Restoring a 1964 Ampeg Reverberocket

[Schlumpfmeister]

Is there hum or is it hiss and buzz? Is it audible with the volume control turned all the way down?

Wide range spectrum starting at 100Hz, all the way up into the 7-8kHz, constant and not intermittent.

Turning volume or dimension up or down changes the signal somewhat, but not much.

I always have a shorted connector plugged in, otherwise the noise level would be much higher (seems a RR problem, the inputs are not grounded).

 

[TwoBear]

Here is the restoration thread for my recent $100 pawnshop find, a 1964 Ampeg Reverberocket, model R-12-Rb -- which I'm slowly starting to decipher and reconstruct between other projects. At first I was elated to find this model amp for so cheap, having serviced (and drooled over) a couple of Ampegs previously, but at this drier-lipped stage I must say that this is the worst case of an abused and neglected amp I have ever come across in my shop (and I've seen some doozies!).

The amp reeks for one thing. Between the dirt, gunge, bilge, blown caps, brittle wire, and sad attempts by former abusers to "maintain" the amp (something tells me Dymo tape labels on the control panel and a microwave-over power cord are not exactly 'vintage correct') it's hard to decide where to start. (One of my friends commented: it looks like it's been dunked in beer.)

Rescuing stray puppies would be so much easier.

Anyway, here's some BEFORE shots of the amp and my inital stages of R&D before I get down to more serious (and dirtier) work. I'm going to have a bunch of questions and decisions as I go along, so I hope the good folks here will check in from time to time to offer bits of advice.

R-12-RB -- BEFORE RESTORATION:

Front of cab, with original grill cloth and logo, non-original generic vinyl covering:

View attachment 585612

Rear view -- The Pros: Original reverb tank and footswitch, mostly original circuit (except 2 resistors), some original tubes that work, functioning transformers

The Cons: spewing electrolytics, nameless mystery power tubes (possibly 7591s, to be determined), grey microwavish AC-cord with broken ground pin & WTF "safety ground" soldered to a cap clamp, chassis unwelded on one end, and oddball Phillips hi-fi speaker.

View attachment 585613

Top of the amp: The silkscreening wore off over the years -- or under the beers -- so somebody helpfully tape-labelled the panel (making sure they were nice and straight!) -- oh, and replaced the original knobs with what you see here.

Pro: The components and handle are original and in no worse condition than any other amp of this vintage.

View attachment 585620

Schematic on the back panel (which, unlike the rest of the cab, has its original blue check tolex). The schematic is dry as a bone and crumbly, so I carefully removed it and am going to re-draw it (See Step 1, below).

View attachment 585621

I was just at Hoffman amps el34 world schematics page and there's a schematic that might be yours down the page at number 144. I couldn't grab the link but I don't think you have to be a member to look there.

 

[timfred]

Wide range spectrum starting at 100Hz, all the way up into the 7-8kHz, constant and not intermittent.

Turning volume or dimension up or down changes the signal somewhat, but not much.

I always have a shorted connector plugged in, otherwise the noise level would be much higher (seems a RR problem, the inputs are not grounded).

Oh wait did you ever try disconnecting the NFB?

 

[Schlumpfmeister]

Oh wait did you ever try disconnecting the NFB?

Just tried it by lifting a leg of the 5,6k resistor off the board. No improvement unfortunately.

Here a sound sample of the noise, everything fully cranked, then first turning volume down and up, then dimension. Taken with mobile phone recorder, 1ft distance from speaker.

20220108_072701.m4a

Shared with Dropbox

www.dropbox.com

 

[Schlumpfmeister]

Hm, no comments, strange.

I since have bought a digital scope and with this I was able to re-read the power supply and I think it is no longer suspicious: All AC readings are quite low (A being highest with 12v, further downstream the ripple is < 1v), and though there is some ripple and background noise, the difference is almost 80dB.

I went back to V3-1, and strange enough, the inputs from preamp (by way of cathode) and from reverb (by way of grid) are almost flat line, yet the output from V3-1 shows substancial ripple. Interestinlgy, above the plate resistor the supply power seems to have no substancial ripple again. I replaced the screen resistor (again) with no effect, it seems ok.

So far I have not had mentioned that some of the tubes were worn and I have replaced some of the worn 6SL7GT with Russian 6H9C. I have just realized that the visible (scope) hiss level out of V3-1 is much higher with the Russian tube than with a "real" but worn 6SL7. The audible difference isn't quite as big. I so war was under the assumption that the 6H9C is a perfect replacement for 6SL7, is it not? It almost seems that the amp oscillates with the Russian tube?! However, regardless of the tube used there is always a low frequency ripple on the output of V3-1, although neither the inputs nor the power supply show such effect!?

EDIT:

Output from V3-1 (pin 2) with Russian 6H9C tube:

Same with old tube (unknown 6SL7GT)

So I guess I found the source for the hiss, it is the noise from several (2 currently) Russian 6H9C tubes.

Right now I run the amp with the power tubes, the PI and V3 (6SL7GT), and there is still some audible hum. I don't know where this is coming from, maybe also some imission?

So I guess I will next order a whole set of new tubes. Any objections to a set of JJs, it would be the least expensive solution?

 

[Schlumpfmeister]

Ok, the new set of tubes is in the amp, the oszillation is gone, so is most of the hum. For a while the amp showed some intermittend static, but this seems to disappear on its own right now. Why the amp is so much more stable with the new tubes I dunno. I had all tubes in a tester prior to running it, and I had purchased 3 Russian replacement tubes with excellent tester parameteters!?

What remains and becomes very obvious now is a pretty pronounced background hiss which I attribute to the resistors now. I am not sure wether I want to replace all resistors on chance in the hope for reduced hiss!?

 

[Schlumpfmeister]

The hiss or static will not disappear and the amp is quite noisy static / hiss wise. Here my sympotmatic analysis:

Analysis of PI and input to PI
- If I pull tubes 1 through 3 and run power tubes and PI only and if I ground the input to the PI at pin 1 the amp is dead silent.
- The same if I ground upstream anywhere behind the 100nF coupling cap from the plate of V3-1 (pin 2) which is the stage prior to the PI.
- Test wise I disconnected the plate resistor at/above V3-1 from power supply point E. If I ground the plate pin before the coupling cap: silient. If I ground above the 47k plate resistor: static.
- Test wise I replaced the resistor - no difference.

Assuming that the above last test was pointless due to high ohmage at ground and low ohmage at the PI I look at V3-1 effects.
EDIT: Which I don't understand, a tube is high impedance, the input resistor from grid to ground is 470k (it measures ok, and i have already tried a replacement part with no improvement). So why would it not short to ground through the 47k plate resistor?

Analysis of V3-1 (stage before PI):
- I disconnect all inputs to V3-1 by grounding pin 1 and removing the 22k resistor connecting the cathodes of V2-1 and V3-1. This is exactly the situation I reported in my post of 1-12-22 prior to the tube swap, but now all new tubes.
- Strange enough: static, similar to the last test, grounding above the V3-1 plate resistor.
- I measured the resistance between the grounds at V3-1 and PI, and it reads 0,1 Ohms which I consider 0.

So why is the amp quiet with PI grounded, but not with V3-1 grounded? The V3-1 plate resistor seems ok, so seem all solder joints involved (at this point I have resoldered pretty my anything inside of this amp!)

Is the observation of no effect of grounding above the V3-1 plate resistor ok or are we facing trouble right there?

I am again clueless and happy for any insights or advise!

 

[hedberg]

The hiss or static will not disappear and the amp is quite noisy static / hiss wise. Here my sympotmatic analysis:

Analysis of PI and input to PI
- If I pull tubes 1 through 3 and run power tubes and PI only and if I ground the input to the PI at pin 1 the amp is dead silent.
- The same if I ground upstream anywhere behind the 100nF coupling cap from the plate of V3-1 (pin 2) which is the stage prior to the PI.
- Test wise I disconnected the plate resistor at/above V3-1 from power supply point E. If I ground the plate pin before the coupling cap: silient. If I ground above the 47k plate resistor: static.
- Test wise I replaced the resistor - no difference.

Assuming that the above last test was pointless due to high ohmage at ground and low ohmage at the PI I look at V3-1 effects.
EDIT: Which I don't understand, a tube is high impedance, the input resistor from grid to ground is 470k (it measures ok, and i have already tried a replacement part with no improvement). So why would it not short to ground through the 47k plate resistor?

Analysis of V3-1 (stage before PI):
- I disconnect all inputs to V3-1 by grounding pin 1 and removing the 22k resistor connecting the cathodes of V2-1 and V3-1. This is exactly the situation I reported in my post of 1-12-22 prior to the tube swap, but now all new tubes.
- Strange enough: static, similar to the last test, grounding above the V3-1 plate resistor.
- I measured the resistance between the grounds at V3-1 and PI, and it reads 0,1 Ohms which I consider 0.

So why is the amp quiet with PI grounded, but not with V3-1 grounded? The V3-1 plate resistor seems ok, so seem all solder joints involved (at this point I have resoldered pretty my anything inside of this amp!)

Is the observation of no effect of grounding above the V3-1 plate resistor ok or are we facing trouble right there?

I am again clueless and happy for any insights or advise!

Boy, this is confusing to me. I'm looking at three different schematics for this amp (two from the Ampeg site and the one reconstructed by the thread starter -- flyswatter) and they are all different and the way that V2 and V3 stages are assigned is different in each. From your description, it seems that on your amp the stages for V2 and V3 would be, on the schematic from left to right, V21, V31, V22, V32. V22 would be the reverb driver and V32 would be the reverb recovery stage. V21 would be a gain stage driving the cathode follower mixing stage V31. Is that all correct or am I fubar? Furthering my confusion, the schematic reconstructed by flyswatter shows a route from the plate of V21 to the grid of the cathode follower stage, a route not on either of the Ampeg schematics and which you don't find on your amp, if I understand correctly.

So, I'm not sure I can offer any useful help. But, If the stages are oriented as I infer, if you pull V2, it will eliminate both the reverb stages and the driver of the cathode follower mixing stage (V31). Reverb pot set to no reverb jfthoi. Then, the only source of your noise would be V31 and the following circuitry to the PI. There are a couple of caps in there -- .1uf and 1nf. Have these been eliminated as the source of noise? I think I've read everything preceding, but if I've missed that, I apologize.

Anyway, connected to the V31 plate you have a 100k and a .01 uf and a 1nf cap. The PI gets its signal from the 1nf 470k junction. If you attach your scope to the junction, you'll see the noise. What happens if you connect your scope to pin 2 of V31, is the noise still there?

Another thought. You've said of the noise that it's pretty much pink to 8khz or so. Shouldn't that 1nf cap tend to roll off the higher frequencies to ground? I'm guessing that they put that cap in there to attenuate problem high frequencies but is it working if the noise is getting that bright?

I'm just making guesses here, I apologize if I'm covering old ground/not staying in my lane. I'm certainly no expert on troubleshooting things like this.

 

[hedberg]

Oh, just noticed something else. In one version of the schematic (and in Flyswatter's amp) the trem circuit works by modulating the power tube bias, if I understand the schematic correctly. I'm not seeing how this could cause your problem though. The other schematic shows a roach of some sort operating on the output of the first gain stage.

 

[David Barnett]

First question: I'm wondering the best way to replace the large 60/40/10/20 uF can cap, seen below. Fliptops sells a replica, but it's $47 U.S. and that's going to run me double that in CDN by the time it crosses the border.

I'd bite the bullet on that one.

Here's an equivalent for a few bucks less: https://www.tubesandmore.com/products/capacitor-ce-mfg-60-f525v-40525-10450-2550

 

[Schlumpfmeister]

Boy, this is confusing to me. I'm looking at three different schematics for this amp (two from the Ampeg site and the one reconstructed by the thread starter -- flyswatter) and they are all different and the way that V2 and V3 stages are assigned is different in each. From your description, it seems that on your amp the stages for V2 and V3 would be, on the schematic from left to right, V21, V31, V22, V32. V22 would be the reverb driver and V32 would be the reverb recovery stage. V21 would be a gain stage driving the cathode follower mixing stage V31. Is that all correct or am I fubar? Furthering my confusion, the schematic reconstructed by flyswatter shows a route from the plate of V21 to the grid of the cathode follower stage, a route not on either of the Ampeg schematics and which you don't find on your amp, if I understand correctly.

So, I'm not sure I can offer any useful help. But, If the stages are oriented as I infer, if you pull V2, it will eliminate both the reverb stages and the driver of the cathode follower mixing stage (V31). Reverb pot set to no reverb jfthoi. Then, the only source of your noise would be V31 and the following circuitry to the PI. There are a couple of caps in there -- .1uf and 1nf. Have these been eliminated as the source of noise? I think I've read everything preceding, but if I've missed that, I apologize.

Anyway, connected to the V31 plate you have a 100k and a .01 uf and a 1nf cap. The PI gets its signal from the 1nf 470k junction. If you attach your scope to the junction, you'll see the noise. What happens if you connect your scope to pin 2 of V31, is the noise still there?

Another thought. You've said of the noise that it's pretty much pink to 8khz or so. Shouldn't that 1nf cap tend to roll off the higher frequencies to ground? I'm guessing that they put that cap in there to attenuate problem high frequencies but is it working if the noise is getting that bright?

I'm just making guesses here, I apologize if I'm covering old ground/not staying in my lane. I'm certainly no expert on troubleshooting things like this.

Thanks! The flyswatter schem is correct for the R12RB with the exception of the incorrect route as mentioned above. Also the trem is bias, not roach.

The amp noise has apparently developed further, right now it has reduced to a constant low volume level hiss and a (again relative to the rest more pronounced) hum.

I have eliminated anything around V31 test wise, with no success.

Unfortunately, the scope is not of great use here, the noise levelsmdont really show very well there, it seems too low. I will try again later.

I'd bite the bullet on that one.

Here's an equivalent for a few bucks less: https://www.tubesandmore.com/products/capacitor-ce-mfg-60-f525v-40525-10450-2550

That’s the one I have put in.

I will clean all connections today including the star point to chassis and the tube sockets, as I more and more suspect a bad connection somewhere (not that I haven‘t tried before though!)

 

[Schlumpfmeister]

I will clean all connections today including the star point to chassis and the tube sockets, as I more and more suspect a bad connection somewhere (not that I haven‘t tried before though!)

I have (again) checked all ground-to-chassis connections (star point next to the can cap, dimension pot, input plugs) and cleaned as well as secured them all. Also I cleaned all tube socket pins and tightened them, as well as the reverb plugs.

Noise level (hiss and hum) has not changed at all. As a final test I will go through all stages and see if there is a point where the hiss/hum increases. But I am at a point where I will accept this noise level as normal for such an old amp.

 

[timfred]

Hmmm I’m stumped. Is C9 (first coupling cap) good?

 

[andrewRneumann]

Does this amp have elevated heaters? How is the heater circuit grounded?

Sometime nasty voltage spikes make their way into the heater secondary via coupling with the HT secondary. Put that scope on the heater secondary and see what it looks like.

 

[Schlumpfmeister]

Hmmm I’m stumped. Is C9 (first coupling cap) good?

I will double check. By now, most of the coupling caps are replaced.

Does this amp have elevated heaters? How is the heater circuit grounded?

I have sent the scope back, sorry.

But I have tried
- Stock: heater transformer CT at chassis ground
- heater transformer CT at „K“ which is power tube cathode level
- artificial CT by way of 2 x 100 Ohms from the heater cables to either ground or power tube cathode

It has no impact, the hum always remains.

Since I have also drawn a schematic including all ground points. I cannot see any ground loops, although there are many chassis connections. This is well thought out, the chassis serves as bus and as far as I can tell all chassis ground connections are either decoupled by a cap, or they provide ground to a select segment of the circuit. There seem to be 2 separate ground areas on the board, one the far right side for the input stages, connected to the chassis through the input plugs, and one roughly in the middle of the board for PI and reverb, connected through the dimension pot housing.

 

[andrewRneumann]

I will double check. By now, most of the coupling caps are replaced.



I have sent the scope back, sorry.

But I have tried
- Stock: heater transformer CT at chassis ground
- heater transformer CT at „K“ which is power tube cathode level
- artificial CT by way of 2 x 100 Ohms from the heater cables to either ground or power tube cathode

It has no impact, the hum always remains.

Since I have also drawn a schematic including all ground points. I cannot see any ground loops, although there are many chassis connections. This is well thought out, the chassis serves as bus and as far as I can tell all chassis ground connections are either decoupled by a cap, or they provide ground to a select segment of the circuit. There seem to be 2 separate ground areas on the board, one the far right side for the input stages, connected to the chassis through the input plugs, and one roughly in the middle of the board for PI and reverb, connected through the dimension pot housing.

Sound like it isn’t the heaters. Still I’m left wondering. I have heard that a way to rule out heater hum is to rig a switch that momentarily cuts the heaters while leaving the rest of the amp energized. If the hum goes away in those few seconds with the heaters off, then we have our culprit.

Not sure that helps, but if you really want to rule out the heater circuit, that’s what I would do.

 

[Schlumpfmeister]

Final update, the amp is sort of finished now:
- A very slight hum has remained, and it is not from the heaters, neither is it from bad filtering or whatever, but from an asymmetry in the OT. This was reproduced by way a simulation. It could be remedied with an active filter network utilizing some silicon or with some inductance within the filter network, but this would be very difficult to implement given the space constraints.
- The hiss is gone, it results from selection of tubes, in praticular, V3. By now I am back running it with a mixture of old tubes and it works fine. Overall the amp is super sensitive to changes in tubes, and I am not referring to voodoo NOS, but one or the other. Even functions like reverb will greatly change when switching from one tube to another, both tubes new or with good tester readings!
- Point blank these amps have no ground loop issue. I have researched this inside out and changed the grounding concept all together to a straight star system with no improvement. The original concept makes a lot of sense and works well. Don't touch it.
- The same applies to the supposedly-low-quality caps utlilized. I replaced all electrolytes and all tiny chiefs and neither has had an impact on the amp nor could I identify any problem with any of the parts removed.

So overall I have spent an enormous amount of time hunting for ghosts, I am left with an amp that works well and sounds good.

I am surprised how mellow it actually sounds, compared to a blackface amp, it has literally no sparkle. Because of this I have eliminated the 1nF cap to ground between V3-1 and PI. I have been experimenting with the tone stack, but the 2nF cap there does already serve as sort of bright cap and I was unable to get a lot more treble through the amp, this mellow sound seems to be its trademark.

 

[andrewRneumann]

A very slight hum has remained, and it is not from the heaters, neither is it from bad filtering or whatever, but from an asymmetry in the OT. This was reproduced by way a simulation.

Wow! How did you figure that out with a simulation?

 

[Schlumpfmeister]

Wow! How did you figure that out with a simulation?

I measured the components, one of the gentlemen in a German forum used LTSpice to implement the circuit and simulated it. It showed very clearly that an asymmetry w/in the OT increases hum although the NFB will try to compensate for it, and also the asymmetry will generate more uneven harmonics which sounds more pleasant.