Designer’s Guide Community

dandelion Community Member Offline Posts: 98	The bjt and cmos process Jul 22^nd, 2008, 7:00pm Hi, I am desiging an active filter. I have the cmos and bjt process availble and they all have enough Ft to meet the bandwicth requirements. I wonder which process you advised me to use? E.g, from the point of lineraity characteristics, which one is better? Thanks
Back to top	IP Logged

HdrChopper

Community Fellow

Offline

Posts: 493

Re: The bjt and cmos process
Reply #1 - Jul 23^rd, 2008, 4:51pm

Hi Dandelion,

The selection will depend on which parameters are most important to optimize in your design.
For example, if offset and noise are important to be kept low, then bjt would be the choice since offset levels are much lower than their MOS counter part. At the same time 1/f is significantly lower for the BJT.

Another thing to consider is the available area. Depending on the passive components sizes and the relative weight of your amplifier on the total area the CMOS option might be the better one (I guess you could implement an smaller amplifier in CMOS than in a bjt process for similar specifications)

Hope this helps.
Tosei

Back to top

Keep it simple

IP Logged

vivkr

Community Fellow

Offline

Posts: 780

Re: The bjt and cmos process
Reply #2 - Jul 23^rd, 2008, 11:36pm

dandelion wrote on Jul 22^nd, 2008, 7:00pm:

Hi,
I am desiging an active filter. I have the cmos and bjt process availble and they all have enough Ft to meet the bandwicth requirements.

I wonder which process you advised me to use? E.g, from the point of lineraity characteristics, which one is better?

Thanks

Linearity? Well, that would depend a lot on the design techniques you use, and the supply voltage you have available to you. I would say that it is possible to make highly linear filters using either MOS or BJT. The high input impedance of the MOS gate makes the design a lot easier though. Otherwise, there is not much to choose between the two if the 1/f noise and offset issues are not critical (you only mention linearity).

I would say the following: Go with the one you know much better. The chances are that you will make a better MOS design than a BJT one if you are used to MOS.

Regards
Vivek

Back to top

IP Logged

buddypoor

Community Fellow

Offline

Posts: 529
Bremen, Germany

Re: The bjt and cmos process
Reply #3 - Jul 24^th, 2008, 1:38am

In addition to the comments/recommendations made by tosei and vivek I think the choice also should be based on the filter topology chosen (filter order, cascade design, direct design, voltage or current mode, gm-C, MOSFET-C, ...) .
Most probably the technology determines filter topology and vice versa. It�s really not easy to find the best combination for a certain application (which always will be a compromize between conflicting parameters).
Good luck and regards to you.

Back to top

LvW (buddypoor: In memory of the great late Buddy Rich)

IP Logged

Paul

Community Fellow

Offline

Posts: 351
Switzerland

Re: The bjt and cmos process
Reply #4 - Aug 22^nd, 2008, 1:03pm

Beyond technical aspects, the application/motivation for the design should be considered. If it were a commercial design, the system-level integration would determine the technology choice. Assuming this is a research project (MS/Ph.D.?), I would go for CMOS, because later on there will be far more companies interested in your job application if you have done CMOS design rather than BJT design.

My two cents
Paul

Back to top

IP Logged