Lateral PNP

Thread Starter

blah2222

Joined May 3, 2010
582
Hi,

I have been looking all around for information on these devices but I am not too familiar with how they work.

I understand that they're main selling point is that they can be made with the same masks as vertical NPN transistors as their forms are very similar.

I have see some datasheets show a circuit symbol for them as a PNP that has its base tied to what seems like the collector of the other half of itself, acting as a current mirror of sorts if that made sense.

Does anyone have any information on how these funny things operate?

Thanks
 

crutschow

Joined Mar 14, 2008
34,285
Lateral PNP transistors basically work the same as normal transistors but because of the limitations of the process, they have poor characteristics, such as frequency response
 

WBahn

Joined Mar 31, 2012
29,979
My experience with these devices is in CMOS processes in which you can't make normal BJT devices. In an typical N-Well process, you would make a lateral PNP by putting two P-active regions within an N-well. But since the N-well is intended to be used for PFET bulk and since the two P-active regions have the same doping (typically emitter implants are much more heavily doped), the performance is pretty poor. But there are some circuits, such as bandgap references, some photodetector devices, and protection circuits, where the use of BJTs is valuable and you can tolerate poor performance (of course, "poor performance" depends on what performance parameter is being discussed -- the effeciency of photodectors made with lateral devices can be quite good, but it is still a pretty poor transistor by most measures).
 
Top