For ploy resistor and mim cap, which has better matching

[abbot_1860]

for ploy resistor and mim cap, which has better matching


thanks

fei

 

[chuckey]

We not understand your question. Never heard of "ploy" and "mim".
Frank

 

[abbot_1860]

oh, my ***

 

[ZekeR]

This is a question best to ask your process engineer or a representative from the foundry, because I think it's probably process-dependent. The matching of capacitors will depend on:
A) How consistent the oxide thickness is
B) How "ragged" the edges of the capacitor are
I think it's probably a toss-up between poly and mim caps. And for the designer's part, one should keep the topology of devices surrounding the caps consistent.

I don't think you should be especially concerned about cap matching though, because generally caps will be fairly large (and therefore fairly well-matched). You will most likely see much more mismatch in the current sources that feed the capacitors.

 

[abbot_1860]

This is a question best to ask your process engineer or a representative from the foundry, because I think it's probably process-dependent. The matching of capacitors will depend on:
A) How consistent the oxide thickness is
B) How "ragged" the edges of the capacitor are
I think it's probably a toss-up between poly and mim caps. And for the designer's part, one should keep the topology of devices surrounding the caps consistent.

I don't think you should be especially concerned about cap matching though, because generally caps will be fairly large (and therefore fairly well-matched). You will most likely see much more mismatch in the current sources that feed the capacitors.

i've been told that for most processes, mim cap has better matching than poly resistor. Is that true, or just because for most applications cap has much larger area?

thanks,

fei

 

[ZekeR]

i've been told that for most processes, mim cap has better matching than poly resistor. Is that true, or just because for most applications cap has much larger area?

Mim caps certainly have better tempo and voltage coefficient, and less parasitic cap, because they are father away from the substrate, so they are generally more precise for most timing applications ... But thats not the same as matching. I don't think I've ever heard that mim caps have better matching, so it could go either way.

In any case, the matching of the surrounding active components will probably dominate.