Voltage on the base of PNP transistor, when no voltage applied directly to it

Thread Starter

phippstech

Joined Jan 27, 2020
19
I have a strangle thing happening to my circuit and wanted to know thoughts on what might be happening. I have a circuit with three PNP transistors and they all are behaving the same. The voltage on the emitter is 5V and while I'm not applying any voltage directly to the base of the transistor, I'm reading 4.4 V on the base. J2 is a six pin header and I haven't connected anything to it yet.
1597886252543.png
 

MrChips

Joined Oct 2, 2009
30,712
The base-emitter junction looks like a PN junction diode. The forward voltage of a silicon PN junction is about 0.6V.
Hence 5V - 0.6V = 4.4V which is exactly what you are seeing on a "floating" base.

Your DVM very likely has internal resistance of 10MΩ. When you connect the DVM between base and ground you are putting the emitter-base junction into forward bias with about 0.5μA.
 

Thread Starter

phippstech

Joined Jan 27, 2020
19
The base-emitter junction looks like a PN junction diode. The forward voltage of a silicon PN junction is about 0.6V.
Hence 5V - 0.6V = 4.4V which is exactly what you are seeing on a "floating" base.
Thanks for the timely reply. I figured it had something to do with that voltage drop. So, Connecting it to the relay/MCU/or whatever is going to trigger the base will make the circuit work as planned?
 

MrChips

Joined Oct 2, 2009
30,712
No. It is common practice to use an NPN transistor in common emitter configuration. Doing it this way you only need about 1V to fully turn on the transistor.
 

dl324

Joined Mar 30, 2015
16,845
I figured it had something to do with that voltage drop.
The issue is that your meter was affecting circuit operation. You always need to consider how a DVM will affect measurements.

A comment about about your drawing style. Conventions for power (positive polarity) and ground is that power points up and ground points down. The transistor orientation you chose is rotated 90 degrees counter clockwise from "normal". We also try to line up components and make the labels right reading when it makes sense.

The transistors have the same "function", but you drew the collector resistors 3 different ways, i.e. inconsistently.
1597893186672.png
 

ScottWang

Joined Aug 23, 2012
7,397
Unless you have a special reason that you must be use the pnp transistor, otherwise you better use an npn transistor as MrChips mentioned in #4, or use an Uln2003, because the initial status of Mcu is Low, and it will active the pnp transistor, but that is not what you want.

If you need to use pnp transistor, the other way is to use an npn + pnp two stages of transistor, or an Uln2003 + pnp.
 

Thread Starter

phippstech

Joined Jan 27, 2020
19
Unless you have a special reason that you must be use the pnp transistor, otherwise you better use an npn transistor as MrChips mentioned in #4, or use an Uln2003, because the initial status of Mcu is Low, and it will active the pnp transistor, but that is not what you want.

If you need to use pnp transistor, the other way is to use an npn + pnp two stages of transistor, or an Uln2003 + pnp.
There are four other PNP transistors in this design, so I think the previous designer wanted to use all the same transistor from a cost standpoint. But I agree, if I was to design this circuit from the ground up, I would have used a NPN.
 

ScottWang

Joined Aug 23, 2012
7,397
There are four other PNP transistors in this design, so I think the previous designer wanted to use all the same transistor from a cost standpoint. But I agree, if I was to design this circuit from the ground up, I would have used a NPN.
For npn driver then you can try Uln2003(7 Drivers) or Uln2803(8 Drivers).
 

Ylli

Joined Nov 13, 2015
1,086
Whether an NPN or PNP transistor would be more appropriate here depends on two things. If the source driving the base(s) can sink current but can not source current, then using the PNP is the right solution. If the source can source current but not sink it, then an NPN would be more appropriate. If the source can both sink and source sufficient current then that point is moot. Second issue is the output. As configured with the PNP's, this circuit can source a significant amount of current, but sink somewhat less. Using NPN's would reverse the situation - the NPN output could sink a lot of current, but not source as much.

So whether using NPN's or PNP's is really application specific.
 

dl324

Joined Mar 30, 2015
16,845
if I was to design this circuit from the ground up, I would have used a NPN.
It would be helpful if you gave more context regarding what you're trying to do. You're using PNP transistors as inverters. If we knew what was driving the transistors and how the outputs were being used, there might be a simpler way to accomplish your task (e.g. replacing the 9 components making the 3 inverters with a hex inverter or buffer IC).
 

BobaMosfet

Joined Jul 1, 2009
2,110
I have a strangle thing happening to my circuit and wanted to know thoughts on what might be happening. I have a circuit with three PNP transistors and they all are behaving the same. The voltage on the emitter is 5V and while I'm not applying any voltage directly to the base of the transistor, I'm reading 4.4 V on the base. J2 is a six pin header and I haven't connected anything to it yet.
View attachment 215161
For surety, you can use a 'pull' resistor to pull the base whichever default direction you want, when the MCU isn't actively controlling it.
 
Top