Negative Feedback Adjustment?

[hesamadman]

Sorry to make another post regarding my last build, but this is a whole different topic of the amp that I would like to start a discussion not only for debugging purposes, but for furthering knowledge on the topic.


So, I posted about my 20 watt amp having issues in the FX loop. Seemed to have been affected by high frequencies mostly. The amp played 100% fine until tremolo was in loop. Also when nothing was plugged in there was a high frequency noise dependent on how the tone stack was adjusted. If you cranked highs and gain knob all the way it squealed. Was odd because input is grounded out obviously.


Well it turned out not to be the FX loop at all but the NFB Loop. I cut it out and noise with nothing plugged in the amp went away and tremolo sounded fantastic in the loop.


Im going to back up a little bit and express what I learned so far about NFB on my last few builds. A couple years ago when I was building a handful of AC30's with an added gain stage, I noticed a very resonant frequency in some notes. Low notes. The open A string was VERY resonant. I posted a thread about that. Couldnt figure it out. Hated the tone of it but a lot of people who played the amp liked it. It occurred to me during this 20 watt build that "oh the AC 30 didnt have a NFB". Thats where that crazy resonance comes from. My 20 watt amp with the NFB installed is exactly how I want it. Well now that I cut it out guess what? That resonant frequency is there and I have to get rid of it.


I realize this is how resonance controls work. I've installed them on my Marshall style builds. I may even implement one here. on this amp. I copied and pasted it exactly (other than the cathode resistor value) from the Hoffman Princeton Reverb. You will notice that I even have my cathode of the gain stage setup just like the hoffman layout. I have a 2.7k bypassed by a 22u and a 47 ohm in series to ground. My NFB connection is between those two points and the goes to a 22k resistor on the 8 ohm tap of the OT. (i only have an 8 ohm tap on the OT)


This is a different type of phase inverter than the marshalls, but would it be better to locate the NFB at the cathode of the phase inverter or can I just make some adjustments to the values here? Or what if I just implement the resonance control mod? Maybe even adjust it to a setting I like and doesnt cause noise and solder that value in?


What would you do?


http://el34world.com/Hoffman/files/Hoffman_PrincetonReverb.pdf

[shooter]

What would you do?

your sketch shows NFB incorporating the TS within the FB loop, I've been counseled against this, to much weirdness happens.  If you look at Doug's PR that you referenced the TS is outside the NFB

[hesamadman]

That's what I've found out. But where to put it instead?

[sluckey]

Try swapping the OT primary plate leads and reconnect the NFB. Any better?

Where did you come up with this schematic?

[hesamadman]

I'll try the OT lead switching.


I just put together a simple AC 30 style normal channel with a full tone stack and MV. I wanted a clean channel and a tone stack without the loss.

[sluckey]

You are likely to still have issues with your circuit. Connect the NFB to the PI exactly like most Marshall amps.

[hesamadman]

That's what I wanted to try but I wasn't sure how cathodyne works compared to long tail. I'll give it a whirl today and report back.

[sluckey]

Forget what I said about Marshall amps. I thought you had an AC-30 power amp, but I guess maybe you have a Princeton power amp. Can you show us your entire schematic?

[mwelch55]

your sketch shows NFB incorporating the TS within the FB loop, I've been counseled against this, to much weirdness happens.

I have seen several original Fender schematics that have NFB right after the volume control and tone stack (like the original post).

Also, I did that on my last Fenderish amp and it worked great.  What sort of weirdness can I expect to see?

Mike

[sluckey]

It's not a problem putting the NFB loop ***AFTER*** the volume and tonestack. But that's not how madman did it. His NFB loop is before the tonestack.

[92Volts]

The most obvious "weirdness" from putting it before the tonestack is the NFB would "try" to cancel any changes the tonestack makes to the frequency response. I'm sure it has some (more complex) effects on distortion created later in the preamp, and/or may be prone to oscillation, but in general cancelling the effects of your tonestack is just not the effect most people are looking for.

[mwelch55]

Sorry, I didn't see the TS after the NFB.  It makes sense that you shouldn't do that.