Cathode follower without elevated heater voltage

[Chuck H]

The 3.3M resistor (At least that's the value on the schematic) that's connected to the Grid of V1, is 1M in the amp. I know it can't be the cause of the amp's problem, 'cus it's only on one channel, but what effect would this have?

That resistor is part of a short feedback loop. Probably to stabilize the circuit? Such loops appear elsewhere in the amp too I notice. Not sure why they're there since gain is already handled in other ways and reduction of harmonic distortions probably wasn't the goal.?. I don't think you'd notice much difference between 3.3M and 1M for that circuit. The 3.3M would give a little more gain. Which probably isn't necessary.

[Dark Mavis]

Okay, so here are idle DC voltages for the preamp tubes and filter caps. Grid voltages were too small to measure accurately. I noticed there was a lot of popping and scratching when I was trying to measure those. I don't mean just when the probe first made contact, like you'd expect. I mean it would randomly set off a period of scratching sounds until I measured somewhere else or gave the chassis a top with the probe. I think It might be a good idea to pull the tubes and clean the contacts, and spray some contact cleaner into the pots.
Here's the schematic, with the voltages on it.

I think next, I'm going to roughly reassemble my guitar, so I can record a proper demonstration of the problem

[Dark Mavis]

There seems to be a problem. And the problem is that the amp is working now. I think, when I tested it with that electric whisk thing that what I heard wasn't the old problem. It was just an amp being loud. It was so brief and so loud, I couldn't really tell. I think it's been fixed since I rewired it. So all along, it was either the filter caps, or the incorrectly wired cathode followers in the preamp.
Basically, this amp has been sitting there, in working order, since September, without me knowing it

[Chuck H]

There seems to be a problem. And the problem is that the amp is working now. I think, when I tested it with that electric whisk thing that what I heard wasn't the old problem. It was just an amp being loud. It was so brief and so loud, I couldn't really tell. I think it's been fixed since I rewired it. So all along, it was either the filter caps, or the incorrectly wired cathode followers in the preamp.
Basically, this amp has been sitting there, in working order, since September, without me knowing it

The static lingering after probe testing the grids is still anomalous behavior. It can happen for a couple of reasons. It might be dirty contacts? Certainly clean the contacts.

When you tough a grid with the probe there's going to be a signal. A noise signal, but still an audible. If there are DC voltage conditions in the amp that linger, like capacitors taking a charge until some other condition discharges them, this might also cause noise in an unstable circuit. The effect is very nebulous though I've experienced a couple of times. By nebulous I mean that I'm spit balling the potential cause. In an amp like that, with some age grime and so much lead wire and vapor deposited flux, etc. I wouldn't be surprised if there are slightly conductive paths on some surfaces. This might also hold a static charge that disturbs the audio until discharged by some other circuit condition. Carefully cleaning with a soft brush and then some alcohol might help. Though in that chassis it seems like a monumental chore without knowing if it'll actually fix anything.

[Dark Mavis]

The static lingering after probe testing the grids is still anomalous behavior. It can happen for a couple of reasons. It might be dirty contacts? Certainly clean the contacts.

When you tough a grid with the probe there's going to be a signal. A noise signal, but still an audible. If there are DC voltage conditions in the amp that linger, like capacitors taking a charge until some other condition discharges them, this might also cause noise in an unstable circuit. The effect is very nebulous though I've experienced a couple of times. By nebulous I mean that I'm spit balling the potential cause. In an amp like that, with some age grime and so much lead wire and vapor deposited flux, etc. I wouldn't be surprised if there are slightly conductive paths on some surfaces. This might also hold a static charge that disturbs the audio until discharged by some other circuit condition. Carefully cleaning with a soft brush and then some alcohol might help. Though in that chassis it seems like a monumental chore without knowing if it'll actually fix anything.

I thought about taking the board out and re-doing all the solder connections, then giving it all a good isopropyl wash, but the tube sockets are mounted from the outside. I'd have to desolder every pin. I can at least take the pots out so I can flip the board over. I gotta replace one pot anyways cus the shaft is broken.
The important thing is that the problem it's had for almost 20 years is fixed. I dunno what fixed it, but I'm still calling it a win. It was either the lead dress, the mis-wired tubes, or the filter caps. Though Apparntly the caps had all been subbed at some point, so maybe not.
I wish I could remember exactly how those tubes were wired before. It's really hard to make out in the photo

[Dark Mavis]

Okay, so it turns out the problem is not fixed. But I can turn the amp up way, way louder before it occurs than before. So I'm thinking it's a lead dress/grounding issue. I think I need to look at where all the ground points are, and create a more optimal ground scheme.

[Chuck H]

Okay, so it turns out the problem is not fixed. But I can turn the amp up way, way louder before it occurs than before. So I'm thinking it's a lead dress/grounding issue. I think I need to look at where all the ground points are, and create a more optimal ground scheme.

There are a lot of threads here that are basically grounding tutorials. That's how I learned how to do it. And in all immodesty I've never had a problem with the ground scheme in any of my builds***. We're here as always if you have questions.

*** There are always exceptions when it comes to reverb circuits I've had to rework grounds on a couple of those.

[Dark Mavis]

There are a lot of threads here that are basically grounding tutorials. That's how I learned how to do it. And in all immodesty I've never had a problem with the ground scheme in any of my builds***. We're here as always if you have questions.

*** There are always exceptions when it comes to reverb circuits I've had to rework grounds on a couple of those.

Have you ever started reading about something and wished you hadn't started? There seems to be an awful lot of people confidently saying that the things that other people are confidently saying, are incorrect. I guess if it really mattered, there would only be the right way, and then all the ways that don't work at all. But when the very first two articles you read, completely contradict each other on several things, you know you're headed down the rabbit hole. I read an article abut it on Randall Aiken's page, then an equally long article elsewhere that basically said the reverse.
So what I'm gonna do is just this: https://el34world.com/charts/grounds.htm Cus this one has a picture. I'm gonna put a terminal lug by the input jack for a ground though, even though they're not isolated jacks. Occasionally I get static that a tap on the jack plug can get rid of, so I don't trust the contact those sockets are making. I guess a ground loop won't be an issue if the wire is short. I know all my power tube cathodes are grounded to the chassis at each tube's socket. I don't think the pots are grounded at all, beyond their physical connection to the chassis, and all the filter caps except the last one, share the same ground.
What about the filter cap that's connected to the driver tube plates (and to the reverb transformer)? Should that one be grounded at the input jack end, or by the power trans?
I got a bunch of lock washers and nylock nuts, and I'm going to clean up the chassis and make all new ground points. Might as well try to clean up the tube sockets while I'm at it and see if I can get rid of the snap, crackle and pop. I have no tools for this though. I think I'll just spray contact cleaner on a tube's pins and push it in and out a few times, then maybe rinse with isopropyl.
I'll also take the opportunity to replace all the plate resistors, and restore the values of those 470k ones to 220 or maybe 100.
More contact cleaner in the pots, too, when I have them out. With them in place, I dunno if I was getting it on the actual track.
What about the filament's false center tap dealie? Right now, each is grounded where the resistors are bolted into place. Where should I ground those?

[Chuck H]

What about the filter cap that's connected to the driver tube plates (and to the reverb transformer)? Should that one be grounded at the input jack end, or by the power trans?

Ah! The reverb grounds. The reverb power supply is often the same as that for the screens in most classic designs. It seems your amp isn't like that. Here's what I do for amps like yours that have a "s" ton of circuitry going on. It's technically wrong but fundamentally right and it makes building easier and helps keep things tidy in the chassis. I use three star ground points. One by the input for the preamp, one at the power supply end and one at the mid point for, ah, well, anything that's NOT the preamp or power amp/power supply. Lemme splain it.

The preamp grounds would be anything that pertains to the dry signal chain and the power supply filter cap grounds that power those circuits only. That's easy. The power amp and power supply is the power tube plate, screen grid and cathode circuits and the power supply filters cap grounds for those circuits. That's easy too. So where does the PI belong? Is it preamp or power amp? Well it's mostly preamp, but sorta power amp idn'it. And in your case the reverb shares a power supply with it. It seems Dan thinks a little like I do here because the reverb isn't exactly part of the preamp, it's an effect. Then there's the presence control and any post PI master volume circuits.?. They're really part of the power amp circuit even though the controls are on the front panel. So I use a third ground point sorta near the power amp end that's for the reverb power transformer and driver tube cathode grounds (not the reverb signal grounds, I'll get to that), the PI grounds, any tremolo circuit, the presence control and any associated filters for those circuits. Now, the reverb signal ground would include the grid leak resistor for the driver tube because that's actually still part of the preamp. As is the reverb tank output and recovery tube. So ground those with the preamp. There's a chance your reverb drive power filter is different from your reverb recovery filter. Use your best judgement here and remember that you want to ground power supply filters with their circuits. And...

Daisy chains are often bad. Those places on the board where one eyelet is a ground point for three different things and then one lead goes to ground. When you do this ALL those grounds share the resistance of the wire. Wait? WHAT! The resistance of wire? Yes, hypothetical you. Wire has a tiny resistance and in something that is very high gain, like a guitar amp, it can matter a little. Get a bunch of peripheral circuits together with other leads flying all over the place and it can matter more. Worse is when you have more than one of these multi ground points and then sorta tack a wire between them (often under the board because it looks neat) and then use one lead to the ground point for the whole array. Do not do this. Now you CAN get away with clumping some grounds together on the same eyelet IF you know how. You can share a ground eyelet with a single lead for individual circuits. Say, a tone stack with an associated volume control where the signal leaves one tube plate and terminates at another tubes grid. The tone stack, volume control and any associated voltage divider or bleed cap grounds can share an eyelet. You might even get away with having the sending triodes cathode grounded there or any grid to ground resistor for the following tubes grid. DO NOT clump or daisy chain grounds of like phase from different preamp circuits and have them share a lead to the ground point. If that's confusing you're not alone. When in doubt just run an individual lead to the ground point.

As to pot cases I am out on a limb compared to most vintage amp specialists here because there is no guru mojo or lore associated with my logic on the matter. So here you go. Don't bother grounding the pot cases and don't use the pot cases as a ground point. There. I said it. The pot cases are not connected to any of the lugs and the resistive elements inside are ALREADY IN THE CHASSIS LIKE EVERY OTHER RESISTOR. No one is taking steps to add a grounded case to the rest of the resistors in the amp, so why the pots? Nope. If a pot lug is schematically grounded then run a lead to the circuit ground. Tack soldering a long copper wire across the backs of the pots and using it as a ground buss is a fundamentally crappy idea. Pot to chassis contact can corrode. You could end up with your whole preamp grounded up chassis by the damn power switch. There's a tendency to use the buss inappropriately and share grounds that should be otherwise be separate (like the presence ground, which is a power amp ground, and the preamp grounds). You should literally ignore any advice that suggests the pot cases should be grounded together with a buss wire. Like the Hoffman drawing. With all the pots soldered together AND the ground bar at the filter caps there is a ground point IN the chassis near the power amp end and multiple ground points along the front of the chassis. Including the unisolated input jack. There should be ONE ground point for the preamp and it should be nearer the input end of the chassis.

When running a lot of individual leads ground points can get crowded fast. I use these little five lug terminal strips that I use to get at Radio Shack, but now have to buy on line. The center lug on these is grounded via being bolted to the chassis and I just solder a solid copper wire across the lug rivets so I have five fat lugs I can solder grounds to.

Sometimes you don't get a choice on where to ground something. As with non isolated jacks of any kind. In this case the - jack terminal is grounded at the chassis. You still want to run an individual ground lead to the ground point, but only use that - jack terminal as a ground lug for cable shields and resistors that connect the jacks + terminal directly to ground.

Shielded leads should only be grounded at one end.

Reverb cables often break this rule and reverbs often hum like they don't know the words. With the exception of the input grid the reverb drive circuit and all the shaking springs is a FUNCTION circuit and not part of the signal chain or preamp. The reverb recovery circuit IS in the signal chain and is the tanks output transducer and the tube stage that amplifies it up to a useful level. Old school reverb circuits tend to clump grounds together for all this and they often (though not always) hum. When I repair I do the best I can with what's in the amp. When I build I always get a tank with an isolated output jack and I even isolate the RCA jack on the amp chassis so I can choose my ground points for the recovery circuit which IS part of the preamp.

The green lead on the power cord gets securely bolted to the chassis near where the cord enters the amp. This is a safety ground and not a circuit ground. The Hoffman layout is wrong on this matter. It's not a problem if you want to do it this way, just know that it's a different function. A safety measure. So be certain it's secure.

Ok, I dissed the Hoffman drawing three times here. It's not "bad" and it's how most vintage amps were/are done and it "works". It can be improved upon in practice unless you're cloning a design.

[Dark Mavis]

Ah! The reverb grounds. The reverb power supply is often the same as that for the screens in most classic designs. It seems your amp isn't like that. Here's what I do for amps like yours that have a "s" ton of circuitry going on. It's technically wrong but fundamentally right and it makes building easier and helps keep things tidy in the chassis. I use three star ground points. One by the input for the preamp, one at the power supply end and one at the mid point for, ah, well, anything that's NOT the preamp or power amp/power supply. Lemme splain it.

The preamp grounds would be anything that pertains to the dry signal chain and the power supply filter cap grounds that power those circuits only. That's easy. The power amp and power supply is the power tube plate, screen grid and cathode circuits and the power supply filters cap grounds for those circuits. That's easy too. So where does the PI belong? Is it preamp or power amp? Well it's mostly preamp, but sorta power amp idn'it. And in your case the reverb shares a power supply with it. It seems Dan thinks a little like I do here because the reverb isn't exactly part of the preamp, it's an effect. Then there's the presence control and any post PI master volume circuits.?. They're really part of the power amp circuit even though the controls are on the front panel. So I use a third ground point sorta near the power amp end that's for the reverb power transformer and driver tube cathode grounds (not the reverb signal grounds, I'll get to that), the PI grounds, any tremolo circuit, the presence control and any associated filters for those circuits. Now, the reverb signal ground would include the grid leak resistor for the driver tube because that's actually still part of the preamp. As is the reverb tank output and recovery tube. So ground those with the preamp. There's a chance your reverb drive power filter is different from your reverb recovery filter. Use your best judgement here and remember that you want to ground power supply filters with their circuits. And...

Daisy chains are often bad. Those places on the board where one eyelet is a ground point for three different things and then one lead goes to ground. When you do this ALL those grounds share the resistance of the wire. Wait? WHAT! The resistance of wire? Yes, hypothetical you. Wire has a tiny resistance and in something that is very high gain, like a guitar amp, it can matter a little. Get a bunch of peripheral circuits together with other leads flying all over the place and it can matter more. Worse is when you have more than one of these multi ground points and then sorta tack a wire between them (often under the board because it looks neat) and then use one lead to the ground point for the whole array. Do not do this. Now you CAN get away with clumping some grounds together on the same eyelet IF you know how. You can share a ground eyelet with a single lead for individual circuits. Say, a tone stack with an associated volume control where the signal leaves one tube plate and terminates at another tubes grid. The tone stack, volume control and any associated voltage divider or bleed cap grounds can share an eyelet. You might even get away with having the sending triodes cathode grounded there or any grid to ground resistor for the following tubes grid. DO NOT clump or daisy chain grounds of like phase from different preamp circuits and have them share a lead to the ground point. If that's confusing you're not alone. When in doubt just run an individual lead to the ground point.

As to pot cases I am out on a limb compared to most vintage amp specialists here because there is no guru mojo or lore associated with my logic on the matter. So here you go. Don't bother grounding the pot cases and don't use the pot cases as a ground point. There. I said it. The pot cases are not connected to any of the lugs and the resistive elements inside are ALREADY IN THE CHASSIS LIKE EVERY OTHER RESISTOR. No one is taking steps to add a grounded case to the rest of the resistors in the amp, so why the pots? Nope. If a pot lug is schematically grounded then run a lead to the circuit ground. Tack soldering a long copper wire across the backs of the pots and using it as a ground buss is a fundamentally crappy idea. Pot to chassis contact can corrode. You could end up with your whole preamp grounded up chassis by the damn power switch. There's a tendency to use the buss inappropriately and share grounds that should be otherwise be separate (like the presence ground, which is a power amp ground, and the preamp grounds). You should literally ignore any advice that suggests the pot cases should be grounded together with a buss wire. Like the Hoffman drawing. With all the pots soldered together AND the ground bar at the filter caps there is a ground point IN the chassis near the power amp end and multiple ground points along the front of the chassis. Including the unisolated input jack. There should be ONE ground point for the preamp and it should be nearer the input end of the chassis.

When running a lot of individual leads ground points can get crowded fast. I use these little five lug terminal strips that I use to get at Radio Shack, but now have to buy on line. The center lug on these is grounded via being bolted to the chassis and I just solder a solid copper wire across the lug rivets so I have five fat lugs I can solder grounds to.

Sometimes you don't get a choice on where to ground something. As with non isolated jacks of any kind. In this case the - jack terminal is grounded at the chassis. You still want to run an individual ground lead to the ground point, but only use that - jack terminal as a ground lug for cable shields and resistors that connect the jacks + terminal directly to ground.

Shielded leads should only be grounded at one end.

Reverb cables often break this rule and reverbs often hum like they don't know the words. With the exception of the input grid the reverb drive circuit and all the shaking springs is a FUNCTION circuit and not part of the signal chain or preamp. The reverb recovery circuit IS in the signal chain and is the tanks output transducer and the tube stage that amplifies it up to a useful level. Old school reverb circuits tend to clump grounds together for all this and they often (though not always) hum. When I repair I do the best I can with what's in the amp. When I build I always get a tank with an isolated output jack and I even isolate the RCA jack on the amp chassis so I can choose my ground points for the recovery circuit which IS part of the preamp.

The green lead on the power cord gets securely bolted to the chassis near where the cord enters the amp. This is a safety ground and not a circuit ground. The Hoffman layout is wrong on this matter. It's not a problem if you want to do it this way, just know that it's a different function. A safety measure. So be certain it's secure.

Ok, I dissed the Hoffman drawing three times here. It's not "bad" and it's how most vintage amps were/are done and it "works". It can be improved upon in practice unless you're cloning a design.

Then I think, what I need to do first is mark the current ground points (Not things like shielded wire grounds) on a photo or layout drawing of the amp, and label them. And then mark each ground on the schematic with the letter/number of the ground it's connected to. I think that will help me visualise what's going where, and it will also make it easier for someone to communicate where I went wrong, if I do.
I already have one ground point by the input jacks, one by the mains transformer, and one or two in the middle of the amp. Space is the problem. I don't really have room for terminal strips, so it's gotta be those little solder tags. I've never liked the idea of using a bunch of those with a single bolt. It seems like too many physical connections where corrosion could get in-between. I guess if I tin the whole tags and bolt them down securely enough, it will be fine. It's better than what's in there now. One ground point is just a bunch of tags bolted to the circuit board, so the electrical connection to ground is only through the head of the bolt. I dunno how critical this stuff is, but that doesn't seem ideal to me

[Dark Mavis]

So here's what's in the amp. The points marked #1, #2 and #3 are the existing grounds. #1 and #3 are tricky to get to without moving other stuff. #1 is under a transformer, and #3 is under the circuit board. So I'm gonna move them to A, B, and C. All the filter caps except the one over by the input jacks are grounded at #1
Now I'm gonna mark all the grounds on the schematic

[Dark Mavis]

Here's what's in the amp. There's a lot of stuff connected to the middle ground point, that probably shouldn't be. And I think the last 2 filter caps should be grounded at G1. The next one - the one providing B+3 - I assume that one should be on the middle ground, along with the driver tube cathodes and some other stuff. The power tube cathodes are each grounded at that tube's base, so I guess they should all be taken over to G3, as should the ground from the false center taps on the filament strings? The negative feedback from the speaker is grounded at the middle point. The ground from the reverb trans is grounded by being connected to the reverb jack.
There's also that 47k resistor to ground, that exists on the original schematic, but is not in the amp. The schematic seems to be from a slightly different model to the circuit board drawings in the manual. I dunno what effect its ommission would have, but it's a resistor to ground from the signal path, so I guess it's there to reduce output. Could too much gain be the cause of the amp's instability? Bear in mind, that since the re-wire, the amp can go way louder without breaking up than before.
Also: I recorded a demo of the original problem. I'll post that here when I upload it. Basically, the amp plays normally until you turn it up to a point where maybe hitting the strings harder, or hitting a chord instead of a single string, causes it to make a loud machine gun sort of sound.
Also: I noticed the mains transformer isn't bolted down tightly. I'm guessing that's why the amp is humming more.

[Dark Mavis]

Then there's the presence control and any post PI master volume circuits.?

Presence control is just adjustable negative feedback, right?

[Chuck H]

Presence control is just adjustable negative feedback, right?

Exactly. So it's a combination PI tail and power amp circuit. I usually ground it, and the PI tail at your G2.

[Helmholtz]

Presence control is just adjustable negative feedback, right?

Yes but frequency-dependent. It allows reducing NFB at higher frequencies and thus acts as an active treble control.