NFB / PPIMV question

[Fred G.]

I put a Torres post PI MV in my Fender Bandmaster 11 years ago. I have wrestled with this amp for 11 years trying to tweak it into sounding good, but always had an annoying problem with excessive shrill high end, no matter what I tried. It always "almost" sounded great, but continually frustrated me when I tried to tweak it into "that sound".

Tonight, I decided to get drastic (for me, anyway). I first disconnected the NFB. Noticed "some" difference, but nothing earth-shattering (keep in mind I have a sleeping wife on the second floor, so I couldnt crank things up too much, even though I'm in the basement).

Next thing I did was to "standardize" the PPIMV, since the Torres version is "backwards" (the wipers are "fed" from the first set of caps, instead of the wipers "feeding" the 2nd set - did I explain that right?).

Anyway, now the amp is a whole different critter. The shrill highs vanished, as if by magic. The excessive rumbly farty bass has been tamed. It doesn't have the same amount of grind that it did before, but it just sounds "better".

I feel like I have something to tweak and work with now to finally get this amp where I want it!

I'm not real sure why reversing (or "standardizing") the post PI MV connections made such a big difference, tone-wise. I don't think disconnecting the NFB wire was the big factor as much as changing the PPIMV connections was.

Can somebody explain to me just what I really accomplished here?

Thanks,

Fred G.

[Fred G.]

MAybe I should rephrase my question

I suppose what I'm really asking is why changing the way the pot was wired had such a dramatic effect on the tone in a 4-cap PPIMV - I wouldn't have thought it would make much difference, but it obviously did.

Thanks,

Fred G.

[Ray Ivers]

Fred,

The post-PI MV you're using now - with the PI-output coupling caps connected to the tops of the pots, and the MV outputs to the PA grids taken from the wipers - maintains a consistent/light loading on the PI output and a consistent frequency response from the PI coupling caps during most of the pot's rotation. At higher MV settings, when the PA grids are conducting or trying to, things change drastically - but even then, any remaining resistance between the wiper and top of the pot will still provide some PI/PA isolation, up to the pot-maxed setting (when both designs should sound identical).

OTOH, the "backwards" wiring you mentioned will progressively load down the PI output, eventually causing PI distortion (I'm not passing judgement on this aspect at all, just noting it). However... it will also vary the RC time constant of the PI coupling caps, first reducing just the bass response and then mids too as the MV is turned down; IMO this is exactly what you don't want a MV to do, given the human ears' reduced sensitivity to bass at lower volumes.

Evidently some people prefer the second "backwards" type, but FWIW my assessment is the same as yours. Regarding NFB - I would highly recommend disconnecting it in any post-PI-MV setup, for reasons that are a whole 'nother post in themselves.

Ray

[Fred G.]

Thanks Ray - that's the kind of info I was looking for, because I really didn't understand WHY it made such a difference, but I knew that to my ears, it sure DOES.

As to disconnecting the NFB - I read somewhere in my surfing that, if you disconnect the NFB loop you should change the 82k/100k resistor pair to a 100k/100k combination, but the reason for this was not given. Is this really necessary, and if so, why?

Also, Instead of disconnecting the NFB circuit, could it also be rendered "ineffective" (I can't think of a better choice of words) or less of a factor, let's say, by using a 22k series resistor with the 1500 ohm ground resistor?

This whole NFB/PPIMV issue confuses the hell out of me!

Fred G.

[Ray Ivers]

Fred,

To disable NFB in the 6G7/6G7A/AB763 Bandmaster circuits, I would recommend grounding the PI end of the feedback resistor (either 10K or 100K) to the main PI ground point. Yes, I realize this will ground the PI tail resistor and the grid-AC-grounding cap - exactly what you want to do when removing NFB from these circuits IMO.

I read somewhere in my surfing that, if you disconnect the NFB loop you should change the 82k/100k resistor pair to a 100k/100k combination

I've read this, too. I've also read the exact opposite. My own personal recommendation would be to replace the 100K PI plate resistor with an 82K in series with a 50K internal trimmer, and adjust to taste (or for perfect PA balance with a 'scope, as I would - not for 'hi-fi' reasons, but for getting-a-sound-I-really-like reasons ).

Ray

[Fred G.]

Thanks again, Ray - I appreciate your patience in revisiting this stuff with me, because I realize this is one of the more heavily repeated topics here.

I disconnected the NFB right at the speaker jack.

I've not had much opportunity to fiddle around with the amp since I made the mods. I've been pondering maybe bleeding a little bit of NFB back in (as in an old bandmaster,100k/5k with a 25k presence pot - also very "Marshall"), or maybe a soul countrol, or just leaving it disconnected.

I'm just happy as hell to have finally gotten some degree of control of the tone of this amp - I'm astounded at the difference that rewiring the MV connections made. I have another amp (Deluxe clone) with a 4-cap PPIMV that hasn't given me a bit of trouble (it always had the "standard" wiring scheme), with the exception of getting a bit too bright at certain settings in the master control - i.e., from 7 to about 8.5 or so. Above that, the power amp goes into overdrive very nicely and diminishes the excessive highs. I'm pondering greatly reducing the NFB in this amp as well, or adding resistors ala Bruce Collins' PPIMV.

Fred G.

[Doug H]

Fred,

The post-PI MV you're using now - with the PI-output coupling caps connected to the tops of the pots, and the MV outputs to the PA grids taken from the wipers - maintains a consistent/light loading on the PI output and a consistent frequency response from the PI coupling caps during most of the pot's rotation. At higher MV settings, when the PA grids are conducting or trying to, things change drastically - but even then, any remaining resistance between the wiper and top of the pot will still provide some PI/PA isolation, up to the pot-maxed setting (when both designs should sound identical).

OTOH, the "backwards" wiring you mentioned will progressively load down the PI output, eventually causing PI distortion (I'm not passing judgement on this aspect at all, just noting it). However... it will also vary the RC time constant of the PI coupling caps, first reducing just the bass response and then mids too as the MV is turned down; IMO this is exactly what you don't want a MV to do, given the human ears' reduced sensitivity to bass at lower volumes.

Evidently some people prefer the second "backwards" type, but FWIW my assessment is the same as yours. Regarding NFB - I would highly recommend disconnecting it in any post-PI-MV setup, for reasons that are a whole 'nother post in themselves.

Ray

That sounds backwards to me. Standard 4 cap version most of the pot is in series at most volumes, whereas if you wire the pot backwards that series resistance is added to the bias feed and parallel with the pot's ground bleed to get the load on the PI, which will ramp up with volume, at least at first, which is a good thing since the nfb is ramping up as well. Maybe all the reverse wiring needs is a couple leading 82k resistors to avoid excessive high pass filtering?

[HTH]

Theres a similar thread on the Metro forum at the minute. I'd offer the same advice in your situation...

* remove the presence pot & the NFB altogether
* ground the bottom of the PI where the 'top' of the presence pot currently connects
* change the two PI to a 820ohm and 47k (leave the two 1M resistors)

You will find the amp tighter and less fizzy using this setup - I did ALOT of testing when I was tweaking a PPI-MV a couple of years back.

You might like a 56k over the 47k as the 'tail' resistor in the PI - just slightly less gain and more focus in the bass end.

[Guitarist]

HTH,

Those values, could they work as a simpler PI design to avoid the neg. feedback and complexity-born errors that some may create?

Sometimes I question the presence circuit's necessity. Anyone else?

[HTH]

HTH,

Those values, could they work as a simpler PI design to avoid the neg. feedback and complexity-born errors that some may create?

Sometimes I question the presence circuit's necessity. Anyone else?

It all depends on the tone you're going for. If you want a raw little rock machine, then having no NFB works in your favour, however you'll have sweet FA in terms of headroom and end up with a bigger output section to create headroom.

With NFB you can keep a smaller output section and retain a fair degree of headroom with the correct NFB setup. For instance - try removing the NFB from a Bassman and see how raunchy it gets very early.