The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> MiM cap vs transistor decoupled cap
https://designers-guide.org/forum/YaBB.pl?num=1285240212

Message started by newic on Sep 23^rd, 2010, 4:10am

Title: MiM cap vs transistor decoupled cap
Post by newic on Sep 23^rd, 2010, 4:10am

For a given area like 10u x 10u, I found out that the transistor decoupled cap (with s & d tied together) has larger capacitance value than MiM cap for the high gate voltage around vcc. Is it better to use transistor cap to form decoupled cap rather than MiM cap.

Title: Re: MiM cap vs transistor decoupled cap
Post by raja.cedt on Sep 23^rd, 2010, 5:43am

hi,
MIM cap only preferred only in case of high linearity requirement. For decap i guess you don't have any such requirement.

Thanks.

Title: Re: MiM cap vs transistor decoupled cap
Post by newic on Sep 23^rd, 2010, 6:19pm

Is it normal that the transistor cap capacitance value is much larger than MiM cap value. MiM is formed by insulator, which is supposedly have large dielectric, thereby larger cap value. That's why I always presume MiM will have larger capacitance value.

Title: Re: MiM cap vs transistor decoupled cap
Post by swolf on Sep 24^th, 2010, 7:57am

yes, the gate oxide is usually thinner than the insulator in the mim.

Title: Re: MiM cap vs transistor decoupled cap
Post by raja.cedt on Sep 24^th, 2010, 10:35am

no man, MIM has very less value because distance between two metals is depends on the inter metal thick-ness where as mos-cap has very thin oxide hence very high cap.

Thanks.

Title: Re: MiM cap vs transistor decoupled cap
Post by carlgrace on Sep 24^th, 2010, 4:00pm

If your process allows it, you can get maximum bypass cap density by laying out a MIMcap on top of the MOScap to make a bypass cell. Be careful about the Q of the bypass cap, you may need to add a small resistor in series with the bypass cap. To know, you will have to simulate with the expected amount of bypass cap and the expected inductance in the power rail.

Title: Re: MiM cap vs transistor decoupled cap
Post by love_analog on Oct 11^th, 2010, 6:14am

NewIC
What process technology is this in. I have heard that in finer geometries (32nm etc), the MIM cap have higher resistance. In which case Q may not be an issue.
Otherwise whatever carlgrace said is bang on.

Title: Re: MiM cap vs transistor decoupled cap
Post by rfmagic on Oct 11^th, 2010, 8:17am

usually for low noise design MIM is better. MOS CAP has higher noise than MIM

Title: Re: MiM cap vs transistor decoupled cap
Post by RobG on Oct 11^th, 2010, 12:52pm

rfmagic wrote on Oct 11^th, 2010, 8:17am:

usually for low noise design MIM is better. MOS CAP has higher noise than MIM

And of course the capacitance of MOS caps is strongly dependent on bias voltage so that is the main reason why it isn't used more. It makes a good decoupling cap, however.

rg

Title: Re: MiM cap vs transistor decoupled cap
Post by ACWWong on Oct 11^th, 2010, 1:57pm

yes moscap makes good decoupling cap in terms of density, but as well as voltage dependancy, they leak current which is heavily temperature dependant.... this may or may not be a problem in your application.

Title: Re: MiM cap vs transistor decoupled cap
Post by newic on Oct 12^th, 2010, 5:08am

thank you :)

Title: Re: MiM cap vs transistor decoupled cap
Post by ci on Oct 12^th, 2010, 11:20am

One additional thought:
MOS caps have one of the plate capacitively coupled to the substrate, so what you have there are actually two capacitors in series. You decouple two voltage supply rails but you couple one of them with the substrate through this additional capacitor. So depending on the design you may inject noise into substrate (if that supply rail is noisy) or you may capture noise from the substrate into a sensitive supply rail.

Cosmin Iorga, Ph.D.
founder NoiseCoupling.com
http://www.noisecoupling.com

Title: Re: MiM cap vs transistor decoupled cap
Post by newic on Oct 14^th, 2010, 6:36am

thank you :)