Adjusting Negative Feedback

[Sir Cuitous]

I just finished a custom amp with this configuration:

Pre amp: 5E3
PI + power amp: 5D5
Negative Feedback loop and Presence control: 5D8

The amp has beefed up transformers and filter caps. It was designed for 6L6 tubes not 6V6s, but it was built inside of a Deluxe chassis.

Since this amp works and I am happy with the tone and very low noise level, I am posting here in the Theory and Design forum.

The issue is that it is not as loud as it should be. I suspect that there is too much negative feedback. I want to know if anyone can tell me how to modify the feedback loop to do two things. 1) decrease the feedback. 2) control more of the overall feedback instead of just particular frequencies.

In particular, what is the result of changing the feedback resistor? And, what happens if I change the cap or remove it from the presence control?

I believe I can also feedback from the 4ohm tap of the OT instead of the 8ohm tap to lower the feedback.

My gratitude as always to those who take the time to help out.

[defaced]

First make sure that the NFB is the problem. Disconnect the feedback resistor from the OT and the presence pot from ground.

In the NFB designs I'm familar with, the feedback resistor and the cap form an RC pair which determine the cut off point of circuit. The feedback resistor also determines the amount of feedback. Increasing it's value will decrease feedback and visa versa (wire in a pot here to really get a feel for the sensitivity).

If you find that the NFB loop is the issue and want to reduce the amount of feedback but keep the same frequency point, use this formula to determine the cut off point of the original circuit, then using your new resistor value and the old frequency value, find the new cap size:
f = 1/(2*pi*R*C)

[Sir Cuitous]

I disonnected the feedback loop and it did not produce as big of an increase as I expected. It was not the NFB! It was the interaction of volume controls. If the volume of the channel your not plugged into is turned up past 9, then it interacts with the channel your playing and starts to decrease the volume. The more I turn the opposite channel past 9 the more it attenuates the volume. It almost sounds like it is cleaning up the tone a little. Which is why I though it was the NFB. I suppose that's the normal opperation of this pre-amp.

This amp sounds good without the NFB, but is a little too bassy when turned up (now that I can hear it's full output).

[tubeswell]

I disonnected the feedback loop and it did not produce as big of an increase as I expected. It was not the NFB! It was the interaction of volume controls. If the volume of the channel your not plugged into is turned up past 9, then it interacts with the channel your playing and starts to decrease the volume. The more I turn the opposite channel past 9 the more it attenuates the volume. It almost sounds like it is cleaning up the tone a little. Which is why I though it was the NFB. I suppose that's the normal opperation of this pre-amp.

This amp sounds good without the NFB, but is a little too bassy when turned up (now that I can hear it's full output).

yep that's a typical tweed-era two vol plus tone control. Once the 'other' channel's vol gets past about "7", it starts to eat away the primary channel volume.

[Sir Cuitous]

Yeah, I figured the volumes were supposed to interact that way. I will change some coupling caps tomorrow to get the bass into the right zone and then I believe it is done (for now).

I'm not sure if I will re-install the presence control. I need to hear it with 0.022 caps first.

[tubeswell]

Yep FWIW I prefer my 5E3 clone with .022uF couplers after the 1st voltage amp stages and driver stage.

[Sir Cuitous]

I built it with all four caps as 047, but it's still too bassy. I'll switch them to 022 today. I may also bypass the PI cathode now that I have disconnected the feedback.

[tubeswell]

I may also bypass the PI cathode now that I have disconnected the feedback.

Hi Sir Cuitous

The driver stage lends itself to being bypassed (being a conventional gain stage), but the PI stage doesn't. The PI itself is not supposed to have any gain. If you add a bypass cap to the 1k5 on the PI, you will put something out of balance. There needs to be a voltage swing across the PI cathode resistor for the PI to work. If you put a cap there it will stop (or at the very least, hinder) the voltage swing (, which in turn will diminish one whole side of your output stage).

[Sir Cuitous]

Tubeswell, this PI uses two separate cathode resistors. I did bypass one as in the Fender schematic, but the difference was subtle. Maybe a lower value cap would make a more noticable change.

I changed the coupling caps to .022 - much better.

[tubeswell]

Ah I just saw your earlier post where you said '5d5' PI, which of course is a parahase inverter - completely different to a 5e3 PI (I'm sorry I do apologise)