People ALWAYS say CMOS is more LINEAR than BJT,what's PROOF?

Under what condiction? How to make a FAIR comparision?

1. Same die area? same current density? or same current? How is the test setup?

2. Use VIP3? I doubt the theory behide, it is DC characteristics. Use IIP3 from HB simulatin? or OIP3? BJT has higher gain than CMOS, thus its IIP3 is easily drop lower than that of CMOS. There is always a peak and a valley for these curves (VIP3 IIP3 OIP3), I don't think it is easy to compare.

I am very confused !

I think you just mistake what people is talking about

if the two devices have the same current , the gm of the bjt is much higher than the gm of the mos
cos the current relation is exponetial in bjt , but is second order in mos
so the gain of the bjt is much greater than the mos

khouly

what mistake?

Added after 41 seconds:

I am talking about linearity.

BJT is more linear than MOS devices

Is it? I have read some papers, they say MOS is more linear than BJT. Why do you think the reverse way? Do you have any reference material?

I have always been told the conclusion that BJT device is more linear that the MOS
cos for BJT Ic=BetaIb it is almost linear while for MOS device, it folows the quadratic law, so the linarity is degraded.

I dunt know where you get the conclusion that MOS is linear than BJT

yes I agree

its the current laws my friend.

for BJT ic to ib can be proven linear. (ic = beta*ib)

for MOS, it is polynomial law. (id = K*(Vgs-Vt)^2)

the_penetrator?

I attached 2 papers here. I think when we talk about linearity, we are refering to gm rather than current gain. for bjt, Ic=Is×exp(Vbe/Vt), for cmos, Ids=k×(Vgs-Vt)2.

exp(x) = 1 + x/2! x2/3! + .... x^n/(n+1)! . Thus bjt has more nonlinear terms.

you are comparing current gain with transconductance

i wonder, why don't you give us BJT's and MOSFET's gm? your multiple posts don't help anyone interested in this topic. neither do mine. this is the last time i refer to this thread. your approach is not polite.

of course the BJT is non-linear regarding collector current against the terminal voltages.

Here are the gms.
BJT: gm = IC/Vt (IC: bias current) (Vt: thermal voltage)
MOS: gm = 2k(VGS-Vt*) (Vt*: threshold voltage)

next time put down the equations... i know i will...

regards
the_penetrator?

First of all, I really appreciate your reply. But what do you mean multiple post? I just clicked the "quote" button and wrote a reply, did I do wrong?

Secondly, we are discussing technical questions, why do you think I am not polite? In your previous post, you just compare Gm with current gain, in this post, you give both gm equation. You've never given an conclusion: which is more linear and why. Maybe you are correct, but if you are really not interested in this topic, no one is expecting you here.

Finally, don't put so many here, that can not solve problem.