Super Reverb Tonestack Lift Noise

[ModMan]

Hi,
This is my first post I am still in the learning process not that it ever ends.

I own a 71-73 SF Super Reverb it looks like it started a AB568 but now when I got it all resistor/cap values were AB763 it still has the hum balance pot though.

I was after more gain and reverb out of the normal channel I removed the vibrato and the circuit is now as follows-

Input is standard V1a is 100k plate 1.2k bias 2.2uf bypass cap V2a is standard 820/25uf bias shared with V2b.
The tonestack is .1uf/.022uf/470pf with a 4k fixed mid resistor and .047 coupling cap connected to 330k resistor wich is connected in parallel to V5a [old vibrato tube] and another 330k resistor to ground.
V5a is 100k plate 10k bis resistor then the signal goes through a .022 coupling cap and joins the reverb circuit at Z the junction of 470k/3.3m/10pf the circuit is then as normal besides two switches first bypass' NF resistor with .047cap second removes NF completely.

After all this I had a slightly different voice on the normal channel and slightly more gain but not enough for the heavy rock sound I'm after so I tried one last thing lifting the ground on the 4k fixed mid resistor what do you know there is all the gain I ever wanted and I like the tone it is slightly dark compared to the vibrato channel but it sounds fine.

The problem is 1 it is with the eq ground lifted it is to loud and introduces hiss I tried a 12AY7 in V1and a 12AU7 in v2 this seems to balance it out a bit but don't know if the bias on V2 should now be changed for the12AU7?

Secondly when you turn the volume up past six the hiss seems to be canceled out by something I dont have a scope or anything so I'm worried this could be sqaure waving or something I can't see is the signal level now way to high for the following stage?

Thanks in advance any suggestions would be of help and yes I now I should leave this stuff to someone who knows but I have to start somewhere .
Dan.

[Chuck H]

First, if the "hiss" gets quieter as you advance the volume control that would be the amp going into parasitic oscillation. It has the effect of making the amp work very hard at frequencies you can't hear and quieting others. This happens when there is crosstalk between circuits due to inadequate layout or grounding for the amount of gain your trying to achieve and can only be fixed by changing the location of components, leads, ground points and possibly even transformers. Second, if the amp is too loud when it's producing adequate gain that's a consequence of the circuit design. You need to get the power tubes into clipping to get heavy distortion from these amps. Third, the circuit in that amp is designed a certain way to do a certain thing. You can cut wonder bread into circles and punch a hole in the middle but it still won't be doughnuts. With all the hand wired flying leads and reverb, vibrato, second channel and independant EQ wiring inside those amps were only marginally stabil for the gain level they provided. Increasing gain without a complete overhaul of the layout is a futile effort.

This is going to sound extreme but I would say put the amp back as stock as you can. You don't like this amp, it doesn't do what you want an amp to do. Sell it and use the money to get another amp.

Another alternative would be to put the amp back as stock as you can and use a pedal for distortion. Then you can control the volume of the distorted tones below the amps full output. But if it must be cranked tube amp distortion you need a different amp. Spending too much time trying to make a square peg fit into a round hole isn't advancing your musical experience one bit. To fix the problems your having there's more to know about electronics and guitar amp design than you can learn from a couple of posts. So you need to decide if you want to learn how to tweak an amp or find a tone at this point. Because IMHO to get what you want from THAT amp is going to be a time consuming experience that will require learning a lot about amp design and rebuilding your own circuit inside.


JM2C

[ModMan]

Thanks for the reply JM2C.

The hiss is only a problem with he eq ground lifted and it doesn't decrease gradually as you urn the volume it just seem s to vanish only when you play and after 6 on the volume knob. It actually fixes the problem it is quiet but I am worried it is the by product of he signal level being to high for V1b is there a way to lower it a bit.

I am very happy with the way the amp sounds and the amount of gain now, I just want to now what is causing the hiss and is it safe or can I reduce it. Could I just raise the value of the 4k fixed mid on the bas pot and hope I can find a level where there is more gain but it doesn't do the noise cancelation thing?

[pdf64]

I think Chuck H is correct, for various reasons SF amps were only just stable; your gain increase mods and the symptoms resulting re-affirm this.
However if you are determined, then I advise that you fit grid stoppers to V1 - V5 tube bases, try 22k. For V1a and V2a, the 68k input mixer resistors could be moved from the input jack socket to the tube base terminal 2.
Analyse the lead dress; for any plate circuit that has 2 signal inversions, with regard to a preceding grid circuit, the relevant wiring and circuitry needs to be kept well seperated.
I like a 50k audio pot for the mid resistor. At 50% it's about stock value (6k8 rather than 4k?), after that it's increasing levels of boost.
Consider fitting the pot in the hole for RCA reverb footswitch (unless you actually use the reverb footswitch, in which case consider an additional hole in the back panel behind V1).

'V2a is standard 820/25uf bias shared with V2b'
That's bad practice for cascading gain stages, as any ESR that occurs / develops in the bypass cap may cause oscillation. Best to use seperate cathode resistors.
If you're getting into amp modding, consider getting a scope; low spec cheap ebay is probably fine. But you need to see what's going on.
Pete.

[ModMan]

Hey thanks heaps Pete.

Are grid stoppers to prevent oscilation?
Is that what you also thing is ocuuring due to to high a signal level?
Will a 50k pot be like a variable bypass or partial bypass?
Would this stop the noise cancellation at high levels?
Is it correct the shared 820 ohm cathode resistor should be replaced by 2 by 1.5k if seperated?

Dan.

[ModMan]

Also where would I fit the 22k grid stoppers to v1/v5?

[pdf64]

Fit a grid stopper to each grid, terminals 2 and 7 V1 to V5 inclusive. One end to terminal, the other to joint with wire, sleeved. It is good practice to fit them as they will reduce risk of oscillation, by rolling off the high frequency (>20kHz) response (in conjunction with the tube's Miller capacitence).
If much larger values (220k - 470k) are used they can also reduce blocking distortion, at the expense of losing audio high frequencies.
The 50k variable resistor mid control will act as a partial, variable tone stack bypass. Compared to your current arrangement, that may stop the oscillation (supposed) caused by the total tone stack bypass, but if it does I still advise you to undertake additional oscillation mitigation (eg grid stoppers and lead dress review).
To maintain the same bias level, yes a shared 820 cathode resistor should be replace by seperate 1k5 resistors.

[ModMan]

Thanks mate what joint do I joint the grid stoppers to is it from each pin to ground?
Dan

[ModMan]

Also can anyone tell me to move the 68k input resistors to the valve socket do I have to run a extra signal wire?

[ModMan]

I just read if I change the 100k plate resistor on v1b to a voltage divider I can lower the gain from the lifted tonestack into the v1b and the lower resistor will also act as a grid leak resistor to the next stage v5 is this the same as putting a grid stopper will it prevent oscilation?

Also V5a already has 2 330k resistors in parallel at v2 one in series one to ground does this still need a grid stopper?

Thanks again to everyone who has commented.

[pdf64]

'Also can anyone tell me to move the 68k input resistors to the valve socket do I have to run a extra signal wire? '
Yes

'Thanks mate what joint do I joint the grid stoppers to is it from each pin to ground?'
No they are in series with the signal flow, so
wire - resistor - terminal 2/9
See
http://t2.gstatic.com/images?q=tbn:A...6F-9pz5TrvFbVg
Hope that link works.

Grid stoppers need to be positioned at the tube base to function as grid stoppers, so if you can arrange existing resistors like that - great, if not then fit additional grid stoppers at the tube base. You may find that you get away without all grids having them, but you'll never know if that's just good luck and something is marginal, eg change a tube and the new one may have slightly different characteristics, and oscillation starts up - Pete.

[ModMan]

Thanks. I think you meant 2/7 like in your first comment, so looking at the drawing it is the 470 on the input,
also to correct myself I meant make the plate load a slit load not avoltage divider.
Dan.

[pdf64]

Yes, terminals 2 and 7, sorry.
Re the mid control, the 6k8 original fixed resistor needs removing, to allow the new mid 50k variable resistor to raise the resistance above 6k8.

'I just read if I change the 100k plate resistor on v1b to a voltage divider (split load) I can lower the gain from the lifted tonestack into the v1b and the lower resistor will also act as a grid leak resistor to the next stage v5 is this the same as putting a grid stopper will it prevent oscilation?'
I'm having trouble visualising this, could you sketch it, or is there a schematic available (of a commercial amp) with a similar arrangement?

[JoeM]

....
This is going to sound extreme but I would say put the amp back as stock as you can. You don't like this amp, it doesn't do what you want an amp to do. Sell it and use the money to get another amp....

JM2C

I agree with Chuck 100%. I just think the AB763 type circuit doesnt lend itself to what you need/want. I have a feeling no matter what mods you do, you still won't be satisfied with the result.

[akimball442]

I love modding these amps. If you're not using the vibrato, you could use the tubes to create a more conventional high gain preamp. I like Modding Fenders because they're built like tanks, and reliable.

As for the tone stack lift, it says you have a 4K resistor for the mid resistor? That's kind of odd. The original would have been 6.8K. The 4K is going to suck more mids than stock, effectively reducing the apparent gain. Try something like a 22K and see how you like it, or even 47K. You'll still have a boost, but as you increase this resistor, the effectiveness of the tone controls is going to decrease. You can further increase the mids by changing the slope resistor in the tone stack to a lower value, say 56K or 47K, even 33K a la Marshall. To get completely into that territory, try also replacing the .1 cap in the tone stack with a .022 cap.

The simple answer seems that if the amp works, other than lifting the ground on the tone stack, don't lift the tone stack. Keep it tied down with a resistor of some value. 100K will pretty much get you where you want to go, but may keep the amp stable.