Maximum current allowable through ESD diode of opamp?

[treez]

Hello,
Just checking what is the maximum current that can safely be flowed through diode ESDD1 of the LT6220 opamp?
(page 13 of the LT6220 opamp datasheet shows ESDD1).
Will 500uA be OK?

We have it set up as in U48 of the attached schematic, and sometimes U48 will loose its supply, but the preceeding opamp (U47) will still have its supply on. U47 will never have more than 2V at its output.

LT6220 opamp datasheet
http://cds.linear.com/docs/en/datasheet/622012fc.pdf

 

[KlausST]

Hi,

You have the datasheet by hand, so why don´t you read it?

* Absolute Maximum ratings:
"Input Current (Note 2)........................................... ±10mA"

* Note 2:
"The inputs are protected by back-to-back diodes. If the differential
input voltage exceeds 1.4V, the input current should be limited to less than
10mA."

added:
* Overdrive Protection:
When the input voltage exceeds the power supplies, two
pair of crossing diodes, D1 to D4, will prevent the output
from reversing polarity. If the input voltage exceeds either
power supply by 700mV, diode D1/D2 or D3/D4 will
turn on to keep the output at the proper polarity. For the
phase reversal protection to perform properly, the input
current must be limited to less than 5mA.

Klaus

 

[treez]

Thanks, yes I read that, but I am not referring to a high differential input voltage, and as such I am not talking about the diodes D5,D6, D7 or D8.
Also, I am not referring to the diodes D1,D2,D3 or D4…because these diodes have resistors in their series path, and so they will carry less current than the diodes ESDD1, ESDD2, ESDD3, ESDD4.

I am specifically referring to the diodes ESDD1-4, because these are the ones that will get hit the hardest when the inputs go above the supply voltage (they will carry the most current)….however, the datasheet doesn’t tell anything about the current carrying capability of these diodes.
Do you have any idea what it might be?

 

[dick_freebird]

Any 1kV ESD protection ought to tolerate (for a limited
duration and number) 600mA pulses. But thermally this
would not stand, for quasi-continuous (> uS range)
stress.

I would expect that a current limited op amp output
driving back into the rail would be OK if it's in the
<100mA range and was not applied indefinitely such
as to cause insane local temperature rise or metal
wearout.

Of course a bit of series resistance as a fault limiter
wouldn't hurt (unless it hurts).

 

[D.A.(Tony)Stewart]

the amplifier is protected against ESD strikes up to 3kV on all pins by a pair of protection diodes on each pin that are connected to the power supplies as shown in Figure 1.

This would imply low ESR and higher current rating.
Power dissipation is negligble.

So you are OK.

 

[FvM]

I am specifically referring to the diodes ESDD1-4, because these are the ones that will get hit the hardest when the inputs go above the supply voltage (they will carry the most current)….however, the datasheet doesn’t tell anything about the current carrying capability of these diodes.

The specification quoted in post #2 is exactly referring to these diodes...

10 mA maximum rating respectively 5 mA to avoid polarity reversal is a DC current specification. Diode current during ESD events are much higher, e.g. several 10 A but only 10 or 100 ns duration.

 

[D.A.(Tony)Stewart]

I understand it differently than FVM
D1-D4 protect output reverse polarity if input exceeds supply rails.
D5-D8 protect input excess differential voltage.
ESDD1-4 protect input to rails for ESD and will protect charges up to 3kV ( using standard ESD model for human finger capacitance and resistance)

These would be bigger diodes than D1-D8 and current is not specified.


In any case, if your output @2V will attempt to pull up the lower supply thru the diodes, so a series R is advisable.

 

[KlausST]

Hi,

I re-read the datasheet.
I'd say the datasheet is not 100% clear about your question.

Will 500uA be OK?

I'd say: yes. But no guarantee.
It's far below the specified values.
What can happen?
Heating will not be the problem. And any other destruction caused by 500uA? I can't imagine.

***
A question..
Why overvoltage? How do you limit it to 500uA?

Klaus

 

[treez]

as per schem of top post, when the error amp looses its vaux supply, the preceeding opamp output may drive into it...its 500uA because we have a limiting resistor there by dint of the error amp input resistor.