antenna violation effect- when to use jumper and when diode?

[siddharth1501]

antenna effect

hi all

To remove antenna violation generally we use jumpers or diodes.My question when we will use jumper and when we will use diode ?

 

[erikl]

Re: antenna effect

Jumpers whenever possible and sufficient. But when your metal chain antenna area still gets too large before it hits silicon contact, you need a protection diode.

 

[siddharth1501]

antenna effect

hi eirkl , u mean if metal area is more .we should go for protection diodes.

Added after 3 minutes:

while using we go for higher metal i.e, if had antenna violation at M1 we go for M2.Then wat if i have antenna violation at top metal.

Added after 1 minutes:

while using jumper we go for higher metal i.e, if had antenna violation at M1 we go for M2.Then wat if i have antenna violation at top metal.

 

[erikl]

Re: antenna effect

If you still get an antenna error with any combination of metal(s) (M1, M1+M2, M1+M2+M3, ...), this means that this connection doesn't (yet) have contact anywhere to the silicon (drain contact, diode contact, n-well resistor). If the area of this connection is large compared to the connected gate area (antenna rules will check this), accumulated charge (during IM or metal ion etching) could destroy the gate. In such case you must insert a protection diode (recommended: p+ diode in n-well, cathode = n-well to VDD). Minimum size is sufficient.

If you still get an antenna error when all metals are present, this means there's still no connection to silicon - a very rare case - or a hint for a missing/failing connection.

 

[siddharth1501]

antenna effect

hi erikl
my doubt if i had antenna violation at top i should use only protection diode. can i go for jumper with top -1 mteal

 

[erikl]

Re: antenna effect

my doubt if i had antenna violation at top i should use only protection diode.

Yes.

can i go for jumper with top -1 mteal

You can, but most likely this won't help.

 

[deepak242003]

Re: antenna effect

hi erikl
my doubt if i had antenna violation at top i should use only protection diode. can i go for jumper with top -1 mteal

I think it wont help.. top -1 is already fabricated.. hence effective field area connected to gate wont be effected..

 

[Ryan1234]

Re: antenna effect

hi erikl
my doubt if i had antenna violation at top i should use only protection diode. can i go for jumper with top -1 mteal

I think it wont help.. top -1 is already fabricated.. hence effective field area connected to gate wont be effected..

Hi, dear guys :
If i had antenna violation at metal 1, then i segment metal 1 into several parts.
And using the jmpers to connect them. My question is :
When the metal 2 is fabricated, does the accumulated charges in different parts of metal 1 will convene through the jumpers ?
In that way, the total charges on the gate oxide won't change, so the gate oxide will be harmed. I am so confused.
Thank you!!

 

[erikl]

Re: antenna effect

If i had antenna violation at metal 1, then i segment metal 1 into several parts.
And using the jumpers to connect them. My question is :
When the metal 2 is fabricated, does the accumulated charges in different parts of metal 1 will convene through the jumpers ?

The accumulated charge from metal 1 etch is flushed during a subsequent wet cleaning (rinsing) step (the charges are short-circuited). But during the end phase of the metal 2 etch (i.e. when the metal 2 structures are just separated), the exposed metal 2 areas (which now are connected via vias to metal 1 and via contacts to poly gate) now catch new charge and distribute it upon the total metal 2 - metal 1 - poly gate connection.

In that way, the total charges on the gate oxide won't change, so the gate oxide will be harmed. I am so confused.

Right. But not from the "old" charges from metal 1 etch, but from the "new" caught charges from the metal 2 etch. The total connection area to poly gate area ratio will decide upon how much the gate oxide is jeopardized, and the ARC will flag an error when indicated.

 

[Ryan1234]

Re: antenna effect

If i had antenna violation at metal 1, then i segment metal 1 into several parts.
And using the jumpers to connect them. My question is :
When the metal 2 is fabricated, does the accumulated charges in different parts of metal 1 will convene through the jumpers ?

The accumulated charge from metal 1 etch is flushed during a subsequent wet cleaning (rinsing) step (the charges are short-circuited). But during the end phase of the metal 2 etch (i.e. when the metal 2 structures are just separated), the exposed metal 2 areas (which now are connected via vias to metal 1 and via contacts to poly gate) now catch new charge and distribute it upon the total metal 2 - metal 1 - poly gate connection.

In that way, the total charges on the gate oxide won't change, so the gate oxide will be harmed. I am so confused.

Right. But not from the "old" charges from metal 1 etch, but from the "new" caught charges from the metal 2 etch. The total connection area to poly gate area ratio will decide upon how much the gate oxide is jeopardized, and the ARC will flag an error when indicated.

Thank you very much.
Very clear answer.

 

[gksivas]

Re: antenna violation effect- when to use jumper and when di

I have one doubt regarding ANTENNA EFFECT.

Antenna violation will come to metals. For example if violation came to metal3. then what we will do, we do metal hopping with metal4 and then to metal3, otherwise we can connect diode. But how can we debug the DRC error by connecting diode in this case means (metal area/ gate oxide area).

compulsory we have to fix the DRC error metal3 violation of antenna.

Please clear this doubt

 

[siddharth1501]

hi gkvikas

you will get lvs error. u have to update ur spice for extra added diode.and drc u have to clean u will get mostly spacing violation.