# What's the difference between NPN and PNP transistors?

Suppose that I know how an NPN transistor works.

How different is a PNP transistor? What are the operational differences between a PNP and a NPN?

• @Federico - What leads you to believe that Denilson wants to know the physical differences? Since he accepted the answer as-is and linked to the other question about operational characteristics, I come to the same conclusion as Kortuk: You've changed the meaning of the question. Edits shouldn't be used to hijack threads, instead, edit to clarify the meaning of the post without changing it. – Kevin Vermeer Aug 10 '11 at 17:42
• @Kevin Vermeer: It's perfectly in line with the title, which asks what the difference is. Denilson also asks in the question what the operational differences are, and the accepted answer only talks about how to connect. If there are other differences I think they should be answers to this very question. – Federico Russo Aug 10 '11 at 17:51
• @Kevin Vermeer: I also wanted to avoid a new question to be closed as exact duplicate, because that's what will happen if I do ask. – Federico Russo Aug 10 '11 at 17:52
• @Kevin - I read Federico's addition and I agree with him that it doesn't change the question's intent. "Differences in characteristics" (FR) are part of "operational differences" (DS). I think it should be Denilson to decide on a rollback. – stevenvh Aug 10 '11 at 18:00
• @stevenvh, it definitely does not fall in line with what the poster probably meant based on what answer was accepted. You are correct, we need to OP to weigh in. Every comment put here is flagging him though. – Kortuk Aug 10 '11 at 19:09

PNP transistors work the same way as NPNs do but all voltages and currents are reversed. You connect the emitter to the higher potential, source current from the base and the main current flows into the emitter and then exits through the collector.

$V_\rm{BE}$ will be $-0.7\,\rm{V}$ but it's magnitude should be the same in both PNP and NPN if you use complementary parts.

• What you appear to be describing in the first paragraph is a PNP transistor, which you didn't say. This also doesn't answer the question since that was more about the device physics. You never explained majority carriers, holes, etc. – Olin Lathrop Aug 10 '11 at 11:43
• @OlinLathrop, you could edit to improve the question but based on the accepted answer the OP is primarily interested in operational differences. – Kortuk Aug 10 '11 at 13:11
• @OlinLathrop, I tried to improve the readability of my answer. As Kortuk said I don't believe the OP was interested in the physics at all. – jpc Aug 12 '11 at 10:48
• I see the question has changed in the mean time, or maybe it was merged. The original question I saw did ask about the physics and specifically mentioned majority carriers and holes. – Olin Lathrop Aug 12 '11 at 12:48
• @Olin I have checked the edit history and it seems that the expanded question you saw was because an addition by someone unrelated to the OP. – jpc Aug 12 '11 at 12:59

NPN and PNP transistors are different. Electrons are more mobile than Holes Which means that PNP is not as good as NPN. For Si BJTs the PNP types are behind when it comes to breakdown voltage and really high power. For general purpose devices like BC337 / BC327 things for all intent and purpose are the same but if you wanted to do a off line SMPS it wouldnt be easy or practical at 1KW. For germanium the NPN is supposed to be better but it is not. This is due to manufacturing issues. The AC127 is not nearly as good as the AC128 and the AD161 is not as good as the AD162 and yes these devices were sold as matched pairs. The ratio of electron to hole mobility is a determining factor in how close the PNP will be to the NPN. This is much worse for SiC so one would expect lousey PNP BJTs so they probably wont bother making them. For some reason PNPs have lower noise so they are favoured in diff pair input stages. The abundance of highside driver chips is proof that PNP is not as good as NPN.

• +1 for highlighting the differences in mobility between electrons and holes. A hole is not a "positive-equivalent" of a free electron. For folks puzzled by this comment, please see more here electronics.stackexchange.com/questions/199347/… – akhmed Aug 19 '16 at 0:01

The only difference lies within the functionality of the transistors. In grounded (common) emitter configuration, when a base current is provided (or to be more practical-when base is connected to 5v supply) of a PNP transistor, no conduction takes place as the majority carriers in n region are electrons whose motion is suppressed and no path is formed b/w emitter and collector.Thus no o/p is obtained at the emitter junction. If base current is removed from the transistor a virtual path is formed b/w the emitter and collector which offers certain resistance to electron flow which is subsequently altered by the base current (or voltage). If in such case, the Vcc is directly connected to the collector and emitter is grounded through a resistance(possibly 10k), then Vcc gets a direct path to appear at the emitter junction. Thus if o/p is taken at emitter in case of PNP, the config is that of an inverter while at collector the transistor works as a simple switch or buffer.(This is exactly the opposite of NPN config.) Due to dearth of certain simulation software, i am unable to present a pictorial view. But i hope this would serve the purpose.