Positive ground, same power supply as other effects?

[seedlings]

I used a NPN Germanium transistor (2N388A) instead of a PNP so that I can use a negative ground so that I can use the same power supply as all my other effects.

From http://fuzzcentral.ssguitar.com/rangemaster.php

I just read this, and being new to positive grounding schemes... can this be overcome any other way than using a battery?  Once I decide on a circuit with Ge PNPs, it would be nice to use it with the same power supply as the other pedals.

CHAD

[nocentelli]

You can build a simple mini board to go inside the pedal with a chip and a couple of other components that will invert the voltage to give you -9v instead of +9v: Search for max1044, ICL7660 etc to find suitable links.  I'm sure others will chip in to give even better suggestions.

[Govmnt_Lacky]

You can build a simple mini board to go inside the pedal with a chip and a couple of other components that will invert the voltage to give you -9v instead of +9v: Search for max1044, ICL7660 etc to find suitable links.  I'm sure others will chip in to give even better suggestions.

I would like to jump in here and ask a question about this 

I am getting a bit old and need some reassurance. For simplicity, I will reference the Tonepad layouts here:

http://www.tonepad.com/getFile.asp?id=82

If I were to configure a MAX1044-based charge pump circuit to produce the -9V supply from a standard wall wart, AND I configured the layout above for a POSITIVE GROUND..... would I hook up the -9V charge pump output to the RED wire or BLACK wire inputs?

Using PNP Germs of course 

[Earthscum]

I am mobile or I would drop a link for you. AMZ has an article on just this. I had to make my friends FF I had built play nice on a 1Spot chain. All you have to do is rewire stuff. It works for quite a few circuits. Only issue he has now is when he turns his volume all the way down with the FF on, it hums. More of a nonissue though.

[seedlings]

I am mobile or I would drop a link for you. AMZ has an article on just this. I had to make my friends FF I had built play nice on a 1Spot chain. All you have to do is rewire stuff. It works for quite a few circuits. Only issue he has now is when he turns his volume all the way down with the FF on, it hums. More of a nonissue though.

Hey- how about this:

http://www.muzique.com/lab/fuzzface.htm

CHAD

[Earthscum]

Hey- how about this:

http://www.muzique.com/lab/fuzzface.htm

CHAD

That is the one.

[Govmnt_Lacky]

Hey- how about this:

http://www.muzique.com/lab/fuzzface.htm

CHAD

Thats the old "swap the ground and power" trick. Problem is that it often results in oscillations (which is indicated at the bottom of that link).

I just want clarification on using a voltage inverter (MAX1044) and where to hook it up to with respects to the Tonepad layouts. I have both circuits but I have no idea on where to connect the -9V output to the Tonepad board.

Embarrased that it is so simple but I have a bit of money invested in PNP Germs that I don't want to lose 

[seedlings]

Hey- how about this:

http://www.muzique.com/lab/fuzzface.htm

CHAD

Thats the old "swap the ground and power" trick. Problem is that it often results in oscillations (which is indicated at the bottom of that link).

I just want clarification on using a voltage inverter (MAX1044) and where to hook it up to with respects to the Tonepad layouts. I have both circuits but I have no idea on where to connect the -9V output to the Tonepad board.

Embarrased that it is so simple but I have a bit of money invested in PNP Germs that I don't want to lose 

Perhaps a big 1000uf filter cap will cure any gremlins.  Plan to rewire my breadboard tonight to check it.

Thank you, Earthscum.

CHAD

[davent]

Hey- how about this:

http://www.muzique.com/lab/fuzzface.htm

CHAD

Thats the old "swap the ground and power" trick. Problem is that it often results in oscillations (which is indicated at the bottom of that link).

I just want clarification on using a voltage inverter (MAX1044) and where to hook it up to with respects to the Tonepad layouts. I have both circuits but I have no idea on where to connect the -9V output to the Tonepad board.

Embarrased that it is so simple but I have a bit of money invested in PNP Germs that I don't want to lose  

I get confused easily by this as well but here goes. Not sure of the correct terminology but the way i try to look at it...  in the positive ground circuit the power source with the higher potential is used for ground so the +ve terminal of the battery is connected to the ground points in the circuit (RED wire pad). The  -ve terminal (typically called ground) connects to the points i see as needing the power connection, R2 & R4 (BLACK wire pad). When you hook up the MAX1044 to provide a -ve power rail you now have a 9v power source with a lower potential relative to your old -ve (typically called ground) from the battery.

So the black wire (-ve from your battery), higher potential, gets connected to the RED wire pad (ground) of the Tonepad Positive ground board. The -ve output of the MAX1044 (lower potential) gets hooked up to the BLACK wire pad on the Tonepad Positive ground board.

(And i always stand to be corrected.)
dave

[Govmnt_Lacky]

Thanks Dave  

So, to summarize and clarify:

DC jack Outer (+9V from wall wart) --> MAX1044 inverter board IN  THEN  MAX1044 inverter board -9V out --> Tonepad Positive Ground circuit BLACK pad.

DC jack Center (0V/GND from wall wart) --> MAX1044 inverter board GND  THEN  MAX1044 inverter board GND --> Tonepad Positive Ground circuit RED pad.

That look about right?

[davent]

Nicely put!

[brett]

Hi

Thats the old "swap the ground and power" trick. Problem is that it often results in oscillations (which is indicated at the bottom of that link).

IF the power supply and ground are well coupled, this is unlikely to happen.
Why? Because the supply line is ground, too (at least wrt AC, and all osciallations are AC).
Try a large electro cap and a small film cap between the -9V supply and 0V ground (or 0V supply and -9V ground, or however else you wish to define them).
cheers

[R.G.]

Thats the old "swap the ground and power" trick. Problem is that it often results in oscillations (which is indicated at the bottom of that link).

IF the power supply and ground are well coupled, this is unlikely to happen.
Why? Because the supply line is ground, too (at least wrt AC, and all osciallations are AC).
Try a large electro cap and a small film cap between the -9V supply and 0V ground (or 0V supply and -9V ground, or however else you wish to define them).
cheers

I have to insert my standard advice. You're right - if the power and ground are sufficiently decoupled it is - less likely - to happen.

But it still does sometimes. There are two kinds of people who swap power and ground: those who have problems with it, and those who think they will never have problems with it, because they haven't. Yet.

IMHO, it is a bad idea for someone who has little enough experience to have to ask the question to try simply swapping power and ground, no matter how well they're decoupled. True, it OUGHT to work fine. Some times, even many times it does.

But there are times when it will not. I've provided debugging advice to a fair number of people, including on this forum, where they thought the swapped ground would work fine, it didn't, and un-swapping fixed the problems immediately. The theory says it's fine. In practice, there are some unknown factors. If I knew what they were, I'd publish it. I've been unable to find the magic elixir, and I've looked.

[Earthscum]

I have noticed in a couple factory schematics that there is a resistor to ground for groups, mostly per chip, but I remember seeing a couple that had groups of transistors with grounds going to a low value resistor, and a .1u cap between the rail and the grouped ground. The chip packages are done the same way, but one per package, like in the Marshall IBS (3530pre).

I was thinking about how well my friend's works, and thought about it. He essentially has the same thing going on with his with the "voltage sag" (actually just resistance between the battery and the circuit's negative rail). Since he mostly uses it with that knob cranked, it should be reducing the interaction with the negative rail, keeping more of the signal contained between the two transistors. Am I far off in my thinking here?

[Paul Marossy]

IMHO, it is a bad idea for someone who has little enough experience to have to ask the question to try simply swapping power and ground, no matter how well they're decoupled. True, it OUGHT to work fine. Some times, even many times it does.

But there are times when it will not. I've provided debugging advice to a fair number of people, including on this forum, where they thought the swapped ground would work fine, it didn't, and un-swapping fixed the problems immediately. The theory says it's fine. In practice, there are some unknown factors. If I knew what they were, I'd publish it. I've been unable to find the magic elixir, and I've looked.

I've always thought it was odd how this works sometimes and sometimes it doesn't. Doesn't seem to follow the tried and true rules, does it? 

[R.G.]

I've always thought it was odd how this works sometimes and sometimes it doesn't. Doesn't seem to follow the tried and true rules, does it? 

No it doesn't.

I can promise you - if it did follow any rules I've ever found, or those of other professionals, including the working magicians of the EE world, RF designers, that I've consulted, I'd have a canned fix for it now.

My best guess is that there are many causes, including but not limited to power supply impedance (in the general sense of both positive and negative resistance, inductance and capacitance), decoupling, and layout, and subcategories inside these. I do have a fair amount of experience with the issue, and I know absolutely that it can be fixed... sometimes. 

This has forced me to my current opinion - it's not something for beginners who have to ask, or have read it on the internet and then wonder if it works.

Kinda like marriage - you really ought to be a stable adult before leaping into that, and for some non-obvious reasons.

[Paul Marossy]

I've always thought it was odd how this works sometimes and sometimes it doesn't. Doesn't seem to follow the tried and true rules, does it? 

No it doesn't.

I can promise you - if it did follow any rules I've ever found, or those of other professionals, including the working magicians of the EE world, RF designers, that I've consulted, I'd have a canned fix for it now.

My best guess is that there are many causes, including but not limited to power supply impedance (in the general sense of both positive and negative resistance, inductance and capacitance), decoupling, and layout, and subcategories inside these. I do have a fair amount of experience with the issue, and I know absolutely that it can be fixed... sometimes. 

This has forced me to my current opinion - it's not something for beginners who have to ask, or have read it on the internet and then wonder if it works.

Kinda like marriage - you really ought to be a stable adult before leaping into that, and for some non-obvious reasons.

Yeah, been there, done that as far as marriage is concerned. 

[brett]

Hi again
I think people make a mistake when they ask "What do I do to the stompbox to make it work?".
It is clear that because batteries work in either orientation, but PS sometimes don't, that it is the PS that needs modification.
Batteries produce a voltage differential with non-polar characteristics (internal R and Z, etc), so it's no wonder that they always work fine when reversed.
But external power supplies are all over the place in terms of polarity. Most have regulation, or worse - regulation with reverse-polarity protection. If they affect the free flow of electrons, there might be an issue.

In my experience a large cap (2200uF) is a key ingredient. Why? Looking "back" from a stompbox to the PS, most devices aren't going to "see" anything except that big, fat, battery-like capacitor. I wonder whether an important characteristic, not shared by simple power supplies is that, like a battery, electrons can be "pushed" back into an electro cap. Or at least it is "spongy" wrt electron flow. Clearly, I'm not an EE !!

Any comments?

[R.G.]

It is clear that because batteries work in either orientation,

A battery does not work in either orientation. A single battery has exactly the same problem that a single power supply does: ground can't be tied to both the positive and negative side of the battery at one time without some interesting and - ideally - educational results.. TWO batteries do work fine; it's just not obvious that the difference between one and two is what's getting you. 

On the professional side, I tend to recommend a second power supply unit to supply the -9V. A second power supply, like a second battery, is a slam-dunk, and it's about $20 in the USA. But that's in my professional life, which we aren't discussing.
 

In my experience a large cap (2200uF) is a key ingredient. Why? Looking "back" from a stompbox to the PS, most devices aren't going to "see" anything except that big, fat, battery-like capacitor. I wonder whether an important characteristic, not shared by simple power supplies is that, like a battery, electrons can be "pushed" back into an electro cap. Or at least it is "spongy" wrt electron flow. Clearly, I'm not an EE !!

Any comments?

I haven't actually tried putting a current probe on a pedal to see if the current ever instantaneously reverses. Just a hunch - I suspect that it doesn't for the vast majority of pedals. If you think about it, being able to reverse current into the power supply, even momentarily, means that the pedal has to store up charge in the normal direction until it has enough in some kind of bucket, then try to force it back into the power supply, fast enough that it overwhelms, at least temporarily, the normal current coming in. This could happen automagically if there was an inductor/capacitor resonant circuit inside the pedal that had a good, low impedance path back into the power supply. As an aside, my comment about

including but not limited to power supply impedance (in the general sense of both positive and negative resistance, inductance and capacitance), decoupling, and layout,

carries inside it the ideas that the circuit may try to push back ("negative resistance") and that it may resonate ("inductance and capacitance"). But having looked for some of those in both non-techie and fairly advanced technical ways, I can't find that as what's happening.

I do agree with you that one or more BFCs is a good first step. The only 9V pedal power supply unit that I definitely know the circuit for has as its final "kiss" to the outgoing electrons a couple of big, burly caps. This, as well as my experience trying to cure reversed-ground oscillation in a captive "bad unit" by hanging both big, little and intermediate caps on the power supply to cure it are part of what leads me to my position on the topic. To be sure, a BFC is a great way to cure some of the instances, and a worthy way to start. It will cure some cases. I'd add to that putting a 0.1uF ceramic in parallel with the BFC, as the inductance of bigger caps means that they start looking like inductors above some frequency not that far above audio.

But I think the advice holds: there are multiple causes and only when the lack of a power source which can be pushed back at is not the only one.