Which cap should I reduce next?

[brewdude]

I’ve been working on an amp…
It’s a single channel with a 12AX7, simple volume and tweed style tone knob, a brown era fender LTP phase inverter and a pair of 6K6GT’s.
I like it a lot, but I’ve been trying to shave some low end.

I’ve made a small adjustment to the preamp’s cathode bypass caps already and like where it’s at.

Should I next try changing the coupling cap before the PI, or the two coupling caps after the PI?

Is there any reason to expect one choice is better than the other? If so, why?

[bepone]

put 1nF before PI and 22nF after and try to see if will help

[roberto]

Do you want to cut lows, or to make it seem brighter?
What B+ and OT do you use with 6K6GT? Around 300V and 10k?

[Stevem]

If it’s a cathode bias output stage and it has a bypass cap you can start out by halving that Uf value.

[roberto]

The gnfb will dampen the effect of anything inside the loop, better to intervene outside it.

[TUBEDUDE]

To start, maybe get rid of global feedback and use it only if needed, and just enough for stability. This might shift things in a more desirable direction.

[brewdude]

Thanks for the replies.

Fixed bias, about 300v on the plates, 12k Primary, -22.8v at the grids.

I don’t want it any brighter.

I used the 6G3 as inspiration for this build.
Coupling cap before PI is .01uF,
Caps after PI are .1uF,
GNFB is 56k on 8 ohm secondary.

Is there any benefit or detriment to carving the bass from before or after the PI?
Or, should I reduce them all together?

[Stevem]

If you have other amps that you use your guitar(s) thru and you play them with the bass up less the max then you simply might want to lower the neck pickups on your guitars a little bit to work better with this build and on those other amps just open the bass up more.

[roberto]

Thanks brewdude, very clear.
Before the PI, so out of the feedback loop, any mod is effective.
After the PI, so inside the feedback loop, any mod is dampened by the loop itself.

Don't act on the height of the pickups a priori.
More bass at input is a totally different beast than less at input and then push more later.
Changing the bass pot position will alter mid frequency and overall balance in most eqs.

[sluckey]

Changing the PI output caps from .1µF to .047µF will have a very noticeable bass reduction effect. This is a very popular Fender mod.

[brewdude]

Thank you all for the input.

I notice that fender used a number of different values before the PI in the Brown era amps. The deluxe uses a .01uF and the bassman a 500pF (IIRC). Seems counterintuitive to me.

I think I will likely end up halving the values of the caps on both sides of the PI.

[pdf64]

… I notice that fender used a number of different values before the PI in the Brown era amps. The deluxe uses a .01uF and the bassman a 500pF (IIRC). Seems counterintuitive to me…

For the 60s Fenders, it tends to be those intended to be used with a closed back cab that had a 500pF coupling cap to the LTP, whereas the combos tended to use 1nF.
Of course, exceptions can be found to the above …

[brewdude]

Got it.

That may be the root of my dissatisfaction with the current sound. I’m using a “3D style” 1x12 that is pretty large with side vents, with an Eminence GA-SC64. I think the cab’s response is a little bass heavy, like a closed back might be(?).

[Roe]

....
Coupling cap before PI is .01uF,
Caps after PI are .1uF,
...

I'd try .001uf and .022uf respectively for tighter bass and faster response with less blocking distortion

[teemuk]

Do note that PI coupling caps are reactive components and if inside feedback loop such will effect phase of the feedback signal. If you dimension them too low (corner frequency high enough within audio band) you start to introduce positive feedback at low frequencies and risk amplifier stability. There is probably a decent balance of choosing low enough capacitance to minimize bias shifting and blocking distortion at sustained overdrive, and high enough capacitance to retain decent stability marging.

I would leave deliberate hi-pass voicings to caps that are outside the loop. They are more effective anyway because they don't have to fight against negative feedback.