Possible pickup problem

[firemedic]

I've been unsuccessfully trying to abolish this noise I get when playing my Alvarez Tele copy on my little SE practice amp- it's like a static-ey kind of nasty little fuzz riding over my midrange notes. Sometimes. Sometimes not. Every time I think I've fixed it, it happens again. I've rebuilt the amp several times & it won't go away- I've changed out all the tubes multiple times.

And it seems to only do it on the neck pickup. It sounds fine on my other amps but maybe it's noticeable on this little amp because it's more sensitive. I have reflowed the solder on the pickup switch & everything looks fine.

I have Fender Vintage Tele (lipstick) pickups. I like the sound but this noise issue is a bummer. Has anyone had similar noise issues with pickups? I'm out of ideas!

[G._Hoffman]

I'm not saying this is the problem - I haven't seen the guitar, so it is impossible for me to diagnose any issues - but you should NEVER reflow solder.  Remove the old stuff, and replace it with new.  when you reflow solder you will almost always get a poor quality joint, as there is no flux in there to clean things up.


Gabriel

[firemedic]

Ah. I have a lot of joints to resolder..... I usually just add more until it becomes too blobulous, then start over.
I'm not proud.

[John]

I'm not proud.

 

[jjasilli]

I had that problem too.  I call it "fizzy speaker".  NFB fixed it.  Do you have NFB?  If yes, try more of it.  Post if you want more info.

[firemedic]

Huh. I did steal the basic topology from the Princeton SE design.....
Leo had 22k NFB but I got rid of it. Let me put it back in & I'll report back.
I have rebuilt the amp FOUR TIMES trying to fix this thing. 

[jjasilli]

Note that if you added a cathode bypass cap to the 2nd preamp stage, you need to use the VibroChamp style NFB circuit.  This is a "trap for the unwary".  If you simply inject NFB to the top of a bypassed cathode resistor, the bypass cap will bleed all the NFB signal down to ground.  This defeats the NFB circuit. The VibroChamp cures this by use of a split cathode resistance.  The small cathode resistor at the bottom serves as the NFB shunt resistor.  Now the NFB signal passes up from the top of that resistor, through the bypass cap and around the main cathode resistor, and into the tube.

[firemedic]

Yeah, that's why I got rid of the NFB in the 1st place- so I could have a bypass cap on that stage.

But my problem is this.

It's not a Princeton anymore at all. W/ Tubenit's help it's completely redesigned & the topology is as follows: normal 12A_7 stage->volume-> 5879 direct coupled to 12A_7 cathode follower-> FMV tonestack-> 6V6.

I could do local feedback, maybe a "conjunctive filter" on the 6V6. Where would I insert global NFB from the OT secondary? Would local NFB at the power tube work for this problem? Where in a CF can you insert global NFB? I have Merlin's book but his feedback section refers mostly to NFB with a LTPI.

I'm not on a computer that supports .sch files right now, the "5879 SE redo" thread in the tweaking section has the schem.

[jjasilli]

One amp I had the problem with was a VibroChamp that I hotrodded.  (Much like the 5F2).  See the VibroChanmp AB764.  One thing I did was to remove all NFB.  That turned out to be a mistake.  A. I removed the stock local NFB cap across the power tube pins 5 > 8; and B. removed the entire NFB loop from the OT to the 2nd preamp stage.  It sounded good except for fizzy speaker -- very frustrating!.  It then dawned on me that if Leo used both local and global NFB in this puppy, then maybe NFB is a necessity here.

I started replacing NFB and wound up with this: I re-installed the cap between pins 5 & 8 of the power tube.  Then re-installed the global NFB loop, and with higher voltage.  This was done as follows:  The OT was replaced with a Hammond 10-W multi-tap.  The NFB is now sourced from the 16Ω tap  -- that should yield about 2X the NFB voltage compared to the stock 3.3Ω OT.  Then the 27k NFB series resistor was reduced by trial and error to 2.7K.  Altogether that's a lot of NFB; but the fizz finally evaporated!  The point is that by boosting the gain in the amp, which is a good thing!, you may need a rather strong dose of NFB to suppress speaker fizz.

A conjunctive filter is another alternative.

[jjasilli]

Oh, and for your configuration, you have a number of options.  Local NFB at the 5879; and/or global NFB to the 5879.  Local NFB at the power tube; and or conjunctive NFB after the power tube.  (It's also possible to run global NFB to the 1st preamp stage).

[firemedic]

I'm gonna try to dial in your first suggestion, the vibrochamp global NFB inserted to the 5879 cathode. It should be fairly easy to implement, I'll try to find the minimum values that work. It seems like purely resistive feedback, as opposed to capacitive treble shunting, would be more effective w/ my issue. Thanks so much for the inspiration.

[firemedic]

My man.
That did it. In fact I think just splitting the cathode supplied some current NFB, as soon as I did that the fizz was pretty much gone- there's not much difference when the global FB is applied. Thanks for the reality check.

[jjasilli]

 

If there is now a resistor between the bottom of the bypass cap and ground, then that resistor is reducing the "boost" effect of the cap.  This alone mimics the effect of NFB, which does basically the same job.