BiCMOS Bipolar+MOSFET better than MOSFET

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Advantages of BiCMOS over CMOS

I have had the quite firm idea for a long time
that hybrid designs between FET & BJT
is the way to the best performance in amplifier circuitry
.

We do not as often as I would like see such amplifers.
- The JFET members do just about only JFET amps
- The BJT members do just about only bipolar amps
- The Pass MOSFET members do just about only MOSFET amps
- The TUBE members do just about only Valve amps


Okay, I know several exceptions, where clever guys
did combine and use the best technique for each element in amplifier.
But we can do more here.

To me, lineup,
this is typical case of human difficulties with

Thinking OUT OF THE BOX

.... you enter a specific square box and your mind stays there
.... at least until your next project - and your next box
;)


Transient response:
CMOS:
cmos50_t.gif


BiCMOS:
bicmos_t.gif


Schematics for circuits
with the above Transient Responses
can be found at David Holburn's website.
.
David Holburn, University of Cambridge, Untited Kingdom,
is a Senior Lecturer in the Department of Engineering and leads the Scientific Imaging Group.
He has a separate web page describing his research, IEE and other interests.

.

Homepage of Mister Holburn:
http://www.eng.cam.ac.uk/~dmh/
 
I think you are right concerning the latchup time in digital circuits; this is enhanced greatly. However, in an analog amplifier, you would consider linearity a more great deal, since this does not really improve slewing rate for each single frequency.

I'm no expert whatsoever, but I think this would be no real improvement. Out of my thumb, I believe it could improve tracking of the output signal but would cause large distortions at higher frequencies.

Am I making any sense or just non-sense?

.edit: If you'd want to go hybrid, try Wavebourn's BJT and MOSFET in parallel:
http://www.diyaudio.com/forums/showthread.php?postid=1056830#post1056830
 
thanks for reply, Limhes

the message i wanted to tell is:

- one should not be fundamentalist in choice of transistors
... when is call for JFET, use JFET
... when bipolar does the job best ... use it!
... if MOSFET Power is best in your application .. go for this

try to design your amplifier for best possible function
and use the components that you figure best can do this

.. instead of saying like:

Now I need power amplifier again:
I have to order some Big Power Bipolars again

Now I need a pre amplifier:
Wonder how many JFETs I have at home for my input


Regards
lineup
....who did once called the mosfet maffia: MOSlim FETamentalists :D :D
 
I've had a look at Wavebourns idea of BJT and MOS in parallel in a simulator. Admittedly, most models of MOSFETs are not very good but they do get the basics right - the BJT || MOS approach generated an amazing spread of high order harmonics when set up as a follower. I willprobably revisiti it some time in the future, but for now it's still plain BJT or MOS for me :)
 
I think you are right concerning the latchup time in digital circuits; this is enhanced greatly. However, in an analog amplifier, you would consider linearity a more great deal, since this does not really improve slewing rate for each single frequency.

I'm no expert whatsoever, but I think this would be no real improvement. Out of my thumb, I believe it could improve tracking of the output signal but would cause large distortions at higher frequencies.

Am I making any sense or just non-sense?

.edit: If you'd want to go hybrid, try Wavebourn's BJT and MOSFET in parallel:
http://www.diyaudio.com/forums/showthread.php?postid=1056830#post1056830

This site don't goes open. Please let me know the currently URL - Thank you very much
 
I once made a preamp where I used CMOS logic unbuffered inverters as linelevel amp (..was it 4007 and 4069?) Internally they actually consist of one N and one P MOSFET, and I simply applied feedback as an inverting opamp. I am still using the thing in my setup and it sounds wonderfull.

Lineup: please check if there are unbuffered BiCMOS inverters available in packages like these - then you would be able to make a cheap, easy and maybe good sound lineamplifier.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.