loop box ground lift - any objections? i've never seen it done.

[darron]

hey guys. i wonder why i've never seen an effects loop pedal, or loop box if you will, with one jack isolated for either the send or return. if you're making a big effects loop, then you're making a nice big ground loop too right? especially if you were going to make one of those big loop strips, which i have no use for. i don't like the idea of calling it an 'earth lift'. earth has always been a safety word for me, and it sounds like 'safety lift'. can't we call it neutral lift or something like that?


we can't have true bypass loop boxes with active transformer isolations... what's the story?

[Dai H.]

it would seem that ground lift switches would ordinarily mean lifting a signal ground and never a safety ground (so I'm not sure there would be a confusion since it doesn't make sense to put the safety ground on a switch to enable disconnection by the user). Getting a connection to safety ground via a signal transmission line wouldn't seem to make sense also. It makes sense that a safety ground typically connects and disconnects along with the AC cable (at the same time as connection to an AC power line).

[darron]

thanks for your response Dai H.

it's an issue that i'm not all that sure about. i don't work on amps so i'm not sure exactly how they work, but i believe signal ground actually does connect to the neutral line, which will be earthed.

last year i came up with a design to switch between amps were the 3pdt would only take the audio signal and earth from one amp, completely isolated from the other, and then it would ground the unused amp's signal using it's own ground connection. everyone said it was dangerous and not to lift the earth and to use isolation transformers etc. we don't have any 2 pin plugs here in australia.

a bit of clarification on the matter would be absolutely fantastic.

maybe in america or somewhere, you might have a situation where you have the looper box plugged into an amp with a neutral casing, but then someone plugs into the  return jack their rack unit with an active casing? i'm not sure i understand the danger, be it isolated or not. you'd still get a kick as soon as you picked up the plug.

erg. anyone .... helllllllpppp ?

[R.G.]

hey guys. i wonder why i've never seen an effects loop pedal, or loop box if you will, with one jack isolated for either the send or return.

You have. Well, someone has, if not you. See "The Wrapper", GEO, Feb 9, 2002, "Beyond the Third Generation". I used the differential amplifier version of a ground lifter instead of the transformer version. Both work. Both ought to be more common.

we can't have true bypass loop boxes with active transformer isolations... what's the story?

There's no reason you can't.

Last year i came up with a design to switch between amps were the 3pdt would only take the audio signal and earth from one amp, completely isolated from the other, and then it would ground the unused amp's signal using it's own ground connection. everyone said it was dangerous and not to lift the earth and to use isolation transformers etc. we don't have any 2 pin plugs here in australia.

a bit of clarification on the matter would be absolutely fantastic.

I'll try.

First: I don't know as much about Australia's electrical utilities as I do the ones in the USA. In the USA, the power from the electrical utility transformer comes from a center tapped winding, the centertap of which is earthed at the power pole by a copper wire running down the pole into the ground. Both ends of the winding, the centertap, and the ground wire are taken to the premises. In the USA, the result is one "safety ground" wire, one "neutral" wire that is theoretically at the same voltage as the safety ground, and two power phases, each of which is 120Vac with respect to neutral and ground, and 240Vac with respect to each other. I don't know if you use 240/0/240 in Oz or not. But I suspect the safety ground and neutral are wired the same way.

Neutral carries the return current which comes from the power phase. As a result of Ohm's law, out at the wall-socket, the voltage between neutral and safety ground diverges when any power is being used. You can see as much as 5-6V difference in odd cases. This is how the power system works normally.

Safety ground is there to soak up any electricity which escapes from the power/neutral loop. Electrical safety regulations require that any metal which a person can touch must be connected to safety ground through a resistance of (I forget exactly) ohms, and this connection must withstand something like 30A of current. That way, you are ensured that any conductor that a person touches is at a safe potential and will stay there for all foreseeable electrical failures. At least that's how it works in the USA.

In a guitar amp with a third wire, the metallic chassis and signal ground must be connected to safety ground for both safety and hum reasons. If you disconnect the signal ground from chassis, you get hum. What gets complicated is when you have two amps, each with signal ground connected to its own safety ground. The two safety grounds may be a few volts apart because of the differences in wire resistances and current use on the branches, and also because each amp will leak a small amount of AC if only capacitively. If you then connect them together, the safety grounds settle the issue of who's more at 0V by letting a current flow between them until they satisfy Ohm's law over the different voltages and interconnecting wires. The resulting slight ohmic voltages in the connecting cable is what causes ground hum that's not inductively picked up.

Hold on, I'm getting there!

But as long as both amps have solid connections to their own safety ground, there is no safety issue breached. The problem is that hum you get.

I believe that it is safe to separate two signal inputs to two safety-grounded amps by a transformer. The transformer does not have to be rated for line isolation voltage, because the difference in voltage between two safety grounds can never be more than a few volts, and much less than line voltage in any case of proper electrical wiring. So any connection that leaves both safety grounds connected to the amps, and also always connects the signal ground that you touch to one or the other safety ground is still as safe as it would be plugging into one amp by itself.

Moreover, the differential-amplifier version of an isolator works because one of the signal grounds is alway driven with a voltage representing the difference between the two signal grounds. Again, in this scenario, both signal grounds are still connected to their respective safety grounds.

It is dangerous to lift either safety ground from signal ground inside the amp chassis. If this is done and there is a fault inside the amp, it can put full AC line voltage on the signal ground you touch. Then when you touch both that chassis and another piece of equipment that is properly grounded, you get electrocuted. But as long as safety ground is connected to signal ground inside both amps, no electrocution can take place. The transformer or diffamp take care of removing the different hum voltages from the amp inputs.

[Dai H.]

last year i came up with a design to switch between amps were the 3pdt would only take the audio signal and earth from one amp, completely isolated from the other, and then it would ground the unused amp's signal using it's own ground connection. everyone said it was dangerous and not to lift the earth and to use isolation transformers etc. we don't have any 2 pin plugs here in australia.

Dunno if I can add anything to R.G.'s excellent post, but do you mean you were putting the AC safety ground through the 3PDT and switching it(I'm not quite able to picture how you hooked things up)? My understanding about the hum from connecting the signals of two grounded amps together and creating a ground loop (and thus hum) is as above (i.e. it's an audio issue not a safety). What I've seen as methods to solve/break the loop are things like transformers(allows passing signal and breaking of ground), a bit of resistive isolation between circuit ground and safety ground(DIY article on Elliot Sound Products site), connecting the chassis with a big wire(I guess is to force the potentials to be the same).

[MarcoMike]

here is what came into my mind reading your post:

if you have some effects in your  amp's loop and some of them before the input, they share the same ground because of the power supply. then say the amp's ground is "referenced" to the one of your pedals through the first cable that goes to the input. and the pedals in the loop are referenced again to the same amp ground.

this way there are several ground loops, is it right?

if we take out the ground wire from our loop send and return those pedals will still be referenced to the same common ground through the power supply, but now we have some ground loops less... is this correct? my actual knowledge of electronics is quite shallow...

[darron]

Thank you very much for your replies guys. I was kind of hoping RG would shed some light. Forgive me for not getting back to you right away, I've just been chaotically bust very recently. Hopefully I'll be able to read over it again tonight and continue discussion

Thanks again.

Darron

[petemoore]

  Could swear it was this thread....
  Someone had mentioned real nice audio 1:1 transformer supplier while mentioning that they're usually more expensive for 'x' quality and width of bandpass.
  I don't get that much hum, but should put some isolation transformers in my splitter.

[TELEFUNKON]

  Could swear it was this thread....
  Someone had mentioned real nice audio 1:1 transformer supplier while mentioning that they're usually more expensive for 'x' quality and width of bandpass.
  I don't get that much hum, but should put some isolation transformers in my splitter.

http://www.diystompboxes.com/smfforum/index.php?topic=67803 ?

[darron]

Okay. Back to business! or, away from business and back to leisure. whatever!

Thank you very much RG. I wasn't aware of the centre tapped transformer for the power. Seems things are done slightly differently there. From my understanding, here the power hits the transformer on the power lines, and then your house gets delivered the active and neutral lines (240v AC, 50hz). My house has the three actives coming in 120 degrees out of phase. At the switchboard the neutral gets hooked up to earth via an earthing rod of a meter or so in the ground, so neutral and earth are pretty much the same thing but for convention and logic there are the three pins. Then you have the RCD's etc.

Well, you COULD have the passive true bypasses, but those jensen transformers cost heaps

It's good that you clarified for me that the signal earth does in fact go to earth, and of course to the amp's chassis's earth. i did once hook my signal earth on guitar to my garage's frame and got heaps of noise back, so that makes sense (:

okay. with RG's wisdom we have the theory for why ground loops are (potentially) evil. with explanation, i now have confirmation that there's potential for high voltage AC to get carried along the signal path into your hands, which can then return to the earth via your feet or even worse to the earth of anther piece of hardware.


so for those reasons, it sounds like i SHOULD NOT include a simple ground isolation on an effects loop pedal for the event that one day somebody could use it for evil and plug an amp into the return or something which could fail.

last year i came up with a design to switch between amps were the 3pdt would only take the audio signal and earth from one amp, completely isolated from the other, and then it would ground the unused amp's signal using it's own ground connection. everyone said it was dangerous and not to lift the earth and to use isolation transformers etc. we don't have any 2 pin plugs here in australia.

Dunno if I can add anything to R.G.'s excellent post, but do you mean you were putting the AC safety ground through the 3PDT and switching it(I'm not quite able to picture how you hooked things up)? My understanding about the hum from connecting the signals of two grounded amps together and creating a ground loop (and thus hum) is as above (i.e. it's an audio issue not a safety). What I've seen as methods to solve/break the loop are things like transformers(allows passing signal and breaking of ground), a bit of resistive isolation between circuit ground and safety ground(DIY article on Elliot Sound Products site), connecting the chassis with a big wire(I guess is to force the potentials to be the same).

That's exactly what I did. I have a diagram if you are interested, but it's a bit difficult to wrap your head around. What it did was this:

1- Take the instrument signal and switch it to one of the amps
2- Take the instrument earth and switch it o the appropriate amps earth, while maintaining isolation from the other earth. The instrument earth is always connected to the enclosure.
3- For the unused amp, it's own earth is also routed into it's own signal path to mute.
4- On top of all of that, it will change between two LEDs to show which amp is active!

Sounds far fetched... but it's in this diagram (all same coloured wires, sorry) http://members.optushome.com.au/bluespherecreations/forum/isolating-ab.gif
Like i said, last year when i put the idea forward everyone jumped on it as being unsafe, but i never knew why. Possibly is the 3PDT failed it could become unsafe?

  Could swear it was this thread....
  Someone had mentioned real nice audio 1:1 transformer supplier while mentioning that they're usually more expensive for 'x' quality and width of bandpass.
  I don't get that much hum, but should put some isolation transformers in my splitter.

http://www.jensen-transformers.com/as/as013.pdf

[Dai H.]

okay, as far as I can gather you are NOT switching the safety ground (the connection coming from the wall) just the signal return/shield (which is connected to ground) so I don't believe there is a safety issue, just some confusion from the terminology. I guess this scheme solves the loop problem by breaking it when switching (no noise?)?

[darron]

okay, as far as I can gather you are NOT switching the safety ground (the connection coming from the wall) just the signal return/shield (which is connected to ground) so I don't believe there is a safety issue, just some confusion from the terminology. I guess this scheme solves the loop problem by breaking it when switching (no noise?)?

well yes, that's more accurate. i'm only talking about the signal ground, which then apparently goes to actual earth.

so from what R.G. says: don't disconnect the earth on an amp - obviously. don't disconnect the neutral/ground from the amp's earth and chassis. signal must be earthed to EARTH by at least one connection always.

all this makes sense and it's what i've always thought, but i've always had people come down on me saying that it's a no-go. hmm.

oh, to answer your question about noise, i'm not sure because i haven't built it yet for concerns above. it should be good. i'll give it a shot, maybe tonight.

so i'll keep the ground isolations. cool! i'm pleased that i have an answer, and i'll stick by this answer until i come across a problem. if you don't hear from me, then yeah.....

thanks guys!

[Dai H.]

they probably meant don't disconnect the AC "Earth" connection ( = "safety ground").

http://www.accesscomms.com.au/Reference/powerplug.htm

the way I see those earthed plugs is that the designers intended the EARTH connection to always be connected when the equipment is connected to an AC power source and only disconnects when the equipment is disconnected from the AC power source.

[darron]

they probably meant don't disconnect the AC "Earth" connection ( = "safety ground").

http://www.accesscomms.com.au/Reference/powerplug.htm

the way I see those earthed plugs is that the designers intended the EARTH connection to always be connected when the equipment is connected to an AC power source and only disconnects when the equipment is disconnected from the AC power source.

that's right. the earth pin is actually a bit longer than the active and neutral so it plugs in before anything, and disconnects after everything. everything has an earth pin unless it is double insulated. all this time i thought there were safety concerns with that little ground shielding, which i didn't understand since there could be such high resistance on the ground (compared to fractions of an ohm on an actual earth line)