Ground Bounce explanation - Am I right?

Status
Not open for further replies.

VLSI_Learner

Member level 2
Joined
Oct 26, 2012
Messages
51
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
1,594
My doubt:
I am considering one circuit of a CMOS inverter with NMOS connected to an inductor before it connects to a ground.
When the PMOS is on, Vout= 1 (say 5 volt = '1' and 2 volt = '0')
and in when the NMOS is on PMOS is off. At that time Vout will start to drain out through NMOS, consequently current will flow through inductor.
say after time 't' the voltage across the inductor is V= L*di/dt
At one time this will occur
Vout = V and both are less than 5volt.
Beyond that time a time will come when V will be slightly greater than Vout. At that time NMOS will swap it's source and drain. That means so far what was drain, will now be source. And V will start to drain through NMOS and it will add to Vout.
This process will continue till NMOS is ON. And so Vout will never go to ground. This is ground bounce.

Am I right?
 

First I've heard of it. From reading a few articles, ground bounce is an effect of inductance present in many situations...

However a coil inductor does not necessarily have to be present for ground bounce to occur.

Nevertheless a coil as you describe might serve to aid in demonstrating how ground bounce arises.
 

But no where it was explained how actually the "bounce" occurs so I thought about it a bit and explained it that way. Am I right?
 

I can't claim to understand all about it. It does not seem as though ground bounce reverses current flow altogether.

The article linked below contains schematics which resemble your own description.

https://www.fairchildsemi.com/an/AN/AN-640.pdf

It explains that ground bounce was seen to emerge as a disruptive effect as devices were switched at increasingly faster speeds and increasingly greater power. Otherwise it is often masked by other effects.

Another article explains that ground bounce is reduced by installing the 0.1 uF capacitor which common advice says to install at the supply rails of a device. The capacitor maintains supply voltage when a device switches internally, at the moment that might cause ground bounce.
 

I can't claim to understand all about it. It does not seem as though ground bounce reverses current flow altogether.

If it does not reverse the current flow through the NMOS, then how the Vout "bounces" due to the inductor current spike (think about the inverter schematics, between PMOS and NMOS there is Vout and between NMOS and ground there is the inductor - the source of ground bounce)?
 

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…