.

I have DC blocking series capacitor at amplifier input to block DC from enetring the amplifier input. This should protect the amplifier input from short to 12V/24V volts DC as well. Correct me if my assumption is wrong.

What I understand from all the responses is that there should be a series resistor somewhere in the ESD diode path to limit the current flowing through it in case of ESD event or short to power rails. something like below. Is it fine?

CAP --------------- -----------------------------------||-----| amplifier | | --------------- ------------------ |ESD DIODE| ----------------- | | / \ / current limiting resistor \ / | ------- ---- Ground -

-bhav

nput

are

Bhav,

please pardon my curiosity....

the questions you are asking about the audio ESD protection...

is this a hobby project or are you working for a company as a paid proffesional engineer on this project?

thank you

Mark

nput

are

to

t

/24V

the

e e

oned.- Hide quoted text -

=A0 CAP =A0 ---------------

=A0 =A0 =A0 =A0 =A0 ---------------

Ahh, that will limit the current through the diode. But what's the voltage at the top of the diode? (the input line) The current limiting element is usually put in the signal line... Then after that the diode to limit the voltage. You might put everything behind the capacitor. That should make life easier.

George H.

input

s are

o V

I am doing my short term diploma course in automotive electronics.

-bhav

input

s are

o V

s to

but

2V/24V

th

V

f the

e

rge

ine

tioned.- Hide quoted text -

=A0 CAP =A0 ---------------

=A0 =A0 =A0 =A0 =A0 =A0 ---------------

Thanks George. The voltage at top of the diode is max of 2V line level audio signal. I do not want to put the current limiting resistor in the audio signal line. So looking for alternates. What is the benefit if i put everything behind the capacitor?

-bhav

ut

e 4V

Just a FYI, 8KV isn't much protection for a pin that goes to the outside world. It is a spec (3B I think), but real life is much worse.

There are two basic areas where ESD is of concern. One is for the manufacturer of the semiconductor. The semi just wants the chip to survive assembly and factory handling. [I've seen ESD issues just due to the part sliding down a rail.] Some chips (interface) are a bit more rugged than that, but not much more. The other is ESD for the final product. When HP was HP, nothing left the factory that couldn't take 25kv. However, the box manufacturer uses a much higher resistance in series with the voltage than does the semi. Basically, it is the difference between human and machine model.

Never having had to do ESD for the box (just the chip), my recollection is they have a gun to zap the unit. You can arrange the connection to the outside world so that the spark is more likely to hit ground rather than the input.

Lastly, in the days when people really cared about ESD for the box, there were protection devices that I though were quite the rip off. Basically a bit of resistance and a cap. If you look at the test circuits, it doesn't take much capacitance to shunt the ESD current.

When using clamping diodes, you need some means of limiting the fault current. On the output side, this should not be a problem, since the output series resistor will also isolate any capacitive load from the feedback loop. A 100-1000 ohm series resistor at the output should be sufficient.

On the microphone input side, large series resistors should be avoided, since they will add thermal noise to the signal.

This may sound a bit old fashioned, but have you considered using microphone transformers at the inputs ?

A small resistor (low thermal noise) will limit the primary current so that it does not burn the transformer primary wire.

The low primary DC resistance will dry out any static charge, so the ESD protection is handled this way. The limited transformer bandwidth will also limit the peak voltage transferred from primary to secondary during a quick ESD discharge.

Of course, using microphone transformers will also help in reducing

50/60 Hz mains hum (+harmonics) problems.

thank you Mark

ne input

uts are

he

to

24V

rts to

s but

e 12V/24V

with

-2V

of the

are

he

ey

surge

line

entioned.- Hide quoted text -

=A0 =A0 CAP =A0 ---------------

=A0 =A0 =A0 =A0 =A0 =A0 ---------------

There is a lot less DC current. Behind the cap you only have to worry about how much charge might be on the cap.

A bit of resistance in the signal line is not that bad. What's the input impedance of the amp you are protecting?

George H.