Ground Loops

[JohnH]

Here are some questions which have been bugging me. What problem does a ground loop within a guitar really cause? What does it sound like? Why does it happen? How can one demonstrate this?

Ground loops which exist between pieces of mains powered audio gear I can understand, I’ve caused a few myself. They are related to voltage differences between different parts of a buildings grounding system, caused by leakage currents etc, due to mains powered equipment. The answer is usually to make sure there is only one ground for the system.

Our guitars are not affected by mains currents, the guitar has only one overall ground route, through the lead. Any ground loop effects can only be caused by induced currents in the control wires within the guitar. I could see that some small signals could be induced in a low resistance path around the ground connections of the guitar controls etc, but they are also mostly shorted out, because they are grounded.

I’ve tried to create a ground loop: I put a 33k resistor between ground and hot connection of a guitar lead (representing guitar pups, pots etc), with the ground connection to it being via two separate wires (either the same or different lengths), physically separated. I plugged in and cranked up the overdrive gain to get a nice buzz. Breaking the ‘ground loop’ by disconnecting one of those wires caused no change. Can anyone demonstrate a ground loop?

My guitar also has a ground loop in its circuitry. Since I’ve shielded it its been pretty quiet, but breaking the ground loop caused no change.

Honestly, I am not seeking to question 10 years of Guitarnuts experience. I know ‘Star grounding’ works well for lots of people. But Id like to understand!

regards

John

[bam]

Ground loop is avoided in a input-amp circuitry since it can act as an EM antenna, thus it (they) can pick up any EM frequencies from the outside world, i.e. from fluorescent lights (buzz), radio stations (GN1 veterans, remember the Mexican polka guitar?)..

This is eliminated by doing star ground (a.k.a. putting all grounds together in one place). <1st prevention stage>

The QTB™ also includes another prevention from EM freqs by making a Faraday cage (from aluminium/copper foil) which picks up all EM freqs and send them all to the ground point. (i.e. making a big ground loop that covers the entire electronics which redirects EM interferences to the ground). <2nd prevention stage>

As you have said yourself,

The answer is usually to make sure there is only one ground for the system.

Now the final <3rd> prevention stage lies within your home's grounding system; To put it simple, one must make sure that the ground point (the star) reaches the Mother Earth. If not, then you're making an even bigger ground loop.

I’ve caused a few myself. They are related to voltage differences between different parts of a buildings grounding system, caused by leakage currents etc, due to mains powered equipment.

[JohnH]

Thanks Bam - nice try! Theres no question about the shielding (second stage), nor the issues relating to grounding to the earth (third stage). I'm still looking for more understanding of why a loop within a guitar ground circuit, which overall has only one connection to ground (via the lead), picks up more noise than similar wires, which are not in a loop. It should be possible to replicate this effect, but I can't!

Any further contributions?

[bam]

I think it's more because it acts like one winding of a coil.. ?

[JohnH]

yes agreed, but its a coil that is shorted out, being a loop of low resistance ground wires. Its a bit like bypassing an SC pickup by grounding both ends. Its contribution to sound and hum is negated. There are currents generated in this way, but which ever way I think of it, I cant see how they cause significant effects in the signal (while not denying that they do!). Even when I think of connecting to different parts of the loop, and consider the induced currents flowing and the voltages implied, it still all cancells out.

I really appreciate your effort to help think about this.

[StratLover]

I was thinking the same one winding of a coil myself, but I wanted to be sure, so I dug deep into the primoidial abyss of notes and files on my comp. only to be shut-down by the wifey to go to a honoree's banquet tonight in her honor. 20 years in the education business.

Just got back about an hour or so ago and I see that I was not the only one feeling this.

I really think it only stands to reason that it does actually act as an aid if you will. I had an open guitar on the slab which is wired stock. Curiousity overtook me and I started tapping on the loops just to see if you could actually get an audible response........guess what? You do indeed get a smaller ping from the loops, but I could only get an audible out of positions 1, 2 & 4.

Ok you guys lets figure this one out. WHY in 1, 2, & 4 only, and as differences do occur, which positions based on the wiring does this occur in other wirings.

[bam]

Heh. It's just a banquet :lol: .. you DO love your wife, right ?

.. Okay, this is all MHO.

If the ground loop is made straight (i.e. SC grounded on both end) it won't made any difference; any signals picked up will be sent to the ground; But in guitars with ground loops, the problem lies within the tone cap since it is supposed to bleed some of the signal from the coil to the ground. My guess is, it can work in reverse (sending the signal from ground loop to the + lead). After all, it's not a flat-voltage DC current but a waveform, right?

I think it's on 1 because the 1 pos coil always has higher output (i.e. the hum is picked up by the pup itself), and on 2&4 because the output waveform from the pups is humcanceling, thus when another hum cycle (from the ground loop) gets in it will make another hum.

(hopefully anyone can understand.. Well, just hypothesizing, after all.. )

[JohnH]

I came across the following quote from Chris Kinman (famous pickup maker), which led me to post it on this thread:

Star grounding, is it necessary?: Star grounding in a guitar is not the absolute necessity it is in active circuits like CD players where it is designed to inhibit ground loops. Guitars don't suffer from ground loops. It is practiced in guitars more as a convenience that serves simplicity, convenience and neatness. That means as long as all the ground wires ultimately connect to the output socket ground terminal all will be OK. In some circumstances it may be more convenient to piggyback onto an intermediate grounded point that is different to the central main one (volume pot cover).

It's from the Kinman website, at:
www.kinman.com/html/myProducts/strats.htm

I think he's right - anyone disagree?

John

[bam]

I'm not "disagree" with that, just saying "it could be true".
The fact is, there is no one I've known who had made a hum comparison between 3 guitars:
- star grounded only
- shielded only
- star grounded + shielded (QTB™)
But the interesting point is, in some Cort and Samick guitars I've opened, they looks like "star grounded only" type. And they hums in 60 cycle hum. (meaning, what Chris Kinman said seems to be right )

Hey, now I wonder, is the member "ChrisK" Chris Kinman himself ?

[bluesman13]

Hi JohnH,
Just reviewing some QTB threads as I undertake a new build.Any recent thoughts on this? I'm not sure who I would implement this as Chris Kinman suggests - all the grounds attach to a common point, then attach to the ground lug of the vol pot whchi then attaches to the neg lug of the output jack?

[JohnH]

Hi Bluesman13, no new thoughts on this really. But then 18 years is plenty of time in which to think of something outstanding and then to completely forget it!

Avoiding ground loops and using Star Grounding was a significant discussion item on the original GN and early GN2. But I never understood why it would be important within a passive guitar. I don't think it is so, Id rather make sure everything that needs a ground gets one, by the simplest shortest route, that may be more of a 'chain' rather than a 'star'. or a combination.

[unreg]

Jun 14, 2005 1:59:19 GMT -5 JohnH said:

I came across the following quote from Chris Kinman (famous pickup maker), which led me to post it on this thread:

…Guitars don't suffer from ground loops.

JohnH, my guitar did indirectly suffer from a ground loop.

This is why:

Ground loops which exist between pieces of mains powered audio gear I can understand, I’ve caused a few myself. They are related to voltage differences between different parts of a buildings grounding system, caused by leakage currents etc, due to mains powered equipment. The answer is usually to make sure there is only one ground for the system.

Your words above ring true! My pedals were powered from a separate outlet, on a different wall, than my amp’s power-strip’s outlet. So, obviously a ground loop was present… and that loop caused noise from my amp. I destroyed that loop, thankfully, by plugging my pedals into my amp’s power-strip. Then my guitar playing is noise-free! That’s such a blessing!

(The organist at my church suggested I had a ground loop AND how to attempt to solve it.)

[newey]

unreg-

The key word there being "indirectly". You're not talking about a ground loop inside the guitar's wiring.

Not too long after this thread was started "way back in the day" by JohnH, ChrisK wrote the definitive explanation on ground loops in guitar circuitry. I can't find the link to that at the moment, but the upshot of it was that a ground loop cannot, even in thoery, cause any noise unless it is carrying the guitar signal. Thus, for example, wiring the two tone pot's shells on a Strat together doesn't cause a ground loop, as no signal is flowing through that connection.

Further, even if a ground loop did exist in the signal path, it would be unlikely to cause any noise anyway, for the reasons JohnH stated in the first post above.

[unreg]

newey, I fully understand. Just trying to voice that my guitar’s sound did suffer from a ground loop. Which has nothing to to with my guitar itself, but, at the time, I didn’t know that. Without that ground loop, the buzzing vanished! So thankful and happy!

[bluesman13]

thanks! being that I am not electrically inclined (and probably oversimply things), is it feasible to attach all grounds (pots and pickups) to each other, then attach those to the ground lug of the vol pot with a wire that goes to the output jack? (does that make sense)? is that what Chris K was suggesting?

Also, if may go OT and ask another question. I bought some 250k pots last week and tested them with my multimeter and they tested at 225k. I asked the vendor for an exchange (since I thought they were defective), and he said up to 25% tolerance is within "industry standards". Is that correct/

[JohnH]

Hi bluesman13

Thst grounding arrangement sounds ok to me.

And I think it's true about the pots. They aren't as accurate as fixed resistors, and at 225k I expect they are within spec. Also, that difference wouldn't make an audible change. (It is feasible to open them up if they are full size pots and scrape the track to increase them though, I've done that often...more on that if its of interest)

[bluesman13]

thanks john. I saw somewhere (maybe it was here), that to create a blender/pan pot, the person opened up the pot and cut the inside of the ring to break the continuity. It looked interesting, and I thought about trying since my currently purchased emerson "blender" pot stopped working. When I attached the multimeter to it, there was no output even after turning the knob in both directions. So, if you have some info on changing the output of a pot, I'd love to know how to do it. Since getting COVID in December and not getting better (precipitating my exodus from work), I have lots-o-time on my hands, and this type of detailed work fits my personality perfectly, and gives me some hobbies to do. 😊. Appreciate the assistance.

[JohnH]

hi bluesman, sorry to hear about the long covid.

With pots, there's a few different constructions. There are videos on YouTube, and this one looks helpful:



Assuming simple single-gang pots, not with switches, the starting point for doing stuff to it is to prise up the 4 lugs that hold the back to the front. Just enough to get it apart. These can be squeezed back down later and with care, you can do this 2 or even 3 times, but once should be enough.

Then you'll find the wafer containing the track, and with luck it will be like the second one in the vid and you can separate it to work on. If you find it trapped under a plastic flange, probably back-track and reassemble. The easy ones are CTS style, and I've had no problems with ones on my Strat or LP, and most others of 24mm size.

With the track (with solder lugs) removed, you can clip a multimeter across the outer lugs and measure R to the nearest k. Now you can tinker! If you want to stretch a below spec pot back up to spec, use a fine blade to lightly scrape the edge corners along the sides of the track. Concentrate on the half of the track that is the max volume or brightest tone, since that's where most of the Ohms are. No need to scrape the middle of the track width.

The nice part of this process is that as you scrape, you can watch R increase k by k until you arrive at exactly 250k, if thats your target. Then you have a pot that has 0% tolerance!

Then reassemble, squeezing the lugs back down. You might clean tbe track with a lubricated cleaner first.

The other similar process I've done a few times is to create a no-load pot. For this, you scrape across the end of the track at the max = high resistance end, until the meter goes off the scale (on a 2M scale)

Provided you can risk a pot, then there's not much to lose by trying this if you need to. I've probably tinkered with a dozen or so and not lost one yet.

[bluesman13]

thanks John! I'm going to try this after I finish this build.

[bluesman13]

Holy crap John - that actually worked! (thank you)I had an extra alpha 500k pot that read 425k on the multimeter. I removed the shell and scraped the side of the track as you said and increased it back to 500k. That was fun - and really easy. The pot I used did have a plastic wheel on the inside but I went under it and scraped on the black track below. I wasn't able to tell which direction was max as the casing was off and the spindle just turned freely, but I picked an edge and scraped gently with a small screwdriver. One thing I was wondering, was about the configuration of the lugs - the middle lug is always hot, but the two outer lugs are related to whether the knob turns CW or CCW. The big contention I read about was that CW rotating analog pot when reversed wired (as Fender does on their left handed guitar's) that the sweep is effected, that is that the pot acts like a "on-off " switch. When I made the regular analog taper pot hooked up to the multimeter with the red on the middle lug and the gnd on either of the outside luge and turned the shaft in both directions, the sweep appeared to be gradual. I was expecting that the readings with not change until one point where it would off quickly. Is it because there isn't a correlation between resistance and audible volume?

thanks,

Steve

[JohnH]

Thats great. I like it too, and I enjoy how exact and controllable the process is as you watch the reading gradually rise.

So, any working pot must have some kind of a gradual taper, rather than really on-off, otherwise it's not much use!

Linear pots should be close to 50% from left lug to centre, and centre to right lug, if set at half turn. Usually this is called a B taper.

A standard audio pot, A taper, wired normally for a righty has a resistance that is less in the first half turn. Say you look at it from the shaft side, like looking at the guitar front face, and the lugs are down. At zero volume, the centre lug is connected to the left lug, and if you turn it clockwise half way, it will be about 10 to 15%, between centre and left. Then the rest adds on between half and full clockwise turn.

This approx logarithmic increase, while not linear on the meter, sounds like a gradual volume build up and a nice smooth increase.

But if you wired that A pot as a lefty volume, with max vol at a full anticlockwise turn, then the apparent volume will increase really fast from zero, almost like a switch. Hence, we have reverse audio pots, C taper that are the reverse of A taper.

[bluesman13]

interesting, thanks John. I've been reading threads online about pots - analog, linear, reverse taper (called "left handed"). Do right handed guitars turn CW or CCW to increase the vol? It seems logical to me that a CW  turn would increase the vol/tone? I thought I needed reverse taper pots to get a full sweep, but I'm reading that reverse taper pots turn CCW to increase the vol/tone. Why would a left handed person want to turn a knob CCW to increase the vol/tone, when a right handed guitar the pots turn CW to increase? Am I understanding this correctly?

[JohnH]

Facing the guitar front, a normal right-handed A-taper pot in a normal guitar turns clockwise to increase volume.

But personally, I also don't really get why left handed needs to be reversed, since everyone needs to turn knobs both up and down, whoever they are and whatever they play!

[bluesman13]

thanks John. Do you know if I use right handed pots in my left handed guitar (wired CW), would the pots turn CW to increase with full sweep?

[JohnH]

Yes that'll be fine. You just have to wire them as in a right-handed guitar, ie per all the diagrams here and online.

Looking at the back now, as you wire it up, lugs assumed down:
Right lug to ground
left lug to switch and/or pickup outputs
middle to guitar jack hot.

That, with a normal A taper pot, should give you a smooth build up as you turn CW (looked at from front!)

[bluesman13]

right, got it. thanks. For the life of me I can't understand why a left handed guitar would have (or any player would want) a knob that would turn CCW to increase (the opposite of a right handed guitar). Well, thanks for helping me sort that out John! :-) Steve.

[reTrEaD]

May 25, 2023 17:00:43 GMT -5 bluesman13 said:

Why would a left handed person want to turn a knob CCW to increase the vol/tone, when a right handed guitar the pots turn CW to increase?

I reckon it has to do with the natural rotation of the right wrist vs the left wrist. But this seems like a good question to ask ssstonelover.

[JohnH]

Yes but....my point is...you can't just keep turning it one way...you soon hit the end stop! So you have to turn it the other way too. So did it really matter which way was up?

[bluesman13]

I think you're saying that you have to turn both CW and CCW anyway, what's the difference if it's CW to increase or decrease? It feels more intuitive (to me anyway) for clockwise to increase, as does many other devices with a knob. I suppose I could acclimate to anything, but as a lefty, maybe there's a little chip on my shoulder that I want it to work the same way as a right handed guitar? I wish that when I started playing that I learned righty, things would be so much easier, but at least when I play with others, no one wants mess around with my axe! 😀

[stevewf]

May 19, 2023 22:39:42 GMT -5 JohnH said:

Avoiding ground loops and using Star Grounding was a significant discussion item on the original GN and early GN2. But I never understood why it would be important within a passive guitar.

Old question: wouldn't a shielded cavity be an infinite number of ground loops, anyway? Isn't there a ground loop that goes VolPot-TonePot-Cavity-VolPot? Another that goes EastCavity-SouthCavity-WestCavity-NorthCavity-EastCavity? And then spin that 360 in 3 dimensions (=infinite)?Makes me a non-believer in problems caused by ground loops inside a passive guitar.

...and really, in any guitar, passive or active, that's not directly attached to a mains ground. Is any guitar ever purposefully directly attached to mains ground?