New build: Princeton Reverb with mods - advice wanted!

[klooon]

New draft version:



- I added the Excelsior tremolo circuit. I put the 320VDC "C" tap to it. Hopefully it works like this and for both the fixed and cathode bias mode.
http://ampgarage.com/forum/files/paw...ematic_149.pdf

- I put a grid stopper for the PI according to the advice here: The Valve Wizard

- The standy switch is added and according to this very interesting topic I added a resistor over the switch to limit inrush currents, etc.
The Valve Wizard

- Grid stoppers (470k) on the grids of the power tubes.

Hopefully, this is fine now....??

[tubeswell]

What about the resistors from the 0.1uF caps to the trem circuit? What is their function exactly? I should use the same value (i.e. 470k)?

If you want a depth control, that 470k can be replaced with a 500k pot

[klooon]

If you want a depth control, that 470k can be replaced with a 500k pot

Would that be a dual ganged pot?

[klooon]

I have looked again (and again) at the Excelsior schematic and I am trying to understand why this might work (assuming output from trem. circuit is similar?). They have physically separated the position of the preamp and the power section in the cabinet. One part is at the top and one part is at the bottom of the amp. You see two times a pair of 470k's. In between one pair is the trem signal; the other pair has a ground connection in between. Is this why it might work??

Excelsior trem circuit:

[klooon]

Here is another interesting example of problem reported with a cathodyne PI + tremolo.
The Amaretto | DvNator's Guitar Amp Projects

He ended up with a using a Vibrochamp tremolo (with single triode in combination with a MOSFET as a cathode (source) follower) and that seems to work fine! This could also an option for me?!

[tubeswell]

Here is another interesting example of problem reported with a cathodyne PI + tremolo.
The Amaretto | DvNator's Guitar Amp Projects

He ended up with a using a Vibrochamp tremolo (with single triode in combination with a MOSFET as a cathode (source) follower) and that seems to work fine! This could also an option for me?!

[ATTACH=CONFIG]28381[/ATTACH]

Indeed

[tubeswell]

I have looked again (and again) at the Excelsior schematic and I am trying to understand why this might work (assuming output from trem. circuit is similar?). They have physically separated the position of the preamp and the power section in the cabinet. One part is at the top and one part is at the bottom of the amp. You see two times a pair of 470k's. In between one pair is the trem signal; the other pair has a ground connection in between. Is this why it might work??

Excelsior trem circuit:
[ATTACH=CONFIG]28380[/ATTACH]

Sorry I was looking (not too carefully) at the 1M bleed-out resistor from the coupling cap after the mosfet source terminal thinking you were referring to it as '470k'. The 1M is the one you would replace with a pot to make a depth control. The 2 x 470K you are referring to are there to lessen the AC load on the cathodyne output impedance. You leave those where they are.

[klooon]

I have looked again (and again) at the Excelsior schematic and I am trying to understand why this might work (assuming output from trem. circuit is similar?). They have physically separated the position of the preamp and the power section in the cabinet. One part is at the top and one part is at the bottom of the amp. You see two times a pair of 470k's. In between one pair is the trem signal; the other pair has a ground connection in between. Is this why it might work??

Excelsior trem circuit:
[ATTACH=CONFIG]28380[/ATTACH]

Does anyone know why the Excelsior bias tremolo works and the others don't in a cathode bias circuit? Is it the point of (separate) insertion (at the point of the 470k's instead of a bit further on: the (bias) point that is connected to ground in the Exc. circuit, i.e. the second pair of 470k's)??? PI output impedance vs. Power amp input impedance... ?? Also looking at this comment:

2 x 470K you are referring to are there to lessen the AC load on the cathodyne output impedance

Or is it something in the tremolo circuit itself? The buffer or maybe specific values that output a different (dynamical; AC vs DC; impedance; ...) signal to the power tube cathode bias voltages/currents???

[tubeswell]

Does anyone know why the Excelsior bias tremolo works and the others don't in a cathode bias circuit?

The excelsior is bias-vary trem. It works by varying the grid voltage of the 6V6s. It is a 'fixed depth' trem that has enough strength to get the trem to work (when it is engaged). If you replace R15 with a 1M pot (ground lug to ground and wiper to the 2 x 470k), you will have a depth control but you'll likely find that the trem won't kick in until the pot is about 1/2-2/3 of maximum pot rotation.

Or is it something in the tremolo circuit itself? The buffer or maybe specific values that output a different (dynamical; AC vs DC; impedance; ...) signal to the power tube cathode bias voltages/currents???

The IRF820 buffers the LFO stage against the AC load presented by the grid load resistors for the 6V6s in parallel with the cathodyne load resistors. The 2 x 470k that the LFO connects to the signal chain through help to lessen the AC load on the signal chain; not on the LFO. (They in fact attenuate the signal on the LFO, because - in conjunction with the grid load resistors for the 6V6s, they form a voltage divider for the LFO. If you decrease R15 or decrease the 2 x 470k, you will increase the LFO strength getting through to the grids, but you will also attenuate the signal from the cathodyne stage). That's why it might take a bit of mucking around with to optimise. Which is why I suggested that you might have to try it and see.

[klooon]

The excelsior is bias-vary trem. It works by varying the grid voltage of the 6V6s. It is a 'fixed depth' trem that has enough strength to get the trem to work (when it is engaged). If you replace R15 with a 1M pot (ground lug to ground and wiper to the 2 x 470k), you will have a depth control but you'll likely find that the trem won't kick in until the pot is about 1/2-2/3 of maximum pot rotation.




The IRF820 buffers the LFO stage against the AC load presented by the grid load resistors for the 6V6s in parallel with the cathodyne load resistors. The 2 x 470k that the LFO connects to the signal chain through help to lessen the AC load on the signal chain; not on the LFO. (They in fact attenuate the signal on the LFO, because - in conjunction with the grid load resistors for the 6V6s, they form a voltage divider for the LFO. If you decrease R15 or decrease the 2 x 470k, you will increase the LFO strength getting through to the grids, but you will also attenuate the signal from the cathodyne stage). That's why it might take a bit of mucking around with to optimise. Which is why I suggested that you might have to try it and see.

Thanks a lot for the excellent explanation! Now I can go forward with the confidence that it will work! (trial-and-success instead of trail-and-error...) Then it's also a good idea to tweak a bit to make the pot rotation work intuitive.

[klooon]

The excelsior is bias-vary trem. It works by varying the grid voltage of the 6V6s. It is a 'fixed depth' trem that has enough strength to get the trem to work (when it is engaged). If you replace R15 with a 1M pot (ground lug to ground and wiper to the 2 x 470k), you will have a depth control but you'll likely find that the trem won't kick in until the pot is about 1/2-2/3 of maximum pot rotation.

Would it be an idea then to put a 500k resistor in series between the hot lug of the pot and C6 (0.1uF)? Then the range is extended as well, allowing lower intensities as well (1M to 1M5)?

[tubeswell]

Would it be an idea then to put a 500k resistor in series between the hot lug of the pot and C6 (0.1uF)? Then the range is extended as well, allowing lower intensities as well (1M to 1M5)?

I suggest that you build the circuit before you modify it. If you are seeking to add a resistor to overcome any latency in the pot rotation for such a depth pot, then that would want to be on the 'ground side' of the pot.

[klooon]

I suggest that you build the circuit before you modify it. If you are seeking to add a resistor to overcome any latency in the pot rotation for such a depth pot, then that would want to be on the 'ground side' of the pot.

That's fine. Your remarks help a lot though. Not only during the build and tweaking it will, but I also have spend a bit of time today to read about the basics of voltage deviders etc.

[klooon]

What would be a good way to mount the mosfet on an eyelet board?

OK. About to order the components! Working in parallel a bit on the layout.

[tubeswell]

What would be a good way to mount the mosfet on an eyelet board?.

On a bit of perf board