5E3 oscillating from feedback loop

[six_eight]

I've got a deluxe build that is oscillating and I can't figure out what's causing it.  It sounds like it's got a tremolo going.  It only happens when I switch on the negative feedback mod.  If I remove the mod, the amp seems to work fine.  Voltages all check out fine.  I've built a couple 5E3s before and did this mod and they worked fine.  I suspect it's not the mod that is causing the problem, but something somewhere else in the circuit.  Does anybody have any ideas where to look?

[OldHouseScott]

Try swapping the leads from the OT to the plates (brown and blue wires perhaps?) It won't affect the amp with the NFB switched off and should cure the oscillation if it's due to the NFB.

[six_eight]

Thanks, OHS.  I tried that.  That just made noise even when the NFB was disconnected, which is one of the reasons why I think it's something other than the NFB itself - something minor that would allow the amp to still "work" even when the NFB is disconnected - the NFB just exacerbates the real problem enough so that you can notice it.

It starts out fine.  Then after about 15 seconds, with the NFB on, it's like there's a tremolo on.  But if I play quietly, it won't do it.  It's only when I strum hard that the tremolo kicks in.  Once the tremolo starts, it won't stop till I put it in standby.  It's like a brownface tremolo.  Voltages throughout the entire circuit start oscillating.

[tubenit]

If I remove the mod, the amp seems to work fine.

I suspect it's not the mod causing the problem, but somewhere else in the circuit?

 

With respect, Tubenit

[sluckey]

If I remove the mod, the amp seems to work fine.

Hmmm. Show us the mod.

[six_eight]

If I remove the mod, the amp seems to work fine.

I suspect it's not the mod causing the problem, but somewhere else in the circuit?

 

With respect, Tubenit

I realize that's quite a contradiction.

[six_eight]

If I remove the mod, the amp seems to work fine.

Hmmm. Show us the mod.

I've tried several different variations that all connect to the cathode on the first half of the V2 12AX7.   

1.  A la 5F1 where there's a 100k pot with SPDT switch.  When the knob is "on" it lifts the 25uf cap and decreases resistance between the cathode and the OT output.  There's a 22k limiting resistor in line.

2.  A la 5E6 using a 5k pot with a DPDT switch that lifts the 25uf cap.

3.  A la the 5B6 - just a switch that makes or breaks a 6k8 resistor to the OT.  The ground connection to the 1k5/25uf is removed and replaced with a 47ohm resistor.  The other end of the 6k8 connects here.

They all work.  They produce the sound you'd expect to hear.  It's just that I'm also getting a tremolo effect if I crank the volume on everything.

[six_eight]

If I remove the mod, the amp seems to work fine.

I suspect it's not the mod causing the problem, but somewhere else in the circuit?

 

With respect, Tubenit

Maybe this will help:

The amp seems to have more sag than a typical 5E3.  It's like this even if I remove the mod.  So much so that the background hiss cuts out completely if I hit a big cord and then mute it abruptly.

Also, I switched rooms and it seems to have cured the tremolo problem, which I don't understand since both rooms are on the same breaker.  There are less lights/computers in the  new room.  But I'm still get a massive amount of sag.

[HotBluePlates]

I've tried several different variations that all connect to the cathode on the first half of the V2 12AX7.

You probably have too much feedback.

The blackface Princeton puts a 47Ω resistor on the cathode resistor of V2a, then runs a 2.7k series feedback resistor.

The way you've described the connection, your resistor to ground is 1.5k (V2a cathode resistor). That's ~32x bigger than 47Ω, so your series feedback resistor oughta 32x bigger than 2.7k, or a bit under 100k.

Instead of 100k being the upper end, use a 1MΩ pot, and a series 100k resistor.

Forget the "tremolo" stuff until you get the feedback thing sorted out, if you're gonna keep it.

[six_eight]

I've removed the NFB mod completely so I can focus on the problem with the amp having too much sag, because I think that's what the real problem is.  It only goes into oscillation when I have the amp dimed and I hit a big chord.  You can hear the power not fully recovering and that's when it starts with the tremolo.  If I put it in standby and then back on, the trem is gone till I hit another big chord.

What would cause too much sag?  The 5Y3 is good.  The filter caps are good.  The voltages check out OK.  Could it be that the PT isn't producing enough current?  Where should I test the current and how much current should I expect to see?

[HotBluePlates]

What PT are you using?

How do you know for sure all filter caps have good grounds?

[six_eight]

http://www.classictone.net/40-18021.html  This is the PT that I'm using.

I tested the caps prior to building and then tested for continuity to ground from the negative leads of the caps.  Everything seems OK.

[tubeswell]

Can you post some gutshots of your build?