Negative feedback

[dscottguitars]

Hello,

I have a question about the NFB circuit. I usually like using very little. I like the raw sound. But I just discovered after testing this new amp I made that I didn't hook up the lead from the speaker to the voltage divider as shown in my schematic. When I connected it, I got a little buzzing sound, so I disconnected it and grounded the cathode directly.

Why would the NFB add some noise? I was under the impression it was supposed to reduce and noise.

The amp sounds KILLER! It's the first time I used this design and with a single tone control. I modified the Muff Pi using the Duncan tone stack calculator to get the values. It has a great bass boost, honking mids and piercing treble. I get awesome distortion with gain all the way up and volume very low. When I crank it the EL84s go into overdrive for a searing endless sustain. I will be making another.

Thanks,

Daniel

[kagliostro]

I'm not able to be of great help, however

did you connected to ground one of the OT secondary ?

to me when there is an NFB you must have the secondary of the OT floating (no ground connections)

Also, there is something odd on your NFB + cathode path of the PI arrangement, may be you must not connect to ground the 3W resistor (?)

Franco

[pdf64]

...did you connected to ground one of the OT secondary ?

to me when there is an NFB you must have the secondary of the OT floating (no ground connections)...

Rather, to avoid 'ground loops', the OT secondary should use the same 0V / circuit common point as the LTP.

...there is something odd on your NFB + cathode path of the PI arrangement, may be you must not connect to ground the 3W resistor (?)...

Yes, it would need that 0V connection lifting for the NFB to operate.

[dscottguitars]

Let me clarify: Those two grounds are not connected. The ground coming off the 20K 3W resistor IS grounded. The one from the NFB circuit is not. I only have it there in dashed line to show how I did have it connected before.

I don't have the negative side of the secondary (output jack) grounded to the chassis.

Thanks,

[jjasilli]

As drawn, both sides of the voltage divider go directly to ground.  The NFB doesn't enter the signal path.

[dscottguitars]

As drawn, both sides of the voltage divider go directly to ground.  The NFB doesn't enter the signal path.

Look at my last post. Both of them are not hooked up.

[pdf64]

...I do have the negative side of the secondary (output jack) grounded to the chassis. Is that not good?...

It's not ideal, may be causing the 'buzz'.
A jack that doesn't connect sleeve to chassis needs to be used, and the sleeve then be connected to 0V / circuit common / chassis at the same point as the NFB / LTP tail resistor.

[jjasilli]

The NFB goes directly to ground whether either one, or both of those resistors is conected.

[dscottguitars]

I was mistaken, the ground of the speaker is not connected to the chassis.

[dscottguitars]

...I do have the negative side of the secondary (output jack) grounded to the chassis. Is that not good?...

It's not ideal, may be causing the 'buzz'.
A jack that doesn't connect sleeve to chassis needs to be used, and the sleeve then be connected to 0V / circuit common / chassis at the same point as the NFB / LTP tail resistor.

I have an isolating jack too, it's plastic and has no connection to the chassis.

[dscottguitars]

The NFB goes directly to ground whether either one, or both of those resistors is conected.

See new attachment

[jjasilli]

Nothing attached.

[pdf64]

I was mistaken, the ground of the speaker is not connected to the chassis.

The OT secondary 0V end needs to be connected to circuit 0V, preferably at the same point as the LTP.

[sluckey]

Well, I'm confused. You say the OT secondary is grounded to chassis. Then you say it's not. Which is it?

Doesn't matter how the OT secondary common lead is connected to chassis. Fender used a standard switchcraft jack that naturally grounded to the chassis and the OT secondary lead is connected to the sleeve terminal. If you use an isolated jack such as a cliff type then you must connect a wire from the sleeve to chassis, preferably at the point that the LTP PI gets it's ground. But is is vital that you connect the OT secondary common wire to chassis ground with some means. Otherwise, you will just get a buzz.

[jjasilli]

Good point.  The schematic implies that both the (-) OT secondary and the (-) speaker terminal are connected to the chassis.  Instead the speaker leads should connect directly to the OT.  Otherwise noise in the chassis currents could possibly reach the speaker. 


In theory the OT-speaker circuit could float.  But it's better to ground it.


As drawn the NFB also grounds the other + side of the speaker circuit.  This is not pretty.  NFB needs to be injected into the top of the recipient cathode resistance OR into the middle of a split resistance if a cathode bypass cap is present. 

[sluckey]

As drawn the NFB also grounds the other + side of the speaker circuit.  This is not pretty.

I don't know what you are seeing but on my end of the internet the positive side of the speaker is definitely not grounded. Maybe I'm just not understanding what you are saying?

[jjasilli]

In the first drawing the NFB is improperly going to ground through one or 2 resistors.  The original question is why hooking up this NFB circuit causes noise. Apparently there is no noise without this circuit. 

[dscottguitars]

Well, I'm confused. You say the OT secondary is grounded to chassis. Then you say it's not. Which is it?

Doesn't matter how the OT secondary common lead is connected to chassis. Fender used a standard switchcraft jack that naturally grounded to the chassis and the OT secondary lead is connected to the sleeve terminal. If you use an isolated jack such as a cliff type then you must connect a wire from the sleeve to chassis, preferably at the point that the LTP PI gets it's ground. But is is vital that you connect the OT secondary common wire to chassis ground with some means. Otherwise, you will just get a buzz.

I said I made a mistake. Do people not read? It is not connected.

[jjasilli]

A picture is worth a thousand words. 

[dscottguitars]

Good point.  The schematic implies that both the (-) OT secondary and the (-) speaker terminal are connected to the chassis.  Instead the speaker leads should connect directly to the OT.  Otherwise noise in the chassis currents could possibly reach the speaker. 


In theory the OT-speaker circuit could float.  But it's better to ground it.


As drawn the NFB also grounds the other + side of the speaker circuit.  This is not pretty.  NFB needs to be injected into the top of the recipient cathode resistance OR into the middle of a split resistance if a cathode bypass cap is present.

I don't see what you are referring to. The before and after drawing is correct.

The positive side of the speaker goes to the NFB circuit through a voltage divider which is grounded.

[dscottguitars]

I'm thinking because the speaker is NOT grounded that that is why I got some buzz when I connected the NFB. Could that be correct? As is stands now, without NFB it does not buzz. I will still hook it up and see what happens.

Thanks

[dscottguitars]

Nothing attached.

It was attached two replies above.

[sluckey]

I'm thinking because the speaker is NOT grounded that that is why I got some buzz when I connected the NFB. Could that be correct? As is stands now, without NFB it does not buzz. I will still hook it up and see what happens.

Thanks

yes