Hiss in BF Super Reverb

[Dion]

Been in the process of modding my '66 Super Reverb to a Cesar Diaz spec I have...the amp is definitely players condition...repro cab, non original speakers, lots of road wear, so no worries by me on modding it. It's my player.

Anyway...I'm having some hiss...was having some hum, fixed that.

http://picasaweb.google.com/shannondew/ ... verbIssues

Here are some pictures as she stands right now...I reduced the hum completely by using a spudger and fiddling with the lead dress, BUT the white noise hiss is still there...I fired it up with V1 and V6 in...no hiss...added V2, no hiss...added V4...hiss...Tried a ton of 12AT7's and the hiss remains...I redid all the plate load's tonight and lifted each cathode bypass out of the circuit one by one trying to isolate if it were cathode leakage. For purists looking in horror at the non-original state of the amp...don't worry, I have all the parts I removed/replaced...but it's a "player" condition amp and I intend to play it, not sell it for a profit...

BTW, the amp currently sounds great, but the hiss...not sure if it's just something typical that all amps like this deal with or what...

Any suggestions on what I should be looking at to reduce the hiss?

Below is a history of my changes...

'66 Super Reverb History

-New Filter Caps (not done by me)
-New Bias Cap (not done by me)
-New Power Tube sockets (ceramic)
-.1uF Coupling caps replaced with Orange Drop 6PS .1uF equivalents
-Plate load resistors replaced with Carbon Film equivalents
-Cathode bypass caps (25uF) replaced with Sprague Atom 25uF / 25vdc
-Midrange pot on vibrato channel changed from 10K to 25K
-Vibrato circuit disconnected at Intensity knob
-Split V1/V2 cathode
-New 1.5k/25uF cathode bypass for V1
-Changed 250pF ceramic cap in vibrato tone stack to 250pF silver mica
-Changed 820 ohm/25uF cathode bypass to 2.7k / 25uF cathode bypass on V2 pin 3
-Changed V2 pin 8 cathode bypass to 2.7k/.68uF
-Changed .02uF preamp coupling cap to .04uF 6PS Orange Drop
-Lifted 220k resistor at PI to remove normal channel from circuit
-Replaced 100ohm resistor with 47ohm
-Replaced .001uF ceramic cap on pin 2 of PI with .002 6PS Orange Drop
-1ohm 2w resistors added between pin 8 and ground of both power tube sockets (bias test points)
-Biased to 38.4 mA[/url]

[Structo]

Looks like you replaced most of the likely suspects, carbon comps.
But that would be my suspicion, a noisy resistor somewhere.

[Dion]

Looks like you replaced most of the likely suspects, carbon comps.
But that would be my suspicion, a noisy resistor somewhere.

Any way to isolate it further without replacing all the other carbon comps?

[Firestorm]

You could try replacing the dropping resistors in the PS with metal film of the appropriate wattage. I've cleaned up some hissy amps this way. Maybe worth a try.

[Bob-I]

Good to see you here my friend!!

V4 is the reverb return/gain recovery after the 3.3M/10pF. Plus you have the normal channel disconnected. FWIW that tube should be a 12AX7 but that should be your problem.

I'd start by reconnecting the normal channel, there's no telling what will happen with an open connection like you have now. There's actually no need for the 220k mixer resistor before the PI if you disconnect the normal channel, so I'd put a jumper wire over it.

Also, try disconnecting the reverb pan, you may have an issue with one of the cables or jacks.

[Bob-I]

Looks like you replaced most of the likely suspects, carbon comps.
But that would be my suspicion, a noisy resistor somewhere.

Any way to isolate it further without replacing all the other carbon comps?

Yes, if you don't have a scope, make up a stethoscope to listen to the various stages. I have a 1/4" in line jack connected to a .1uF cap on the hot lead, then to an alligator clip, the ground lead goes right to an alligator. Connect the ground to the chassis, plug in an instrument and listen at the input and output of V4, grid1 then plate1, grid2 then plate2, by connecting the hot alligator clip to these pins on the tube. Be careful when going to the plate, the cap will block the DC from going to the listening amp but nothing will protect YOU. I use my home stereo to listen but any type of amp will do, as long as it doesn't hum.

Personally I think it's most likely from something you did. You may have put in a bad resistor/cap or have a cold solder joint.

[Dion]

Looks like you replaced most of the likely suspects, carbon comps.
But that would be my suspicion, a noisy resistor somewhere.

Any way to isolate it further without replacing all the other carbon comps?

Personally I think it's most likely from something you did. You may have put in a bad resistor/cap or have a cold solder joint.

I'm not totally convinced of that. I've went through the plate loads twice now. All of the caps lifted out fine without any change in the hiss. I reflowed every solder joint I did...I've constantly used forceps/heatsinks to prevent component overheating. I also reversed every change with the exception of the coupling caps and the 25K midrange pot and had the same results

Remember, I started all of this due to the nasty snap crackle and pop and that was resolved by the new plate loads...the hiss was then exposed...

Anyway, I'm also going to look at the 3M3 on the reverb return...that's a strong possibility...

[Bob-I]

Anyway, I'm also going to look at the 3M3 on the reverb return...that's a strong possibility...

That's an easy one, just bridge over it with a clip lead. The gain will go through the roof and the reverb won't work, biut you'll find out if that's the problem.

[Firestorm]

Anyway, I'm also going to look at the 3M3 on the reverb return...that's a strong possibility...

That's an easy one, just bridge over it with a clip lead. The gain will go through the roof and the reverb won't work, biut you'll find out if that's the problem.

The gain will not only go through the roof, it will go through any low-flying airplane that happens to be in the vicinity. The 3M3 could get a little noisy (just replace it with metal film) but not that much noise in "hissy" frequencies is going to make it across a 3M3. The 10pf ceramic, on the other hand, is there specifically to bypass high frequencies around the 3M3, so I'd look there. Replace it with silver mica.
I'd disconnect the other end of the "intensity" wire, too (connected under the board to the 220K mix resistor) I just don't like the idea of a grid wire hanging around in a high signal area with nothing to do.
One more thing, by changing the shunt resistor on the feedback loop from 100R to 47R, you're obviously reducing the amount of negative feedback. All well and good, some folks disconnect the loop entirely. But a stock Super already has less feedback than a Twin or Deluxe from the same era. A lot of the noise cancelled by negative feedback is high-frequency, so that change will always add some hiss.

[Dion]

Anyway, I'm also going to look at the 3M3 on the reverb return...that's a strong possibility...

That's an easy one, just bridge over it with a clip lead. The gain will go through the roof and the reverb won't work, biut you'll find out if that's the problem.

The gain will not only go through the roof, it will go through any low-flying airplane that happens to be in the vicinity. The 3M3 could get a little noisy (just replace it with metal film) but not that much noise in "hissy" frequencies is going to make it across a 3M3. The 10pf ceramic, on the other hand, is there specifically to bypass high frequencies around the 3M3, so I'd look there. Replace it with silver mica.
I'd disconnect the other end of the "intensity" wire, too (connected under the board to the 220K mix resistor) I just don't like the idea of a grid wire hanging around in a high signal area with nothing to do.
One more thing, by changing the shunt resistor on the feedback loop from 100R to 47R, you're obviously reducing the amount of negative feedback. All well and good, some folks disconnect the loop entirely. But a stock Super already has less feedback than a Twin or Deluxe from the same era. A lot of the noise cancelled by negative feedback is high-frequency, so that change will always add some hiss.

Thanks Firestorm, I completely didn't think to increase the NFB when I was reversing stuff trying to track it down...thanks bro, I'll try that.

[butwhatif]

If you lifted the connection between the input to the phase inverter and the 50k intensity pot, it will increase hiss and gain. The 50k pot loads the input to the PI quite a bit. If you disc'd that, there's your issue. And why would you want to disc the norm channel---you should leave it connected. With an a-b box it is a channel switching amp.

[Dion]

If you lifted the connection between the input to the phase inverter and the 50k intensity pot, it will increase hiss and gain. The 50k pot loads the input to the PI quite a bit.

I disconnected it at the lug on the intensity pot.

Lotta stuff to check out tomorrow when the kids are down for a nap

[Bob-I]

Anyway, I'm also going to look at the 3M3 on the reverb return...that's a strong possibility...

That's an easy one, just bridge over it with a clip lead. The gain will go through the roof and the reverb won't work, biut you'll find out if that's the problem.

The gain will not only go through the roof, it will go through any low-flying airplane that happens to be in the vicinity. The 3M3 could get a little noisy (just replace it with metal film) but not that much noise in "hissy" frequencies is going to make it across a 3M3. The 10pf ceramic, on the other hand, is there specifically to bypass high frequencies around the 3M3, so I'd look there. Replace it with silver mica.

IMHO that resistor is much too large. It's there to mix the reverb and since Fender's reverb are useless above about #3 on the dial, I typcally replace these resistors with 1M or even 470K. You still get plenty of reverb but you can drive the PI harder and get some really great breakup.

One more thing, by changing the shunt resistor on the feedback loop from 100R to 47R, you're obviously reducing the amount of negative feedback. All well and good, some folks disconnect the loop entirely. But a stock Super already has less feedback than a Twin or Deluxe from the same era. A lot of the noise cancelled by negative feedback is high-frequency, so that change will always add some hiss.

I think lowering the NFB is the point of changing the resistor. I usually lower the NFB by raising the NFB resistor from 820 to 1.5k, same effect. I haven't noticed any hiss increase on my amp though.

[Firestorm]

I think lowering the NFB is the point of changing the resistor. I usually lower the NFB by raising the NFB resistor from 820 to 1.5k, same effect. I haven't noticed any hiss increase on my amp though.

Sometimes I find the hiss goes up when messing with the feedback. Some amps seem to have more room for tweaking. It may even depend on the interaction between the frequency-dependent impedance of the speaker and pure resistance of the feedback resistor. I was always intrigued by Fender's choice of feedback/shunt resistor values in the BF amps. 8-ohm speakers -- 820R/47R; 4-ohm speakers -- 820R/100R; 2-ohm speakers -- 820R/100R. Why didn't they keep going in the direction they were obviously headed when they got to the Super?
But I digress...

[Bob-I]

I was always intrigued by Fender's choice of feedback/shunt resistor values in the BF amps. 8-ohm speakers -- 820R/47R; 4-ohm speakers -- 820R/100R; 2-ohm speakers -- 820R/100R. Why didn't they keep going in the direction they were obviously headed when they got to the Super?
But I digress...

Agreed. I guess that's why so many folks mess with the NFB on a Super. Many have it disconnected, but I find they sound better if you bring it into the proper range.

I also don't understand why they went with 2 ohms. That's so close to a dead short you'd think they'd use 16 ohm speakers and a 4 ohm xformer like in the bassman.