Question about hum and ground loops in original Fender AB763

[dai h.]

(these are some of my general (hobbyist level) musings) :

as the term "electronic circuits" implies, there are loops everywhere AC, DC or both (seems a good idea to habituate yourself to think in terms of loops)

the (probably overly simplified) essence:

V/R=I and basic transfomer action


V/R=I

current flows over an impedance, creates voltage drop. When part of that loop flows over ground, there is a voltage drop, so an error (not the ideal "zero nothing happening here" situation), so managing (so how do you connect grounds together so the circuit works like it should without problems) the various returns and connecting them in a way that makes sense is key (might be harder when things get more complex)

(probably familiar?) example:

the 1 ohm resistor on the cathode. Typically on a power tube, depends on the tube but maybe 30-40-ish milliamps static (no signal) state. And of course if you've kept monitoring while you played the amp, you notice more voltage drop (so there can be more change over the ground). (The 1 ohm resistance makes things "more obvious", but since ground wires, traces, etc. have finite impedance so there are still things happening, and it's not "just zero".)


basic transformer action (more a layout thing I suppose) :
as mentioned, there are loops everywhere. If you have a loop on top or inside each other (and you want that) then good, but if not bad. So use basic techniques like distance (move it away), keep loop areas small (less inductance, less radiation), turn one loop so (orthogonal? right angle sort of relationship--example on PCBs many times there is a jumper over a trace at a strict right angle. Both are part of different loops, so if you can't avoid them being close cross them at strict right angles to avoid interaction), also shielding (shielding doesn't have to be a shielded cable, you can obtain the effect by laying wire on the chassis--if you have a corner available, you get both side and bottom).

(maybe familiar (to a guitarist)) examples:

-if you get close to the amp with your guitar (pickup/s) to the output transformer and power transfomer some weird stuff will happen, but move away and back to normal

-think about a 12A_7 (assuming AC heaters) which has a noise sensitive input (grid). How did the tube designers deal with the noise source (AC heaters = hum) going into the tube? (I think) it's the relative angle thing. (Basically) the heater goes in, makes a sharp 180 degree turn on goes back out. The cathode is around the heaters, and the grid winding is around the cathode (should be pretty close?). I don't think the grid winding can be perfectly a "T" or cross since it has to run lengthwise (so needs a bit of angle), but generally close? The connections from the tube pins (inside the tube) and the placement of those "L" shaped wire things I would guess are very deliberately placed and worked out (again distance, relative angles at work).

-when you get less buzz when touching the guitar ground (bridge, jack, whatever), I think this is shielding effect at play. Your body is sort of a ground plane under the guitar electronics (and maybe the body to guitar circuit C is more effective because the circuitry is higher impedance?).


like loudthud mentioned with the Fender and relatively low gain, you can probably be more lax with things if gain (and bandwidth?) is lower(narrower) (and maybe if currents are low as in a battery operated efx pedal?). If you "do" high gain I would guess you'd probably have to execute things more technically correctly out of necessity.

[dai h.]

Here is something I found (a passage from an article stumbled upon looking for basic info on op amps) that I think presents a really nice succinct overview of the fundamental issues:

The culprit is current

When planning grounding schemes, you need to think about two quantities: current and impedance. Current has to flow in a circle. Despite whatever path current takes into a circuit, current has to flow through ground to get back to its source. As Brokaw states, “Think where the currents will flow.”

Ground paths and connections always have some finite impedance. Current flowing through this impedance causes voltage drops. These voltage drops, which are the differences between the ultimate reference ground and a local circuit's operating ground, are the crux of the problem (see box, “The origins of noise”). The ideal role of ground is to act as the reference point for all other signals in the system; the practical role is that ground carries current.

Because it's impossible to remove the current or make the impedance zero, your job becomes tightly controlling the current flow and minimizing the impedance. If you can identify and control your circuit's largest currents, you've dealt with the largest potential polluters of ground. Ground planes are the best approach when dealing with high-speed signals, but not for precision circuits with 16-bit or greater precision.

source: https://www.edn.com/edn-access-11-23...-demons-at-ba/

[mikepukmel]

There are a couple of subtle things that Fender did that made hum a non-issue. First is the relatively low gain. Next was the general layout with the power supply and output stage on one end of the chassis and the preamp on the other end. You will usually notice a black ground wire comes out of the dog house and grounds in the preamp and another grounds near the power amp. This is important to keep hum out of the preamp.

Many people have tried to "improve" the Fender grounding scheme. Most failed miserably because they didn't understand how ground works or they tried to implement a star ground.

I did try hard to mimic the original layout, until it came to grounding, then I got completely flustered. I downloaded a whole bunch of photos of original chassis, as many as I could find. I was struck by how clean the AB763 wiring looked. First problem i hit wasn't huge but the brass plate I made was too thick, so the nuts on the pots just barely grabbed. I bought the best pots I could get without going to those wild expensive mil spec pots. But seems the threaded section is shorter, or maybe the plate is thinner. maybe both.

[mikepukmel]

So, finally (thank God) got some time this afternoon, after hacking up the lawn more and finally getting the entire lawn mowed:
- Removed the filter caps I put in based on some "over my engineering pay level" info.
- Removed the second trial attempt at the grounding bar. This one had a mess of wires attached to it, and one final ground wire to the chassis under the far side of the power supply.
- Took a look under the dog house.

Few more questions, if you all don't mind me torturing you a bit:

- I put a few lugs down on the chassis and hit the edge with a bit of solder with the big iron, then screwed down the lug to the chassis with a self tapper. Is it OK to have the center tap wired in one spot, and the 2 wires from the filter caps on another lug about 2" away, or do they all 3 need to be on the same lug?

- I have 2 "mid" pots in there, going to remove those and put back to stock trem, even though I won't use the trem. So, is it OK to solder a wire across the back of all pots and ground that separately, for noise not electrical? I mean, I won't use the back of teh pots as a ground point.

[mikepukmel]

Can I ground the heater reference to a spot right under the light socket, or does that need to get sent to the same lug as the 2 or 3 other "dirty" grounds?

[mikepukmel]

This is the ugly after removing the second attempt at the grounding bar (after removing the brass plate) before going in and doing a nicer job based on some of our build photos.

[loudthud]

The HV center tap is the noisiest ground wire so it's better to keep it off the chassis if possible. I would extend it into the dog house, then run a wire from the first filter to ground. Your picture confuses me. I see two red wires grounded on opposite corners of the PT, but the black stripe seems odd. I like the little bias board.

As for grounding the heater center tap, it doesn't make a whole lot of difference until... you get everthing else perfect. Then just experiment.

The main thing that makes Fender's layout work is the quiet ground for the preamp. Don't be tempted to hook grounds together in the dog house. Keep the preamp filter ground separate and connect it to chassis or brass strip near preamp. One place there is a major conflict is the Reverb driver. B+ comes from the screen node but ground wants to be in the preamp. I have no guaranteed solutions here.

[mikepukmel]

The HV center tap is the noisiest ground wire so it's better to keep it off the chassis if possible. I would extend it into the dog house, then run a wire from the first filter to ground. Your picture confuses me. I see two red wires grounded on opposite corners of the PT, but the black stripe seems odd. I like the little bias board.

As for grounding the heater center tap, it doesn't make a whole lot of difference until... you get everthing else perfect. Then just experiment.

The main thing that makes Fender's layout work is the quiet ground for the preamp. Don't be tempted to hook grounds together in the dog house. Keep the preamp filter ground separate and connect it to chassis or brass strip near preamp. One place there is a major conflict is the Reverb driver. B+ comes from the screen node but ground wants to be in the preamp. I have no guaranteed solutions here.

Thanks Loud! The red wire tied to teh top right corner is actually red with a yellow stripe, its a Hammond PT, not sure that makes a difference. There are no black stripe, they got gunk all over the wires when they manufactured it. Its actually 2 solid red (HV out) red + yellow stripe (HV CT). The two solid reds go to the rectifier tube, its the angle that makes it look like one is grounded. The black wire coming out of the PT is tied to the chasses on the bottom left but not on the PT screw, its soldered to the chassis close to there.

Thanks re tips on the dog house. This version I got the layout from some place on the web a while back, don't remember where. The 2 x 22uf reservior caps, the negative from that is routed back into the chassis, and its hard to see from my stinky photo, but its tied to the chassis under where the resistor comes out of the bias pot.

Thanks re tips and suggestions. Oh I don't feel so bad, when I read the schematics and learned a little bit, the reverb wiring is exactly what I was completely confused about!!!! I thought, there must be a reason for this. Fed from early in the HV, but should be a quiet "thing". Someone suggested here and there, that I could put another dropping resistor (someplace?) then another cap, and feed the reverb from that. Seeing what I ran into so far, I think I will try very hard to get this as close to the Fender photos as possible, with the exception of the grounding bar instead of the brass plate.

In some chassis Ive seen on this site, and some suggestions from other builders, they wire the far end of the grounding bar right up to the first input jack ground. I guess I can sort of see that, but couple of questions: why would they not tue the other end of the grounding bar to the chassis? And, even if the grounding bar is tied to the first input jack, the second input jack has another ground point. (Im guessing, no idea, that is why Marshall used those plastic box input jacks?

[mikepukmel]

My diagram skills are pretty stinky, but this is what I think I saw on some of the cool builds on this site. The fat red line is teh grounding bar, and thinner lines are what would have been tied to the chassis (either at the back of the pot, or from the circuit)
This is sort of close to what you talked about Loud Thud, except running a wire from the HV center tap up to the dog house.

Another question, how do you mount the other side of the resistors that are tied to teh back of the pot in the original layout? I have some old mounting lugs, some kind of hard insulated plastic, a screw on one end that can be bolted to the chassis, and a turret on the other. There are a few in the build above. But they are one heck of a pain to install after the fact, drilling and like that.

[mikepukmel]

Was talking to a good friend on the phone today, he's a real electronics guru like some of you guys, he did electronics for his career (retired now). he was working on one of his marshall amps, tracking down some noise/hum and found an issue in how wiring was routed from the rectifiers to the filter caps. So, I looked at old photos, the previous incantation (2 years old) and found that the dirty signal right off the tube rectifier output was routed way too close to the output tube wiring. I went back and compared it to some photos I have of original AB763, I screwed this one up royally. Also, all the original's I could find, they twisted the output transformer wiring to the output tubes and routed the wiring between the two output tubes, far away from that dirty input wire. Going to fix those two issues. Fingers crossed. Last cleanup is to remove the pile of leads crammed into the top of a few turrets, re-watched some clips of old soldering instruction videos, none of them put wires inside teh turrets, they all carefully wrap leads around the turrets about 3/4 way. This is cool stuff!

[mikepukmel]

Rewiring #39. (only 5 years and I got this far)

Got most of the odd mods removed, and almost all the rewiring getting ready to put the replacement ground bar in.. Decided to leave the mid pots in for now. A few of the turrets still have a bunch of leads jammed in the top, got a few fixed, will work on the bad ones as I go.

The second rewiring after I found that the brass plate would not fit, I got a big iron and soldered bus wire onto the chassis. What I had ended up with, probably rewiring #38 or so , Id describe like this: take a ground point and write the number down on a piece of paper. Then take a ground wire and write that on another piece of paper. Put them all in a bow, mix them up then pick two out at a time. Okay, maybe not quite that bad, but I got plenty of nice raspy hum.

As mentioned above, I melted the output transformer wiring so put some heat shrink on that, and the previous wiring, those wires were parallel and close to the heater and also output from the tube rectifier. I tried to fix that, but with the heat shrink on its hard to twist like the factory photos. But I did reroute the OT wiring between the 2 output tubes, and as close to 90 degrees from the heater wiring and far from the rectifier output wire.

Almost ready to "fire it up" tomorrow. With any luck. the smoke won't come out.

[mikepukmel]

Also, not to be rude, but I've read some of your other posts on the forum and you seem to be having some issues with your builds.

My advice would be this: Dial it back and build some simpler amps. It's not that a JTM45 is all that difficult, but it does have a fair deal of parts. Reverb amps can be very fussy, and I wouldn't recommend them until you've really mastered everything else. If you had issues with the 5E3, that's about as simple as you can get with a push-pull amp, I'd really suggest you put this other stuff aside and build some single-ended amps. AC4, 5C1, 5F2A, BF Champ, AX84 P1. Any of those. And although you read my posts it would probably sound like I would recommend good components. No. I'd buy the cheapest crap you can get and make those sound the best you can with that. Then start modifying, swapping, testing, testing, testing. Then, and only then invest in the big money components. Those will never sound good if you can't master the basics.

All my first amps were SE. None had reverb. All single channel. All used Hammond transformers, usually the universal SE type and whatever PT spec I could figure from the universal tube supplies. All used cheap caps, resistors and tubes. Once I had mastered them, then I started tweaking. And I'll say, you hit a limit pretty fast with a 2-tube SE amp. But see what you can do with. You should be able to make it quiet and sound reasonably good. Then go back to your 5E3 and perfect that. Next I'd say would be the JTM45. And lastly, this Deluxe Reverb.

Just some info, I built this franken amp 5 or 6 years ago, it was my first attempt, and the photos show it.
I added various changes over a few years, when I had time, to fix a hum problem, that is still there, somewhat, although better than the first few tries. It looks nothing like it did when I got the first version working. I had plenty of wiring/lead dress problems. I think at least some of those are gone now. The most recent incantation had many grounding problems.

I have an old scope, reading up on how to use it (haven't used since physics class back in school, and there were no 460v dc in those circuits).

[mikepukmel]

Thanks for all the articles, suggestions, tips. I got the grounding bar in, and fixed some of the component wiring to turrets. Found a few component mistakes, Rewired all the grounds except one (hard to get that last one). Fired up the amp, and it did not release any smoke are sparks, which is nice. I could only play for 10 or 15 minutes with low volume, but the raspy hum is gone. The only guitar I have has 3 single coil pickups, so its very bright and picks up noise from every dang thing, but when I moved around and nulled out the pickups-to-amp, it sounds great. I think it sounds a lot better than with all the hacky wiring. Tone is nice. There is a little bit of him when I crank the reverb way up but I don't use it like that anyway. Will move to the basement during reasonable hours (its after 1am here now), and crank it up a bit and see if it still holds up.
Thanks!

[mikepukmel]

Here is an article that addresses some of your grounding questions: http://www.geofex.com/article_folder...nd/stargnd.htm

I read and re-read a few times, thanks. Where I get stuck, I don't have the engineering chops to make big changes to e.g. a fender style build. So I can get up to a certain point, then not sure where to route some ground wire. I got some photos of you guys cool builds and mimicked that, and its working a lot better. Id like to learn enough to be able to build, e.g. a marshall JTM45 circuit and use information in the article directly.

[mikepukmel]

For all the newbie builders:

its not the way I had hoped it would work out, but, starting about 4 years ago: build something bad, try to figure out what I did wrong, fix that, and so on. So far, luckily, didn't let the smoke out.


Below image is after a few fixes. The white/green stripe wire coming from the rectifier tube, just before this photo, was very close almost parallel to the output transformer brown wire, and heater wire pair, and all 3 were parallel for some distance. I have since shortened the white/green strip wire coming from the rectifier, its now almost straight to the turret board and less excess wire. (Excess wire is a killer in these amps, I learned)

I melted the crap out of the output transformer wiring, so put some heat shrink on. I was not able to do a very neat twist like the original, but re-routed and twisted as best I could. This one change got rid of the nasty sounding hum. It wasn't a smooth sine sounding hum.

Fix:
1) new wire from the rectifier tube, made it shorter than in teh photo below, and routed it as straight to the turret board as possible. its about 90 degrees from the heater tube wiring and also far from the output trans wiring
2) twisted the output transformer wiring, routed them as close to original as I could, flat against the chassis, and between the two output tubes.
3) ran the output transformer wiring around the bottom of the tube, so its far from the heater and rectifier tube wire.

I had read on MEF many places that "lead dress" is critical, but took some mistakes to realize what that means in more detail.

The factory amp wiring, at least this era, look amazingly clean. I think my heater wiring should be flat on the chassis as it rounds the power transformer. Also, in general, my wiring in many places is too long, the wire meanders instead of going directly to where its supposed to go.

I wish I had a way to record the hum before, and how the amp sound now, for comparison.





This was taken at the speaker terminals, years ago, not long after I build the first haywire mess, when I started to diagnose the hum. I don't remember the specs, whether the spikes were 120 or 60. I imagine those sharp spikes were caused by signal getting induced into the output tube wiring.

The other thing that made a difference (thanks to the MEF gurus) was switching the output tubes around. One way, there is noticeably more hum. This is a nicer sounding "sine" hum, but still there. Switch the tubes around and that drops significantly. These are supposed to be matched and burned in tubes. My cynicism tells me that they probably just threw the tube in the box without doing that.