The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Ft and Fmax
https://designers-guide.org/forum/YaBB.pl?num=1312123343

Message started by raja.cedt on Jul 31^st, 2011, 7:42am

Title: Ft and Fmax
Post by raja.cedt on Jul 31^st, 2011, 7:42am

hello every one,
can any one please explain why all foundaries will give ft only, they don't mention fmax? because fmax is more informative than ft.

Thanks.

Title: Re: Ft and Fmax
Post by aaron_do on Jul 31^st, 2011, 7:11pm

Hi,

I don't know the exact answer, but unless you use inductors, fmax is probably not that useful.

regards,
Aaron

Title: Re: Ft and Fmax
Post by buddypoor on Jul 31^st, 2011, 11:55pm

raja.cedt wrote on Jul 31^st, 2011, 7:42am:

hello every one,
can any one please explain why all foundaries will give ft only, they don't mention fmax? because fmax is more informative than ft.
Thanks.

I can imagine (guess) what you mean - however, don't you think it would be appropriate to give us some more details?
What is your definition of fmax? What do you mean with ft? (Probably transit frequency, but nobody knows).
You even did not mention if you speak about an amplifier or a passive part.

Title: Re: Ft and Fmax
Post by raja.cedt on Aug 1^st, 2011, 12:20am

ha budy poor,
i am sorry for giving such a lass informative question, so ft is unity gain short cicrcuit current fmax is unity gain power gain.

My question is why no one specifies (at least up to my knowledge) fmax in their technology. By the way i am talking about amplifiers.

Thanks.

Title: Re: Ft and Fmax
Post by buddypoor on Aug 1^st, 2011, 1:40am

raja.cedt wrote on Aug 1^st, 2011, 12:20am:

Where are your "definitions" from? Which textbook?
I must confess that I never have heard about a "unity gain short circuit current" that defines a frequency like ft.
And what is "unity gain power gain"?

What I know is the following:
* for transistors, the transit frequency is defined for current gain beta=1
* for opamps, open-loop gain Aol=1.

* for opamps, fmax=large signal bandwidth (caused by the slew rate).
_____________
Does this help?

Title: Re: Ft and Fmax
Post by sheldon on Aug 1^st, 2011, 4:14am

Buddy,

I think that the expression is the unity gain short circuit current gain
frequency, see

http://whites.sdsmt.edu/classes/ee320/notes/320Lecture22.pdf

As I remember the short circuit part is the reason that you don't have
to worry about the Rb*Cu term. It is also the reason you can't actually
measure ft, that is, there are no short circuits in the real world.

Best Regards,

Sheldon

Title: Re: Ft and Fmax
Post by buddypoor on Aug 1^st, 2011, 5:12am

Hi Sheldon,

thank you for your message.
However, I think it is sufficient and correct to speak about the "unity-gain frequency" of a BJT since the signal current gain always is specified for a constant bias point (i.e. Vce=const and Ic=const, which is equivalent to a signal voltage short circuit).
Thus, at the same time the EARLY effect is without any influence.

Therefore, I think it is not necessary - and, more than that it is confusing - to state that "ft is unity-gain short circuit current".
(Is ft a current?). That was the background of my answer.
However, I could imagine that the question was related to the transistor transit frequency. On the other hand, I must confess that I try to teach people to ask clear and logical questions.
Example: Some time ago I read the question: What is the meaning of B?
And what about "unity gain power gain"? Is it silly to ask for clarification?
Regards
Buddy

Title: Re: Ft and Fmax
Post by aaron_do on Aug 1^st, 2011, 7:51am

On the one hand, the question could have been more clear, but on the other hand, I think most analog designers would have understood what raja.cedt was asking. Anybody else care to answer the original question?

Title: Re: Ft and Fmax
Post by buddypoor on Aug 1^st, 2011, 8:02am

Hi Aaron,

by the way, what is your answer to the original question:

raja.cedt wrote on Jul 31^st, 2011, 7:42am:

can any one please explain why all foundaries will give ft only, they don't mention fmax? because fmax is more informative than ft.

Regards

Title: Re: Ft and Fmax
Post by aaron_do on Aug 1^st, 2011, 6:10pm

Hi buddypoor,

this was my answer,

Quote:

I don't know the exact answer, but unless you use inductors, fmax is probably not that useful.

I guess my answer wasn't very clear either. What I was saying is that fmax, which is the frequency at which the maximum power gain reaches zero, can only be achieved using resonant networks (from my understanding). So for most analog designers who don't deal with inductors, it isn't very useful. On the other hand, fT allows you to quickly estimate the speed of your analog circuit.

If the technologies were originally developed for analog design, then I think that would explain it enough. But since they are developed for digital IC design, I would be interested to know what fT corresponds to in a digital design.

regards,
Aaron

Title: Re: Ft and Fmax
Post by buddypoor on Aug 1^st, 2011, 11:50pm

aaron_do wrote on Aug 1^st, 2011, 7:51am:

........... but on the other hand, I think most analog designers would have understood what raja.cedt was asking....

Hi Aaron,
the extent of this discussion regarding the meaning of two simple parameters (ft and fmax) - for my opinion - confirms that I was not completely wrong by asking for questions that are more clear.
As you will agree, most of us are rather busy but, nevertheless, take the time and try to answer some questions that are placed here in the forum - in many cases basic problems from beginners.
And because of this I think it is not appropriate that the reader who is willing to answer is forced to assess and to guess what the questioner may have meant.
In the present case, two symbols have been mentioned (ft and fmax) without any additional information resp. explanation. No mention if it is a matter of amplifiers, parts, ADC's or what else.
Therefore, my unpatient response (hopefully not unpolite).
(By the way: Did you really know from the beginning that fmax was related to power transfer only? For example, for my understanding fmax could also be the large signal bandwidth of an opamp caused by the slew rate).
There are so many misunderstandings round the world - should we as engineers not try to express ourself as clear as possible? And should we not try to teach this "art" to other forum members ?
Kind regards
Buddy

Title: Re: Ft and Fmax
Post by aaron_do on Aug 2^nd, 2011, 1:00am

Hi Buddypoor,

as I previously said, the original question could have been more clear. Especially the part where raja.cedt defined the two terms. I have noticed a lot of people (including myself sometimes) ask questions not realising how difficult they are to understand without knowing the context - or just type faster than they can think. "unity gain power gain" is probably an extreme example :D

Quote:

(By the way: Did you really know from the beginning that fmax was related to power transfer only?

Personally, when I read the question, I automatically read fmax as the intended meaning. That is probably because I deal more with RF amplifiers than op-amps. Also, the question specifically related to foundry data, so from that point of view, a transistor's maximum oscillation frequency is more meaningful, since it doesn't relate to specific circuit designs.

I understand it can be frustrating reading some questions, especially after a long day's work, and I hope I haven't offended you since generally your comments are very helpful.

regards,
Aaron

Title: Re: Ft and Fmax
Post by buddypoor on Aug 2^nd, 2011, 7:26am

aaron_do wrote on Aug 2^nd, 2011, 1:00am:

Hi
...I understand it can be frustrating reading some questions, especially after a long day's work, and I hope I haven't offended you.....

No, in no way. - don`t worry.
Greetings
Buddy