Virtual ground connected to enclosure?

[armdnrdy]

A diode drop above ground using a Schottky is only .3 volts......I wonder if that is enough to cause any issues.

I used 1N5819s for the bridge rectifier for the least voltage drop possible when the 18VDC passes through.

Also, since there is no guaranteed voltage rating of an unregulated 12VAC wall supply, ( I have a 200ma that measures 16 volts with no load) I wanted to be sure that I was above the minimum voltage input rating of the regulator.

[R.G.]

The issue with using signal reference ground and also power supply sewer ground as a source of 0.000000000000000000 volts is that any voltage away from ground is a problem. The classical example is that noise generated from currents flowing form one end to the other of a solid copper wire can be a problem, and that can be in the millivolt range.

Not that you can't get away with it sometimes. But you *will* hear from the folks where it didn't work. 

If you're designing a project for random people to make, or for commercial production, you are going to have problems in some cases when someone wants their 18V DC power supply to power this pedal and also another pedal that doesn't have the ground shift issue.  DIY hackers can cope with it, but not the technically illiterate.

If you simply must have both 12Vac and N volts DC be inputs, the simplest thing to do is go half wave on the rectification. This is not as big an issue as you may think. The filter caps only need to be twice as big (about) for half wave as full wave, as there's twice as long between charging intervals. Make C50 be 1000uF, or make it two 470's in parallel and you're at the same ripple as you would be with a full wave bridge.

Some years ago, I spent some time trying to figure out how to make a pedal immune to whatever voltage and polarity was stuffed into it. I could do this with a bridge and a regulator and some messing about, but I could never solve the ground shift problem. I went so far as to make the bottom-side diodes in the bridge be synchronously-enabled MOSFETs for low drop, and got the ground shift down to a few millivolts, and that was fine for most low gain pedals, but given that the battle cry for distortion pedal users is "MORE GAIN!!", there's always a pedal somewhere which will go nutso over a ground shift. True, I was solving a different problem, including reversed polarity, but that's a subset of AC stuffed into the power jack.

[armdnrdy]

Once again, Thanks for the guidance R.G.

It just so happens that I have a ton of room left at the top of the board where the power supply is located. Fitting larger a larger cap and half wave rectifier will be a breeze!

[R.G.]

I always find that the frustrating issues with the general genre of pedals are the things we can't do anything about: the history which has given us 1M single ended signal cables to 1/4" phone jacks, a difficulty dealing with reversed power supplies and AC on the power inputs, sensitivity to DC offsets and noise, and so on.

Most of these are a legacy of the time when tubes and transistors both cost about $5-$20 in 1960s dollars, which would be $50-$200 dollars in today's diminished currency. The electronics designers were often so focused on saving a part that the designs were cut to the bone, sometimes INTO the bone.  

Consider what the guitar pedal world would be like if the first guitar amps could have used a dual triode input as a differential amplifier reading a balanced, shielded input cord. Triodes were too expensive to use both halves for an input amplifier.

[R.G.]

It just so happens that I have a ton of room left at the top of the board where the power supply is located. Fitting larger a larger cap and half wave rectifier will be a breeze!

Having tons of room can fix a lot of problems!   

[armdnrdy]

It just so happens that I have a ton of room left at the top of the board where the power supply is located. Fitting larger a larger cap and half wave rectifier will be a breeze!

Having tons of room can fix a lot of problems!   

This is a rare but welcome occurrence!

[armdnrdy]

One thing that threw me about this rail splitter is the fact that +V is 7.5 volts and the ground is -7.5 volts but only in relationship with the virtual ground (bias).

If +V is used in conjunction with "true" ground, then we have +15 volts to ground and bias to "true" ground is 7.5 volts.

If there was a emoticon with a spinning head.....I would use it!

[samhay]

it's all relative, and probably always will be.
Just wait till we start designing quantum mechanical pedals.

[armdnrdy]

Well....

I feel on track again!

Thanks again R.G.

I made some adjustments to the schematic, drew a 13X26 7.5 lead spacing horizontal capacitor in Eagle, and made the proper rectifier/capacitor changes on the board routing.

It's a good day!  

Just waiting on the main order of parts.

[R.G.]

One thing that threw me about this rail splitter is the fact that +V is 7.5 volts and the ground is -7.5 volts but only in relationship with the virtual ground (bias).
If +V is used in conjunction with "true" ground, then we have +15 volts to ground and bias to "true" ground is 7.5 volts.
If there was a emoticon with a spinning head.....I would use it!

This is a really good lesson in what "ground" is and why I so very much overuse quotation marks about it. The only "true" ground is a conductive rod driven into wet dirt or submerged into a body of water on the planet's surface. Anything else is a "ground" of convenience - that is, a point we choose to call "ground" and by that we mean only a relative reference point against which all other measurements are made.

Is it more proper to call the negative side of your power supply "ground" or the mid-point that much of the circuit is biased to? The only correct answer to this is that for this circuit in isolation, it does not matter at all, as long as you're consistent with your reference point. Utterly no difference, because the point you hook your meter's black lead to makes no difference to the readings on the red lead, other than a shift in reference point. The differences between voltages on all other points will come out the same.

It's when you hook up something else with a "ground" and want the two to agree on the same connected nets being ground that things start getting ugly and complicated.

Remember: it doesn't matter what point you call ground as long as (1) you're consistent about it and (2) other equipment/circuits you hook up to the "ground" you've just designated can work OK with that point being "ground" for both of them.

[tubegeek]

Consider what the guitar pedal world would be like if the first guitar amps could have used a dual triode input as a differential amplifier reading a balanced, shielded input cord.

Well, I wouldn't have to explain to students why we build unbalanced instrument cables and why the instrument inputs they feed are high impedance. The guitarists in the room get it when I relate the historical sequence of events but nobody else really follows what the hell I'm talking about, especially after I've just spent so much time explaining why low-impedance, balanced inputs are the way to go. Would it have killed Leo to throw a transformer in there?

And if I do it in the opposite order it doesn't really make it any better.

You should have heard me today talking about Victrolas and the Radio Corp. of America so I could justify the RCA phono plug. There's another one there's very little excuse for now.

[armdnrdy]

How did you explain RCA?

RCA was a company that was in existence from 1919 to 1986....you know....three years before you were born.

They manufactured electronic parts and equipment.....what was that?......No.....RCA never made iPhones.

Excuse me....No.....they did not make the iPad.

No Sir....that was quite a while before iTunes.

I'm sure it's a "blast" teaching kids about things that they don't give a rat's AAss about!

We live in a very different world now! People wonder why certain outlets don't carry DIY goods anymore. We're a dying breed.
We have turned into a world of consumers!

[tubegeek]

Well, I guess I didn't explain it so much as refer to it. I kind of hoped that the word "radio" still meant something to them - I supposed I should have called it "terrestrial radio" but I didn't have the heart.

[R.G.]

Actually, in a classroom setting, it's a great entre to the real world note to them that they're engaged in a furious effort to learn the theory of how things work, and that as soon as they exit and try to support themselves on that, they'll run face-first into the wall of real-world things they can't change just because something theoretical would be better.

Rather than making the theory less important, it makes it more crucial, because some things can't be changed for historical, psychological, and economic reasons, so anyone wanting to really understand what they're doing has to look beyond the pure theoretical and into how what they do works economically. Then you REALLY need theory to patch around what the dismal economic world has done to you.  

Very few of my professors hinted that some things that are perfectly possible by the laws of physics are absolutely forbidden by economics, and that this was just as important and no less real.

It might make a great extra-credit question to mention something about economic feasibility and then ask it on a final.  

[R.G.]

We live in a very different world now! People wonder why certain outlets don't carry DIY goods anymore. We're a dying breed.
We have turned into a world of consumers!

Actually, we may be past the DIY Dark Ages, at least the most recent one. In the 80s, there was utterly nothing. I'm still blown away on a regular basis that this forum has nearly 30k registered members and massively more watchers.

It was pretty notable to me when I found a few folks on usenet that were interested in building effects. I figured there was one of me until then. To go from that to a world where people go from "Which end of a soldering iron do I hold?" to "We have released our new Belchfire Mark IV, with its utterly new dynamic tone architecture build with entirely NOS germanium point to point parts in the USA" in less than a week is pretty amazing.

To me. But then I'm easily impressed.   

[tubegeek]

note to them that they're engaged in a furious effort to learn the theory of how things work

True only for very small values of "effort," "furious" and "engaged."

[R.G.]

Point!

[PRR]

> The only "true" ground is a conductive rod driven into wet dirt or submerged into a body of water on the planet's surface.

Yeah, butt....

The only thing a dirt-rod is really good for is Lightning.

Audio systems with NO dirt-rod: every car on the road. Also PigNose battery-amps. iPods. Pocket radios.

The concept we should be groping for is "common".

Current flows in loops.

These loops "can" all be isolated. Separate battery for each tube/transistor. Transformers between stages.

Ah, but tube/transistors have 3 legs and three current-loops, so MUST have some loops "commoned" on one leg. (Photo-devices, also carbon mikes, can evade this, but perform poorly.)

Instead of a dozen or more separate loops, it is MUCH simpler/cleaner to "common" many loops. In a car, most lamps and switches return through a common frame/body.

In a pedal we have input, output, power, and internal loops. We usually common some part of each loop to all other loops. Somehow the mis-leading name of "ground" is used for this common bus. It might be better to just say "common".

[R.G.]

> The only "true" ground is a conductive rod driven into wet dirt or submerged into a body of water on the planet's surface.
...
In a pedal we have input, output, power, and internal loops. We usually common some part of each loop to all other loops. Somehow the mis-leading name of "ground" is used for this common bus. It might be better to just say "common".

Actually, that's what I was going on about. The only "true" ground is pretty useless for our purposes and what pedal hackers like to call "ground".

Well, there's that die-if-you-kiss-the-microphone thing, but that's something else again. 

I agree, "common" would be a less-bad word in many senses, but we're stuck with "ground", just like we're stuck with high-impedance unbalanced signal lines, 1/4" phone jacks and 9V battery clips.   

[armdnrdy]

It might be better to just say "common".

In the line of work that I do, the word "common" used to describe a path to earth, or a return could be a bit misleading.

The word common is used to describe just that....anything that is common. You can have common marked on a control switch that is actually where a "hot" connection is made.

Besides...what's in a name? How about color codes?
We'll make +V red, GND black.......oh wait...that's already been done and it's still confusing!