Tracing Operative Current Flow in a 5E3

[peteb]

do you agree that according to the layout drawing in post #5, the only path back to the HT CT from V1 (as well as V2) is via the input jacks and chassis?

yes. There are other paths to the chassis as well, like thru the volume pot, but what you suggest is the only direct path with very little resistance.

You did say, "I don’t think it will happen the way that your wiring diagram is wired,"

I meant that I did not see how AC ripple in the DC operating current is going to affect the signal at the input Jack.

but then in a later post you said, "that is the way I see it too,"

I was agreeing with your assessment of the current flow. I just then understood the current path through the black wire at the U turn.

Fwiw, by "input jacks," I mean the ground or sleeve portion. If pulsating DC passes through there, the little field it creates is in very close proximity to the hot guitar signal, and noise can be induced. 2L man also commented on this.

That is a good point and something that I did not consider.
this could be tested for as peegoo has suggested. Build it both ways.

however, I still doubt it would happen. I think that the ground bus and the ground side of the input Jack all stay at ground potential. the small AC ripple in the DC operating current that flows through the V1 and v2 cathode, will impart a small AC voltage on the cathode resistor, developing across the cathode capacitor. The closest I envision that small AC voltage getting to the input Jack, is the circuit node where the cathode resistor meets the grounded ground bus.

I personally do not see how this small voltage has an opportunity to induce AC onto the signal portion of the input Jack.

If you think that other return current paths exist in that specific (and very common) layout, from the preamp ground bus back to the PT, where are they? Can you identify another path in that example?

like I said above, there are other paths to ground, but I agree with you that you have identified the only direct path to ground with nearly no resistance.

 

[gabasa]

@peteb, thanks for responding.

How can current flow to the chassis via the volume pots, when nothing is connected to the backs of the pots? I don't understand.

 

[peteb]

thanks for responding.

you are welcome.

How can current flow to the chassis via the volume pots

maybe that is a bad example. I am just saying there could be other paths that offer a high resistance to ground, and that they need not be considered because of the high resistance. This is a generalization. if we look at the electric current flowing from the tube, through the cathode resistor toward ground, there are very few options. The current leaves the cathode resistor and hits ground and that is it.


I will concur again that in post five the only viable current path from the v1 cathode to the HT center tap is thru the chassis.

 

[2L man]

When suitaple build comes I must test forcing all the current leak to chassis thru input jack and connect PS zero to farthest chassis corner. I will use switch so it is easy to compare this to "correct" circuit.

I have temporarily installed a switch to two push pull amps and "split grounding" did hum more.

 

[gabasa]

The 4x10 I built last year (see my signature below), had a hum issue. It was completely resolved by changing the grounding scheme to not allow return current to flow through the input jacks and chassis. The difference was startling.

I think I’ve built 8 amps this way since I learned this, and modded 6. In my limited experience, it makes a difference for the better, but hey, who’s going to argue with Aiken, Blencowe and 2L man?

 

[gabasa]

The reason I started this thread was to test my own knowledge through questioning.

@peteb and @2L man, as always, you’ve been very helpful.

 

[peteb]

The 4x10 I built last year (see my signature below), had a hum issue. It was completely resolved by changing the grounding scheme to not allow return current to flow through the input jacks and chassis. The difference was startling.

that is very interesting to hear.

what is your suspicion?

the problem was that the chassis was the return path for ground? Or that the ground path went through the input Jack? Or both?

 

[gabasa]

that is very interesting to hear.

what is your suspicion?

the problem was that the chassis was the return path for ground? Or that the ground path went through the input Jack? Or both?

Probably both, because EMFs from ripple currents can induce noise anywhere in the circuit from transformer action. Because the input jacks are the most sensitive part of the circuit in terms of noise, my gut tells me that most of the difference was there.

That amp, with the volume on 12, has absolutely zero hum now; there’s not even the tiniest amount of low-level hum. Only a bit of preamp tube hiss when really loud, from V1.

That’s why, if you take Merlin’s suggestion to only ground the circuit at the input jacks, the HT CT must be lifted off the chassis. A ground loop will inevitably occur through the input jacks if a connection to the chassis is made at both ends. Here’s how I do it:

That terminal strip is isolated from the chassis. Nothing touches the chassis other than the input jack grounds and the earth safety, which is how I follow Blencowe/Aiken/2L man’s instructions. Even the speaker jack is isolated and has its own wire that goes to the terminal strip.

 

[screefer]

Here’s how I do it:

need link?

 

[gabasa]

need link?

I tried explaining it in the photo I posted (post #28), as well in the text beneath the photo.

I actually once tried to show more detail in a previous thread, but at the time, it didn't generate much interest. Here's a link if you want to have a look:

Another Amp Re-Grounded

I've had good success grounding my amps lately by using a lifted terminal strip and rewiring to make sure that no current can return to the PT via the chassis. This has been based on learning I have done with the guidance of @2L man. I just did this again, as part of an overall upgrade to my old...

www.tdpri.com

 

[peteb]

I think I’ve built 8 amps this way since I learned this, and modded 6. In my limited experience, it makes a difference for the better, but hey, who’s going to argue with Aiken, Blencowe and 2L man?

I am confused.

do these three experts approve or not approve of chassis current?

I have one vintage amp with chassis return current that is dead quiet.

do these three experts approve of grounding at the input Jack or not?

I thought 2L man brought up inductance across the Jack?


I do see now that even if the negative side of the input Jack and the entire ground bus remain at zero volts, there is still a small amount of AC ripple current going through, from one end to the other, the negative side of the input Jack.


if inductance is happening across the input Jack, why would it be a practice to ground there? I see that there is the possibility of many pre amp tubes, like the 5F6-A, all having their current going through the negative side of one input Jack.


why did people start grounding on the input Jack, of all the many places to ground?

Gabasa, which layout on which post best matches your best practices?

And, how did the layout on post 7 do?

thank you


who’s advice does your best grounding scheme most closely match?

 

[gabasa]

Hey @peteb, this is all to the best of my knowledge, I'll do my best to answer.

do these three experts approve or not approve of chassis current?

They do not approve of chassis current. They all state that, other than the safety earth connection, the circuit can only connect to the chassis at one point, and that the input jacks are the recommended spot to do this. Current cannot flow through the chassis under normal conditions when this is done. I've seen people argue this, but think about it; if current flows into the chassis, how can it get back to the HT CT if neither the HT CT nor anything else is physically connected to the chassis? It's impossible without a single, physical connection anywhere else.

I have one vintage amp with chassis return current that is dead quiet.

Vintage amps usually have several grounds randomly soldered directly to the chassis or to a brass plate. Any current that flows through those connections diverts current away from the input jacks. Less current through the input jacks means less noise in the most sensitive spot of the amp, and that's why your vintage amp is quieter. However, most amps that are "dead quiet" have some amount of perceptible hum when the volume is completely cranked up, and even this is eliminated with a better grounding scheme.

do these three experts approve of grounding at the input Jack or not?

Yes, all three recommend it, and non-insulated jacks are fine. It can only be a safety-earth connection though and the amp must be wired so that no return current can flow through them.

I thought 2L man brought up inductance across the Jack?

Yes, he did. The way I see it is that in order for ripple current to exist, there has to be some resistance, just because of Ohm's Law.

I do see now that even if the negative side of the input Jack and the entire ground bus remain at zero volts, there is still a small amount of AC ripple current going through, from one end to the other, the negative side of the input Jack.

Correct, and then from the input jack negative back to the HT CT. In my post #5, all the current from V1 and V2 flows back to the PT via the input jacks and chassis, which is exactly what you want to avoid. Blencowe illustrates it really well and says that ground is, "not some electrical black hole into which current disappears never to be seen again." Once those electrons hit the preamp ground bus in post #5, which is not the preferred setup, they will still travel all the way back to the HT CT, via the input jacks and chassis, to complete the loop. Not good!

if inductance is happening across the input Jack, why would it be a practice to ground there? I see that there is the possibility of many pre amp tubes, like the 5F6-A, all having their current going through the negative side of one input Jack.

Yes, and it's even worse in blackface builds that have more preamp tubes. I think that it's done because people rarely spend time trying to visualize the flow of return current to see what's actually happening in their amps. 2L man has been shouting at us all to change this practice for years, lol. Connecting the input jacks to safety earth is a great practice. Using it in a manner that allows current to return to the HT CT via the input jacks is not a good practice.

why did people start grounding on the input Jack, of all the many places to ground?

Connecting the input jacks to safety earth is a good practice that has been done since the '40s. Setting up the ground bus by splitting it in a manner that direct all preamp tube current flow through the input jacks is a more modern practice. My answer to your question is that forum mentality makes a lot of people approach things the same way without questioning it. No offense to the forums though, I love this place. Look up "5E3 chassis" or "5E3 build" (or anything similar) in Google Images and you'll see that the vast majority of current builds follow the method of routing all preamp current through the input jacks. People just copy one another, and it's exactly what I did for a long time.

Gabasa, which layout on which post best matches your best practices?

Post #7 is my best practice and I believe that 2L man would agree that it's far better than post #5. No current through the input jacks and no current through the chassis in post #7. I read his forum posts for a week last year and wouldn't stop until it made sense to me.

And, how did the layout on post 7 do?

Amazing. Layout #7 is what I started doing after reading Aiken's website, Blencow's chapter and 2L man's posts, and it's what I showed in my chassis photo above. When I redid my 4x10 Bassman and all the hum vanished, it was layout #7. I will never go back. That photo I posted only has two chassis connections: the input jacks and the safety earth. Nothing else, not even the output jack.

who’s advice does your best grounding scheme most closely match?

All the changes I made came from the advice of the three Titans we've been talking about this entire thread. I don't follow Blencowe's "improved bus ground" layout, but he recommends that for more complex or high-gain amps, which I don't really do. In a vintage, non-master volume amp, this is perfect. I need a beer!

 

[andrewRneumann]

I've read the whole thing. Your fingers must be burning, but can you please elaborate on this statement?

The way I see it is that in order for ripple current to exist, there has to be some resistance, just because of Ohm's Law.

 

[peteb]

do these three experts approve of grounding at the input Jack or not?

Yes, all three recommend it, and non-insulated jacks are fine. It can only be a safety-earth connection though and the amp must be wired so that no return current can flow through them.

we may be talking about two different grounds.

do these three guys recommend grounding the pre amp to the input jack as post 5 suggests?


thanks for the detailed response.

 

[gabasa]

I've read the whole thing. Your fingers must be burning, but can you please elaborate on this statement?

Lol, my day job often requires a lot of typing, I'm used to it. This is how I understand it, and if I'm wrong about anything, hopefully someone will come and correct me.

Here's the AC voltage coming into the amp, constantly swaying up and down:

Here's the DC, which pulsates because it's just rectified AC:

When filtered with smoothing capacitors, the pulsating DC turns into "ripple". The amps don't have pure DC, which would be a straight, horizontal line:

With the ripple DC pulsating up and down a little, the current will also pulsate up and down through any resistance because voltage and current are directly proportional. I=V/R. This is what is meant by ripple current.

When we talk about ripple through ground wires, we have to assume that in those ground wires, the voltage ripples up and down, causing the noise that gets into our guitar signal via induction. For any voltage to exist anywhere, there must be a resistance because V=IR, so if resistance = 0, then voltage = 0. This means that ripple in a ground wire would be impossible, unless there is a very small resistance within the wire itself. There is a resistance in the conductors we use because there is no perfect conductor, and I think that this is what is meant when Blencowe et al. speak of unavoidable series impedances within your ground loop.

However, I may be wrong about this, lol, and I'm happy to have someone correct my interpretations of this all, but this is how I understand it.

 

[peteb]

Yes, he did. The way I see it is that in order for ripple current to exist, there has to be some resistance, just because of Ohm's Law.

like I said when I now see it,

without a resistance, there can be no voltage drop,


however, there can still be current flow. Think flat, Stillwater, not a dropping river, there is still water flowing in and flowing out Of a lake.



the input Jack negative side and the ground bus may be at zero volts, but if current flows into the bus it must then also flow out of the bus, the Law of Current Flow.

 

[gabasa]

like I said when I now see it,

without a resistance, there can be no voltage drop,


however, there can still be current flow. Think flat, Stillwater, not a dropping river, there is still water flowing in and flowing out Of a lake.



the input Jack negative side and the ground bus may be at zero volts, but if current flows into the bus it must then also flow out of the bus, the Law of Current Flow.

Bingo.

 

[gabasa]

we may be talking about two different grounds.

do these three guys recommend grounding the pre amp to the input jack as post 5 suggests?


thanks for the detailed response.

peteb, I have to go run errands, I'll answer your question better later, but for now, it's a "yes."

 

[2L man]

we may be talking about two different grounds.

do these three guys recommend grounding the pre amp to the input jack as post 5 suggests?


thanks for the detailed response.

Pete please define what the "ground" is to you?

 

[andrewRneumann]

When we talk about ripple through ground wires, we have to assume that in those ground wires, the voltage ripples up and down, causing the noise that gets into our guitar signal via induction. For any voltage to exist anywhere, there must be a resistance because V=IR, so if resistance = 0, then voltage = 0. This means that ripple in a ground wire would be impossible, unless there is a very small resistance within the wire itself. There is a resistance in the conductors we use because there is no perfect conductor, and I think that this is what is meant when Blencowe et al. speak of unavoidable series impedances within your ground loop.

Ok, so you meant to say "ripple voltage" and not "ripple current" in your original statement?

in order for ripple current to exist, there has to be some resistance, just because of Ohm's Law.

I'm not sure you even need ripple from the power supply to get noise. If I was supplying clean DC, I still wouldn't use the chassis to return it or let that current concentrate in the sleeve terminal of the input jack. I'm not arguing with you about the best way to wire an amp; I'm a Blencolyte myself. But does Blencowe say the source of the noise in this context (ie preamp return current) is "ripple current" or just "current" (DC or otherwise)? Because if it's just "ripple" that we are fighting, the solution can be as simple as more filtering and we wouldn't have to worry about ground paths.

And just to keep you honest, do you have a reference that backs up this statement?

noise gets into our guitar signal via induction

...because that's news to me my friend.