Diff pair mismatch and tail current

[Tahar]

Dear All,

Is differential pair mismatch reduced by a higher tail current or a smaller one ?

Did not figure out yet....maybe something to do with Vov.

Thanks.

 

[watersky]

pair mismatch is usually effected by area. you should find some things from report of foundry.

 

[gpwu]

The differential pair mismatch is reduced by lower bias current if W/L is fix.

 

[Tahar]

Thanks, any idea about hot to derive the equation showing that ???

 

[jerryzhao]

ΔVth=σ/√W*L you can get σ in mismatch report from fab. σ depend on process, about 4~7mV.um

 

[Tahar]

I know this equation (hopefully), this is not what I am looking for.

I am looking for an equation of the form: mismatch=f( Ibias).


Thanks anyway.

 

[elbadry]

A higher Vgs (i.e. a higher current) would give better mismatch for a tail current.

Intuitively, this can be explained by the fact that for a given delta Vgs mismatch, higher Vgs would mean smaller impact ( since selta vgs is small compared to vgs, the corresponding delta I would be likewise small).

A derivation of the equations can be found in Grey & Meyer's book "Analysis and Design of Analog Integrated Circuits, 4th Edition"

 

[gpwu]

The equation of mismatch caused by I is:

Vos= sqrt(I/(2K'W/L)×(Δ(w/L)/(w/L))1-2 - Δ(w/L)/(w/L))3-4)

Hope this can help!

 

[markjiang]

there is a key equation at P.379 of Razavi. about the offset.

the summary is the more offset when the bias current is larger.

 

[layes2]

squar root WL