Resistor values in a LTP PI - affect on tone?

[wizard333]

Playing with adding presence and resonance controls to amps that dont otherwise have them and curious about how changing the resistors in the PI affect tone.

I'm familiar in a basic way with how an LTP PI works, but not so familiar with how changing the resistor values affects the sound.

As a for instance:

Blond Bassman: Has presence control, cathode resistors 820/6.8k, 4.7k to ground; presence is standard 25k pot and .1 cap bypassing the 4.7k so the high end isnt included in the nfb.

Blackface Bandmaster: Cathode resistors 470/22k, 47ohms to ground.

The nfb resistors are wildly different at 100k (blond) and 820 (black) but the tonal affect there is something I'm very familiar with.

So if I put the presence control on a Bandmaster for instance,

1) Do I need to switch the cathode resistors to the blond values? What tonal affect does changing from 22k to 6.8 have?

2) Should I use the common value (for amps with presence control) of 4.7k to ground? How does changing from 47ohms to 4.7k affect tone?

[MWJB]

Let's take each part step by step...

Cathode resistor - Sets the current drawn by the PI tube. 470R was fine in the tweed amps with a 12AX7 PI, probably because the 12AY7 input tube dropped gain at that end of the circuit, providing a degree of headroom (compared to 12AX all round). Roll on the Blonde/Brown tolex amps which had 12AX7 input & 12AX7 PI tubes and Fender pushed up cathode value probably to increase headroom. But, it wasn't enough, guys were still looking for more jangly cleans so Fender subbed the 680/820 cathode resistors bacck to 470 and used a 12AT7 PI tube.

Tail resistor - large values give better balance at the PI, but also drop gain & make for a looser sound. During the Tweed era tails dropped from 10K to 6.8K, increasing gain/punch/fidelity, balance doesn't seem to be an issue. Again by the BF era & using a 12AT7 27K then 22k were used. 27K was used in conjunction with 2x100K plate resistors, 22K utilised the more common 82K/100K set up, presumably to restore balance with the slightly smaller tail. 27K might be overkill if using a 12AX7 in the PI the amp may become too dull (assuming global NFB in all cases). No reason not to try 10K or even 22K with the blonde circuit, limiting factor will be how loose the amp feels with the higher values.

NFB loops - Blonde has a 4.7K load resistor with 100K dropping resistor, a ratio of 22:1. E.g. for every volt at the OT secondary 1/22 of a volt is fed back, out of phase, to the PI. BF has a load of 47ohms and a dropper of 820, therefore a ratio of 18.4:1...not a world away from the blonde. Typical BF/SF Fender ratios run from 9:1 for 2ohm & 4ohm outputs, with 18:1 being tyical for 8ohm BF amps & 58:1 for BF/SF Champs/Princetons. Higher ratios tighten up the amp, lower ratios make the amp feel looser/softer. Centre territiory for Fender might be considered around th 15:1 mark?

It's therefore the ratio that is critical, rather than specific resistor values. You simply wouldn't swap 47R for 4.7K on it's own ...or 100K for 820R because you screw up the feedback ratios. Too low and you get poor output, too high and the amp will sound brittle & may become unstable.

The 25K presence pot on the blonde is just an available size that allows the 0.1uf cap to be turned off as the series resistance of the pot is increased (turned down).

If I was putting a presence control on the BF Bandmaster, I just lift the Blonde presence control & NFB loop in their entirity, then I'd ear test PI tail & cathode values...if using a 12AX7 PI tube you might go stright in with blonde values, if using 12AT7 you might go in with BF values. So long as you like the tone a 12AT7 and a 12AX7 will work in either circuit.

1) no, but you might want to try them if running a 12AX7?

2) Yes, I would. Using 47ohms would require a bigger cap, say 1uf & a pot 5-10 times the size of the 47ohm load resistor...bonde parts are cheaper/more readily available.

Consider the NFB load & dropper as one circuit, consider the cathode & tail as another.

[wizard333]

Excellent info, thanks!!

Regarding a resonance control, if I'm understanding that circuit correctly, the larger the pot the greater the effect, and the smaller the bypassing cap the higher the frequencies affected?

Seems like the pot reduces nfb, but the cap limits the reduction in nfb to low frequencies thereby "boosting" them (or just not attenuating them as much via nfb)?

Any idea on suggestions for a pot size to start with there? I referenced some Peavey schematics but they dont list the pot value, only the part number

[novosibir]

The resonance pot's value is 3...10 times of the NFB resistor's value (dropper's value). Use a log pot - and to avoid scratchy noise, you can put a 1µF foil cap in series to the pot/NFB resistor, but it's actually not required.

Larry

[wizard333]

Thanks!

[mooreamps]

Instead of using a tail resistor, has anyone built a PI using a choke driven tail ?


-g

[wizard333]

If you want to see a strange PI, see the schematics for a Mesa DC-3.