Earth Ground

[Papa Jim]

On the amp schematics I look at usually the only earth ground to the chassis is the mains earth ground. Is it just assumed that the amps ground eventually gets connected to chassis? I always use a star ground for everything then a connection from the star to chassis. I think I know the answer but just want to be positive about it on my builds.

[vampwizzard]

thats a general symbolism.. grounded earth for electrical equipment will refer to the largest ground plane available. The other symbol (little triangle commonly known as ground vs the rake for earth ground) is used to denote relative ground within this system.

Thats not incredibly important when referring to pedals and amps per se, but it does get important when youre interconnecting systems like all the amps that make up a hammond organ.

[Papa Jim]

thanks

[PRR]

> Is it just assumed that the amps ground eventually gets connected to chassis?

Hum/buzz from the room is greatly reduced if all low-level audio is enclosed in metal and that metal connected to the average potential of the power-lines in the room.

This would include the guitar pickup shield.

With the ordinary guitar plug, this means guitar pickup shield connects to 1/4 plug shell, connects to amplifier chassis, connects to wall power ground.

And for circuit simplicity, one side of every stage of the amplifier also connects to chassis somehow.

This is not inviolatable. Old broadcast gear with truly floating inputs and outputs had no connection from circuit to chassis. There was a jumper to connect them, or opened so that signal shielding would follow signal path throughout a studio chain, connecting to power ground at some distant point (possibly on another power source). But that's far outside a guitar amp's field.

[pdf64]

I think that the use of the term 'ground' to cover both mains ground (ie planet earth) and circuit common connection leads to considerable confusion and muddled thinking. eg folks mentioning that they try moving the chassis mains grounding point in an attempt to get rid of a hum issue.
It's a heritage we've got from texts dating from the golden age of tubes; modern electronic text books tend to use more specific terminology, so it's about the only aspect in which Merlin's document on the topic could be improved.

[jjasilli]

@pdf64:  Wikipedia agrees with you:  "The use of the term ground (or earth) is so common in electrical and electronics applications that circuits in portable electronic devices such as cell phones and media players as well as circuits in vehicles may be spoken of as having a "ground" connection without any actual connection to the Earth, despite "common" being a more appropriate term for such a connection. This is usually a large conductor attached to one side of the power supply (such as the "ground plane" on a printed circuit board) which serves as the common return path for current from many different components in the circuit."    https://en.wikipedia.org/wiki/Ground_(electricity)

Verbally, we always have the option to specify "common", "chassis ground" or the like, vs. "earth ground".  Pictorially, in a schematic it should be clear from the ground symbol used.

[silverfox]

So then why not drive a ground rod outside the back window if you want a real earth ground?. Seriously, how would that work? Isn't that what the old HAMS did? Would you need a filter on the ground plane to match impedance to ground and pass noise?

silverfox.

[PRR]

> Isn't that what the old HAMS did?

Hams are very strange people.

Specifically they try to throw radio waves around the world. They can do this without a dirt-rod, but a di-pole takes twice the tower suspended "far" from earth. At TV frequencies this is convenient (rabbit-ears). At low frequencies it is much better to use the Earth as "the other side" of the dipole.

Hams also have the problem that any large structure will attract lighting and that lightning WILL go to earth! (Though if there is some expensive gear along the path, it will divert though that just for fun.) So those dirt-rods are vital safety measures.

As Wiki implies, much audio can be done withOUT any dirt-rod. E-guitar in rooms with wall-power is a marginal exception: we get less buzz if tied (even loosely) to the power system. And with the hackered-up gear some musicians and PA dudes use, 120V can be leaking all over and it is wise to drain it back to the power system (not earth).

The power company has a problem like hams: their poles attract lightning and in practice they MUST dirt-rod frequently in self-preservation. 105 years ago this was not clear and UN-grounded power was advocated. Easy to show in a lab that you can grab either side of a floating power source no-shock. But experienced power operators knew that all real power lines are riddled with cracked damp insulators, tree branches, leaky transformers, and lightning, and that grounded power is the only way to go.

[vampwizzard]

This has evolved nicely.

Earthing in this case.. driving rods into the ground.. is made to protect the whole enchilada (full power system) especially under a fault condition. Lightning strikes are fault conditions but not massive if arrested properly. Common grounding just gives you the zero plane with which to derive relative voltage.

Most of the time we are close enough to the transformer outside to make sure we meet earthing requirements at the electrical servicing entrance. You need 25ohms or less resistance for a decently grounded power system. If you dont have that then they'll start driving rods.

Most people dont have to worry about that.. but when youre making power centers to go underground for coal mines theres a whole slew of protections put on that ground loop. Current protections, continuity protections, overvoltage, you name it. We get maxed out on cable run lengths as well.. a boon to safety and a companys bottom line!

[PRR]

> You need 25ohms or less resistance for a decently grounded power system.

That's an arbitrary, and almost useless, spec.

Put full 125V into a 25 Ohm ground. Only 5 Amperes will flow. This will not blow a fuse.

Power substations and radio towers aim for much less than 1 Ohm. They may have several miles of ground wires plowed-in.

This is quite impractical for homes. I do not know how the NEC came up with "25 Ohms". And that only means: if you can't prove <25r, you must have a second dirt-rod. Here on thin soil, two rods is still over 50 Ohms, more in drought. I now have 5 rods in various places and I still would not bet on <25r. (I do have a "ditch wire" and most of the year that one may be 10 Ohms; but it is 200' from the entrance.)

[vampwizzard]

25 ohms is a hand me down from the telegraph days. its also why we use 120V.. just the largest transmittable voltage for DC back in Edison's days. They estimated circuit resistance in a different unit of measure.. 1 mile of telegraph wire. The 25ohms comes from the lightning arresters they had back in the day... which also brings us to the biggest fault in EE: If it worked before, lets not explore why or whats most efficient, lets just believe the magic. 

The equipment I worked had an acceptable range of 0.25 - 0.5A allowable ground current. Installs were a nightmare.

[pdf64]

...we always have the option to specify "common", "chassis ground" or the like, vs. "earth ground".  Pictorially, in a schematic it should be clear from the ground symbol used

In forums etc, I always try to use the terminology 0V / circuit common / chassis 0V; the problem is that the things are so ingrained that the response is sometimes 'do you mean ground?' 
Regarding the schematic symbols, in many cases the people that draught them (presemably competent engineers) don't seem to get it either, and the 'earth ground' symbol gets used throughout; a fairly recent Vox design comes to mind for that, I'll try to root it out.

[trobbins]

Yes I find it sad when a schematic has little relevance to how circuit nodes should be portrayed (as I see it), and try wherever possible to present at least the basic 3 links and circuit nodes - ie. mains AC protective earth connecting via a link to a chassis ground, and a single link from a chassis ground to a 0V node for circuitry.

In some circuit configurations, the 0V or B+ of a local stage can be conveniently joined together in a schematic and shown with just one generic 0V or B+ symbol.  That can make it easy to compartmentalise circuitry where just one incoming and one outgoing 0V lead, or B+ lead(s), would connect to and from the next stage - if your aim was to use distributed star power distribution. 

Modern schematic software sometimes doesn't make it easy to show a local star node (eg. for a stage's 0V or B+ nodes).  Some vintage british (as I recall - see link below) schematics made quite some effort to show a local stage node with multiple connections radiating from it - that would have been a useful guide for people wiring up amps - and of course they had much much better schematic capture software back then 

https://tubedata.tubes.se/other/Mullard/TenWattValveHighQualityStereophonicAmplifier.pdf

[Ritchie200]

> Isn't that what the old HAMS did?

Hams are very strange people.

Specifically they try to throw radio waves around the world. They can do this without a dirt-rod, but a di-pole takes twice the tower suspended "far" from earth. At TV frequencies this is convenient (rabbit-ears). At low frequencies it is much better to use the Earth as "the other side" of the dipole.

Hey! I resemble that remark!  Grounded dipoles are for very very low freqs.  Navy used it for communicating with subs, not any more.  Russia still does.  Megawatt input with just a couple watt output will get you under the ocean half way around the world.  Hams use radial grounds on vertical antennas to lower the transmitting lobes - less up, more out.  Hams also use grounds on beam antennas as reflectors.  Hams direct ground their radio chassis to eliminate static and any charges that might build up on the antennas.  This even with and because of, as PRR said, a very unreliable service ground.  Hams will also ground a tower and use a lightning arrestor on the coax. 

If you trace your outlet grounds back to your panel you will notice that they are bonded to the neutrals and you "should" have a separate dedicated ground wire the goes out to the meter base outside (depending on panel and year of home).  You should also see a solid copper ground wire that comes out of the meter base and goes down to the ground, sorry soil.  You may or may not see the ground rod there.  If you have rocky soil or shallow bedrock, you might find a 12" rod, or if you are lucky an 8', or none, or one back at the pole. True ground reference on your grid is a roll of the dice of application and soil.

Jim

[PRR]

> you "should" have a separate dedicated ground wire the goes out to the meter base outside (depending on panel and year of home).

Not required to go to meter. Only from main entrance panel to dirt.

Yes, in many situations the meter is near dirt and the panel isn't. If the meter is on the house it may make much sense to come from the panel out the service entrance hole and then drop to dirt.

I had an exposed dirt-rod which was bent on top. When I pulled it out it wasn't 2 feet. No matter because no wire was ever run to it. In further digs near the meter, which here is on a pole 50' out from the house, I found two 5/8" rods laying horizontal 18" down--- not full code but as good as could be done on this thin land. Since then I added two under the new room, and one by the garage horizontal into the weepy end of the septic field.

I don't remember saying "grounded dipoles". (OK, I see my words were unclear.) The two classic antennas are the dipole (off ground) and the UNI-pole (vertical stick) where the Earth "mirrors" the vertical stick; alternatively we can say it induces EM sheer between sky and dirt. In that vision improving ground conductivity (plowing in much wire) tends to throw waves that "hug the ground".

Yeah, that sub-talker was just some north of where I am sitting.

[jjasilli]

Even 0V can be misleading to the unitiated.  Though the chassis may be 0V relative to the circuit it houses, it may have a voltage potential (not = -0-V) relative to Earth.  This could even be dangerous to the unsuspecting person.


OV may not be literally true; it's a useful convention.  The chassis could be -400V relative to the tube plate.  Then the plate is +400V relative to "chassis ground".  So the tube will function happily.

[Tbone55]

The electrical system in a car is grounded but not to earth. You've got rubber tires which insulate the car from the earth. If you were in a lightening storm in a car you would be safe as along as you don't touch anything metal in the car. When you look at an electrical schematic for a car the ground points all relate to the cars' chassis, not earth ground. The electrical panel in your house is earth grounded via a large copper stranded wire that usually connects to the main water line coming into the house. As I understand it, this is for protection in case your house get hits by lightening. I to find grounds a bit confusing.

[PRR]

> in case your house get hits by lightening.

That; also if the utility company line is hit or near-missed by lightning; and for some utility company mishaps which put high voltage on your electricals.

And you are no longer allowed to have ONLY a water pipe for dirt-bond. (My last house, the copper turned to plastic 6 inches outside the cellar wall.) You should have a full 8 foot rod. Without testing or engineering, in recent years you are normally supposed to have two rods spaced apart (there are alternatives but not for dummies). You *in addition* bond all known metallic underground piping, so everything is on the same plane. (Gas piping is a very touchy topic.... let your gas company be the boss on that.)

[Tbone55]

When I built my house (31 years ago) I had 3/4" copper piping going all the way from street connection into the house to the water meter. I guess these days with the greater use of plastic piping there's probably a greater chance of the main connection not being all copper. I don't know if the building codes may have changed but I would imagine they had to address the grounding issue in some manner. 

[Ritchie200]

You *in addition* bond all known metallic underground piping, so everything is on the same plane. (Gas piping is a very touchy topic.... let your gas company be the boss on that.)

Speaking of lightning and gas lines....  CSST requires bonding back to the panel due to pin holes being blown out of the CSST to the nearest ground potential when there is a close by lightning strike.  That could ruin your day.

Jim