Super Reverb

[pbiagi]

Initially I didn't like the thread being hijacked by what I regarded as a distraction however I think it had to be.
One more point - this is a great forum, truly... There are other forums where superficial or nonsense advice is given by just about anyone.
Personally I really appreciate the rally to keep this forum smart.

I don't follow as closely as you all but I happened on another thread where SGM's advice talked about European electricity... since I live in Europe.. thought .. uh.. what?

You can label my ideas as wacky. Take it or leave it.

If you make this your signature, we'd probably all be OK with your posts.

Last night I drafted a reply covering several points.. but hit the 'Reply to Thread' button instead of the 'Post Quick Reply' .. and lost all my text.. which I'm re-writing here. And.. just figured there is a recover button - wish I had seen that.

To the Super Reverb -

"mint condition-looking vintage amp" to me says "sounds like crap, I was afraid to play it out."

LOL, it's not my favorite amp but it doesn't sound so bad.. I hesitate to gig with it because I'm afraid to bang it up (even regret that it's cosmetically so well).

It is the AB763 circuit. I have not checked the cap values but only change from AB763 I've noticed is the extra 100 ohm resistors on the heater filament. Maybe someone added that.

I'll take it slow, remove the 100 ohm resistors to ground on the heater filaments, replace the bias cap with 100uF/100v and remove the death cap.
It already has the 3 prong chord and the electrolytics were all already replaced. I'll keep the Mallories.
I like silver mica caps (especially in hifi), and may change out the ceramics in the signal path. Had already ordered the exact ones Steve indicated.
Steve, Thanks for the treble cap experimenting tip - will try that. Anything removed will be kept for restoration.

You might also want to replace the 3M3 resistor with 2M2 as Mesa Boogie did with some of their early amps.

(Since I'm too lazy to look it up,) what tonal difference will this make in the vibrato (tremolo) circuit?
Somewhere I once read of a mod which makes the tremolo more choppy - would be interested in trying that (especially if easily reversible).

I may also try some of G. Weber's (book) mods, like disabling the negative feedback circuit.
Will do these one at a time to hear what they're doing.

[Steve A.]

(Since I'm too lazy to look it up,) what tonal difference will this make in the vibrato (tremolo) circuit?

While the EE students would compute the knee frequency of the RC network I look at things more intuitively. Ignoring the cap we have a voltage divider circuit with the 3M3 resistor on the top and the equivalent of a 150k resistor on the bottom which means that roughly 5% of the pre-reverb signal is passed on to the grid of the following stage. With a 2M2 resistor that goes up to roughly 7%.
Adding a cap in parallel with the resistor increases the high frequencies in the signal. Just as a treble bleed tone control shunts high frequencies to ground the 10pF cap allows more high frequencies to pass through. There are two different things going on. On one hand a smaller pF cap will pass higher frequencies but on the other hand the higher pF caps seem to pass a stronger signal (I believe that has to do with the ESR- Equivalent Series Resistance.) A good example would be the BF treble cap. Looking at the RC charts you would think that 50pF cap would be very much brighter than a 250pF cap. However at least in my experience the 50pF cap passes much less of the signal so its extra brightness is compensated by a comparatively stronger signal coming from the bass and mid circuitry. Increase the 250pF cap to 750pF or 1000pF and you have a strong midrange boost. [The treble pot is basically a mixer with the signal from the treble cap on one side of the pot and the signal from the bass and mid circuitry on the other side. So it makes a really big difference if the treble cap is audio or linear- much more so than the bass or mid pots which are wired as variable resistors.]
Getting back to ESR it seems to me that a mica cap will pass through more of the original signal than a ceramic cap. So just replacing the 10pF ceramic cap with a 10pF will pass on more signal to the following grid.

I may also try some of G. Weber's (book) mods, like disabling the negative feedback circuit.

The stock negative feedback resistor is 820 ohm which gives you a very tight bass response. Removing it can give you more volume but it can also make your bass flabbier. (All descriptions are my own. YMMV) A good compromise is to wire in a 25K linear pot in series* with the 820R resistor (Marshall guys might put in a 100k resistor.) For a vintage amp like yours there is a good place to mount the added pot: the hole for the ground switch. If you want to put in a second pot you might consider removing the Ext Spkr jack. Getting into hack'n'butcher territory the bottom of the chassis is a good place to add mode switches, like pentode/triode or fixed bias/cathode bias. But I digress...

* It is a good idea to put a "tail" on a pot when you use it to replace a fixed resistor since reducing the resistance to 0 ohms could have undesirable results.

Steve Ahola

P.S. Punch "Super Reverb" into the search bar here and you get 5 pages of hits.

[pbiagi]

Wow, thanks Steve. I've got plenty to digest now.. It'll be a couple weeks before I get started and I think I'll be pouring over your every word.

[trobbins]

Getting back to ESR it seems to me that a mica cap will pass through more of the original signal than a ceramic cap. So just replacing the 10pF ceramic cap with a 10pF will pass on more signal to the following grid.

I think many people expect ESR to be much more influential than it is - just because it is a 'value'. But in reality a change in capacitance is pretty much the only thing noticeable.

[timrichter9]

I agree on wiring a pot into the NFL. I did this on my '70 non-master volume after reading about the mod in the 90s. I think it was touted as the "Soul Control".
While it is a nice rhyming term to describe it, I do not think it added "soul" (although the tie-dyed grillcloth did that nicely.

[Steve A.]

I think many people expect ESR to be much more influential than it is - just because it is a 'value'. But in reality a change in capacitance is pretty much the only thing noticeable.

I had noticed a louder and stronger signal passing through a mica cap compared to a ceramic cap of the same value before I ever heard of ESR. I had also expected the Vox AC30/Trainwreck Rocket tone stack to be extremely bright with its 50pF treble cap compared to the 250pF treble cap of the Fender BF tone stack. But the 50pF cap does not pass through as much signal as the 250pF cap so it works out fine.

Just my own observations...

Steve Ahola

[pdf64]

'I had also expected the Vox AC30/Trainwreck Rocket tone stack to be extremely bright with its 50pF treble cap compared to the 250pF treble cap of the Fender BF tone stack'
Steve, I don't follow your rationale with that.
At the first approximation, the BF 250pF treble cap feeds a 250k pot track, resulting in a high pass turnover frequency of 2k5Hz, whereas the Vox 50pF treble cap feeds a 1M pot track, giving a high pass frequency turnover of 3k2Hz.
Given component tolerances, my feeling is that's pretty much equivilant.
But if the treble cap value is halved, I don't perceive that the control becomes 'brighter'; rather it becomes less effective, probably because the frequency range it affects will be in the high frequency roll off area of the speaker's response, and generally there's less sound 'content' up there anyway. So the amp becomes less 'bright'.
To extend the principle, as the treble cap value is reduced to 0, the control will become less and less effective, and so the amp will become less and less 'bright'.
As an example, the Fender Super 60 uses a regular BF tone stack but with a 100pF treble cap (along with a power amp presence control but no bright cap); my perception was that it adds a silky sheen to the tone, rather than the strident top end that the regular value supports.
http://www.webphix.com/schematic%20h...er_60_rack.pdf
Pete

[pbiagi]

Ok, got back to this amp.
- disabled the heaters artificial center tap (since the green-yellow wire was grounded already)
- removed the death cap
- replaced the bias circuit electrolytic cap with 100uF/150V

.. and for the fun tone stuff -
Replaced the 3M3 resistor with 2M2 and replaced the 10p cap with silver mica. Wow, that is a wonderful change! Reverb is richer and deeper.. I'm starting to like this amp again!

However for the treble cap, seems someone already changed that - it's a 500p mica cap. (The mid is 0.022uF.)

Which brings me to my next challenge.
A guitar with p90 pickups sounds nice.
However the strat (and tele), with volume up at moderate level, have a sort of saturated knock on the treble side. Sort of like compressed pick noise. .. and none of this is good saturation/compression, it's annoying, however I'm exaggerating to make the point. And more confusing to describe, after the attack, the treble is thin...
I'm wondering if the issue is with the speakers (which are original) - maybe there's one bad?. I may rig a patch system for my 5f6a speakers, to hear the amp through them.
Or.. maybe it's the 500p mica cap? Maybe I should try changing out to 250p mica.?

Thoughts?

In any case the amp sounds much nicer already. Thanks for the advice!

[Randall]

As a pro player myself, I get real antsy with the idea of improving upon Leo Fender designs. I have a '65 BF Super Reverb in my shop right now. I will replace defective components with the original value, unless it is going higher voltage or wattage rating for reliability and know it won't change the sound. My thinking is if that's the way it was designed and built by Mr. Fender, and it is an amp well known for it's sound, then that's the way it should stay. I'm sure glad nobody fixed or improved many of the dirty amps on many of my favorite blues records!

Anyway, I really enjoy and learn a lot from this forum. A lot of serious talent here.

[soundguruman]

Ok, got back to this amp.
- disabled the heaters artificial center tap (since the green-yellow wire was grounded already)
- removed the death cap
- replaced the bias circuit electrolytic cap with 100uF/150V

.. and for the fun tone stuff -
Replaced the 3M3 resistor with 2M2 and replaced the 10p cap with silver mica. Wow, that is a wonderful change! Reverb is richer and deeper.. I'm starting to like this amp again!

However for the treble cap, seems someone already changed that - it's a 500p mica cap. (The mid is 0.022uF.)

Which brings me to my next challenge.
A guitar with p90 pickups sounds nice.
However the strat (and tele), with volume up at moderate level, have a sort of saturated knock on the treble side. Sort of like compressed pick noise. .. and none of this is good saturation/compression, it's annoying, however I'm exaggerating to make the point. And more confusing to describe, after the attack, the treble is thin...
I'm wondering if the issue is with the speakers (which are original) - maybe there's one bad?. I may rig a patch system for my 5f6a speakers, to hear the amp through them.
Or.. maybe it's the 500p mica cap? Maybe I should try changing out to 250p mica.?

Thoughts?

In any case the amp sounds much nicer already. Thanks for the advice!

Depending on speakers, 100 or 120 pf sounds much less harsh for the bright cap.
250 or 500 is passe. rakes on the ear too much.
Silver Mica cap, of course. You will get less hum in the audio path with silver mica.
This is sometimes helpful:
http://www.duncanamps.com/tsc/
and then again in the tone stack, I use 120 for the treble control too...
see, it's 250 or 500 passes too much upper mid, and makes the treble too obnoxious.

Bias on the hot side. Bias set too cold is going to make the highs very harsh.
This is the biggest problem we see, and the treble sounds nasty with the bias cold.

[Steve A.]

'I had also expected the Vox AC30/Trainwreck Rocket tone stack to be extremely bright with its 50pF treble cap compared to the 250pF treble cap of the Fender BF tone stack'
Steve, I don't follow your rationale with that.

Try wiring up a switch with 3 different value treble caps. The lower value caps are brighter but not as loud. If you go down to 50pF in a BF tone stack there will be a brighter signal passing through the treble cap but with the lower signal level it is more likely to be overpowered by the lower frequencies coming up through the bass pot.

When wiring up a switch with ceramic and mica coupling caps of the same value the mica cap will pass a stronger signal than a ceramic cap. As to why it is like that I always used the term "ESR" although that might not be technically correct.

Steve Ahola

[Steve A.]

Or.. maybe it's the 500p mica cap? Maybe I should try changing out to 250p mica.?

For a BF tone stack I usually start with a 330pF mica treble cap and will switch in a 390pF mica cap on a push-pull pot for a mid-boost (or "shift" as Mesa Boogie called it in their early amps) and then take it from there.

Two channel BF amps are cool because you can keep one of the tone stacks stock while you experiment with the other.

The Duncan TSC program is very helpful in seeing the effects of component change in the Fender tone stack. I will also experiment with the 100k slope resistor- if I want more highs I might raise it to 110 or 120k, if I want less I might lower it to 91k or 82k.

BTW for intermediate resistor values I will add them together in parallel using the Windows calculator to determine the value needed. To find the net value of two resistors in parallel you add the reciprocals of both resistors and then find the reciprocal of the sum. To figure out what value resistor (R2) to add to a particular resistor (R1) to create the desired resistance (RT) I will subtract the reciprocal of RT from the reciprocal of R1- the reciprocal of the difference will be the value of the resistor you need to add (R2). I do all of that using the M functions- M+, M- and MR and the 1/x key.

We all know that two resistors of the same value R in parallel will have a net resistance of 1/2R- it is adding resistors of different values that gets tricky. And we all know that capacitors in series are like resistors in parallel, and vice versa. So Windows calculator is my friend.

Steve Ahola

[Joe L]

Silver Mica cap, of course. You will get less hum in the audio path with silver mica.

Hum?? Are you sure you meant to say "hum"? Ceramics can be more microphonic than Mica but there is no deficiency in the dielectric that will pass more 60 or 120 Hz hum that what their value would calculate.

Also, according to the test run at The "Sound" of Capacitors, if you use high voltage Ceramics =>1kV, most of the faults are diminished to the point where the advantages of Mica are hard to hear.

Personally, I use Polyethylene for bright caps, Mica for tone stack caps in Fender circuits that I build for myself. And Ceramics occasionally in Marshall type circuits where I want the highs to have a bit of grittiness. If I can't hear the difference, I use the higher quality product.

[pbiagi]

Thanks for the advice Steve!

Making progress.. but new challenges..

The normal channel treble cap was a original ceramic, 250p. I went ahead and changed it out for mica 250p. (BTW, saving all components..)

For the vibrato circuit, replaced the existing treble 500p mica cap with 250p and 10p in parallel, which measured ~270pF on my DMM. I didn't have anything else to get close to 330pF.

Here's where I think I made a mistake -
On both the normal and vibrato channels I measured the slope resistor in circuit at 104K ohms.
In the vibrato circuit, to reduce the slope resistance down to about 100K ohms, I added a 2M ohm resistor in parallel. But surprise, after the addition the slope resistors measure 109K ohms (in circuit).
I thought maybe I made a mistake in the original reading, to bring down the resistance I added again another 2M ohms, the result measures 106K ohms (in circuit).
Tonally the results are better but ...
The bass tone pot controls the bass obviously, but controls the volume to a certain extent, much more than the middle or treble pots. The middle has the least affect.
When the bass is turned all the way down, the amp has very little volume and sounds tinny of course. I didn't notice the bass pot having this control in the normal channel.
My guess is that I was measuring the slope resistance in circuit, and the paralleled resistors out of circuit would likely measure in the low 90K's ohms. (??) Imagine I over did it, and the bass control is taking most of the signal.?

Amp sounds best setting the bass at 4 max., or less. The amp has much less saturated/compressed treble (good).
However I'm thinking I need to verify everything, especially the tone circuit.
The bass is dull, and the mediums can loose string separation (more than I want) at moderate volume.

We have some French holidays coming up this week and my plan is to remove one of the 2M parallel slope resistors, add possibly add a parallel 50pF mica to the treble circuit (if I can get my hands on one before, or replace with 330pF), and measure DC voltages (from memory did that in past and everything was OK), etc. I'd like to hear it over bassman speakers and will rig something. Also may try changing tubes..
Will need to try one thing at at time, not to confuse myself..
I have a scope and sine-wave generator, limited experience.

Paul

[Steve A.]

When the bass is turned all the way down, the amp has very little volume and sounds tinny of course. I didn't notice the bass pot having this control in the normal channel.
My guess is that I was measuring the slope resistance in circuit, and the paralleled resistors out of circuit would likely measure in the low 90K's ohms. (??) Imagine I over did it, and the bass control is taking most of the signal.?

By my calculations adding two 2M resistors in parallel with the 100K slope resistor would bring the net resistance down to 90.9k which should make the tone stack a little less trebly but not cause the behavior you described. Are you working on an eyelet board, a turret board or a pcb? If a pcb they can be very tricky to work on. On a Fender style eyelet board too much solder can cause problems under the board with the components (been there, done that.)

BTW when experimenting if I make a change that creates problems I will undo what I just did. If it still isn't right then I must have done something inadvertently to the circuit and need to troubleshoot that problem.

Good luck!

Steve Ahola

P.S. There are other ways to adjust the balance of a tone stack. In the Champion 30 ss amp that I have been rewiring the treble control was WAY too bright. Raising the value of the treble cap sounded good but not as Fendery as stock. So I tried a different approach:

The stock treble cap goes directly to the 100k treble pot. To limit the range of the treble control to what was 0-5 on the stock amp I added a 68k resistor between the treble cap and pot, and then added a 47k resistor across the two outer terminals of the 100k pot bringing the net resistance down to 32k. To the rest of the amp circuit it still looks like a 100k pot.

That combination has a very usable range of the treble settings but I would like it to be a little brighter when set to 10 so I'm going to try a 56k series resistor between the cap and pot, and an 82k resistor across the outer terminals of the pot for a net resistance of 44k.

All of this is done on a pcb so to add the resistor in series with the cap I mount them like a steeple, with the cap and resistor each soldered to one of the copper pads and the other leads soldered on top (like a church steeple.)

Here is the formula I use to figure out what resistor to add to the outer terminals of a pot to create the desired resistance:
1/(1/Rpot - 1/Rnet) = Radd

For two resistors in parallel:
1/(1/R1 + 1/R2) = Rnet

I might punch those formulas into the Windows Calculator several dozen times a day when working on amp circuits. (I should get a simple handheld calculator that has keys for 1/x, M+, M-, MR and MC. With dual power to run on solar or batteries. I have one that I used extensively for computing HVAC bills and proposals but I wore the lettering off most of the keys.)