The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> ft (transition freq) and GBW
https://designers-guide.org/forum/YaBB.pl?num=1312286378

Message started by luky on Aug 2^nd, 2011, 4:59am

Title: ft (transition freq) and GBW
Post by luky on Aug 2^nd, 2011, 4:59am

Hi, all
under 0.18um cmos process, ft is about 30GHz. But a op with Aopen=6，f(-3dB)=3.2GHz can be designed under this tec? why?
Is there any relationship between ft and GBW(Aopen*f(-3dB))?

THANKS

Title: Re: ft (transition freq) and GBW
Post by buddypoor on Aug 19^th, 2011, 2:12am

I am not sure if I well understand your question.
Nevertheless, I'll try to answer:

Of course, the several components have a higher transit frequency than the resulting opamp. The reason is that - due to stability properties if feedback is applied - the opamp gain function must be shaped like a first order lowpass. This is the main reason for a rather low open loop gain cut-off (only 10...20 Hz for all-purpose opamps).

Title: Re: ft (transition freq) and GBW
Post by harpoon on Aug 19^th, 2011, 7:19am

hi lucky,

i think you will struggle to design a standard amp that has a GBW of 19.2GHz ... (6*3.2GHz) ... especially if you want it working over PVT.

however, never say never.

To achieve 3.2GHz, you will need some "tricks" ... e.g. neutralization, inductive peaking etc ... I think you will find such tricks widely discussed in literature.

Note that there are trade-offs ... and whether they are "robust" enough for production.

there are some rule of thumbs ... and i remember willy sansen mentioning something like GBW = ft/16 ... I would say something like ft/10 is a good target.

in any case ... don't read too much into the reported ft numbers ... it should be treated purely as a GUIDE only !

good luck !

Title: Re: ft (transition freq) and GBW
Post by raja.cedt on Aug 19^th, 2011, 8:09am

hello lucky,
ya as harpen said it is very difficult to reach ft for opamps, you can design with nuralization and some kind of idicutive peaking techniques. Moreoever ft is a some king of extrapolated paaramer and no resister have taken into consideration so just consider Ft is a benchmark. Please refere this paper to get some more information about ft. 10-40GHz design in SiGe-BiCMOS and Si-CMOS –
linking technology and circuits to maximize
performance
Recently i have come across a pap titled (11 GHz UGBW Op-amp with feed-forward compensation technique) where they have reached 11G ugb but with some body biasing technique.

Thanks.