Understanding Negative and Positive polarity (PNP vs NPN)

[milkypostman]

I am new to the DIY stomp box community but have been doing a lot of reading all over about negative and positive ground circuits but something just doesn't make sense in my head.  First, lets assume we have two stompbox circuits N and P.  N has a negative ground and P has a positive ground, and neither is using the chassis as a ground, and further, assume that both have a DC input wired such that the sleeve is positive and the center is negative (BOSS style standard).

What I don't understand is how this causes a short if I use the same DC power supply for both of these.  Forgive my lack of electrical knowledge here, but what actually causes the short?  The way I see it is the positive comes into the N circuit, flows through the circuit, and ends up at the ground, which is connected to the center pin, back on to the DC power chain.  In the P circuit the DC comes in the sleeve but is connected to the circuits ground flows through the circuit (backwards) and ends up at the DC negative pin, back on the DC power chaing.

So what I don't get is with the proper DC jack wiring why this is still a problem.  Note the layout design from GGG that I am linking to below.  It seems like it should play fine with a normal DC power source even when connected to other sources.

What is the actual cause of the short?  As I see it, with proper wiring of the DC jack i'm never crossing positive and negative connections from the DC power source.


Layout:
http://www.generalguitargadgets.com/pdf/ggg_ff5_lo_b69.pdf

[greaser_au]

neither is using the chassis as a ground

there is always a 'signal ground' of some sort (there must be, so that the signals have a 'counterpoint' to be a difference to)...  in this case it's the 'sleeves' of the sockets. This ground is also the 'power ground'. 

In the case of a PNP pedal, this ground will be the positive supply lead, which will also connect to the socket sleeves.  marry that up on your pedalboard with an NPN pedal (where the negative supply lead is connected to the  socket sleeves.  When you  connect them together with a jumper cable you connect the + supply of one to the to the - supply of the other.

Hence the problem, if you are powering them from the same supply...

david

[milkypostman]

AHHHH THE SOCKET SLEEVES!!!

This makes much more sense now.  So by that logic, couldn't you wire up the circuit so that the sleeves are wired to the negative rather than the positive on a PNP circuit?

EDIT:

Maybe that wasn't clear.  I mean, the stereo input jack is just acting like a switch to connect the sleeve to the ring.  On a PNP circuit, rather than wiring so that it's a positive signal being put through the stereo jack, why couldn't you connect the negative signal (by moving the ring/sleeve connection to the other line off the power supply).

[smallbearelec]

Wiring negative ground with PNP transistors has been tried, as you will see if you search and scan previous threads here. Unfortunately, many very knowledgeable people have reported problems with intractable motorboating noise with they did this. It does not happen with every build (and we don't know why), but it happens often enough that it isn't recommended.

Regards
SD

[stringsthings]

I am new to the DIY stomp box community but have been doing a lot of reading all over about negative and positive ground circuits but something just doesn't make sense in my head.  First, lets assume we have two stompbox circuits N and P.  N has a negative ground and P has a positive ground, and neither is using the chassis as a ground, and further, assume that both have a DC input wired such that the sleeve is positive and the center is negative (BOSS style standard)....

So what I don't get is with the proper DC jack wiring why this is still a problem.  Note the layout design from GGG that I am linking to below.  It seems like it should play fine with a normal DC power source even when connected to other sources.

the PSU doesn't really care about negative or positive ground ... it's providing a potential of 9 volts ....

by connecting N and P up to your 9v PSU, you are asking it to provide 18 volts .... it only wants to play with all N's or all P's but not a combination ...

[milkypostman]

there is always a 'signal ground' of some sort (there must be, so that the signals have a 'counterpoint' to be a difference to)...  in this case it's the 'sleeves' of the sockets. This ground is also the 'power ground'. 

So is it true that the signal ground needs to be the same as the circuit ground?  Like, the sleeve of the input needs to be the same as the ground signal of the circuit?

[greaser_au]

Now we're into the engineering model of the 'perfect voltage source' which has a zero internal resistance/impedance... With that theoretical model, there is no reason why the socket sleeves (signal ground) couldn't be connected to the voltage rail rather than the circuit ground, because with a perfect voltage source as the power supply, they are the same point for AC signals. In practice however there is always some significant series resistance/impedance and noise, and with AC supplies, some voltage ripple, which translates into instability, noise or hum at the output.

david

[PRR]

> So is it true that the signal ground needs to be the same as the circuit ground?  Like, the sleeve of the input needs to be the same as the ground signal of the circuit?

There is no "NEED".

But what is "circuit ground"?

In a "simple" one-transistor audio amplifier there are at least three complete circuits. Input comes from guitar, flows base to emitter, and back to the guitar to complete the circuit. Bias flows from battery to base to emitter and back to battery. Power flows from battery through collector to base and back to battery. Output flowsemitter to collector to outside load and back to emitter.

But the transistor only has three legs. There is no way to wire four circuits on three legs except by making one or more legs "common" to one or more circuits. In the "Common Emitter" plan the major circuits are commoned at the emitter (or additional parts in series with the emitter).

We usually design NPN circuits negative-ground and PNP circuits positive-ground. There is a bit of logic here. However it is not necessary. If the power supply is perfectly clean, the PNP emitter may connect to the positive power source just as well as the NPN collectors do. The important thing is that the PNP emitter, sitting way up at plus 9V, has "NO" audio-type noise relative to the negative power lead which is conveniently commoned with NPN emitters and in/out unbalanced signal grounds.